Extracting the Americyl Hydration from an Americium Cationic Mixture in Solution: A Combined X-ray Absorption Spectroscopy and Molecular Dynamics Study

Extracting the Americyl Hydration from an Americium Cationic Mixture in Solution: a Combined X-ray Absorption Spectroscopy and Molecular Dynamics Study Sergio Pérez-Conesa, José M. Martínez, Rafael R. Pappalardo, and Enrique Sánchez Marcos∗ Department of Physical Chemistry, University of Seville, 41012 Seville, Seville, Spain E-mail: [email protected] Phone: +34 955421005 Abstract The Am(VI) solution chemistry differs from that of lighter actinoids, as U, Pu and Np, where the actinyl [AnO2]2+ is the most stable form and plays an important role in nuclear fuel technology. The behavior of americium in solution shows the trend to stabilize lower oxidation states, mainly Am(III). Riddle and col. (J. Radioanal. Nuclear Chem. 2016,309, 1087.) have recently reported the EXAFS and first XANES spectra of an americium-containing aqueous solution where the Americyl species is detected in a mixture. We have developed Am3+-H2O and [AmO2]2+ -H2O intermolecular potentials based on quantum-mechanical calculations to carry out classical MD simulations of these two cations in water. Structural information extracted from the statistical trajectories has been used to simulate EXAFS and XANES spectra of both solutions. For the Am3+ case the theoretical-experimental agreement for both EXAFS and XANES 1 spectra is satisfactory. This is not the case for the [AmO2]2+ aqueous solutions. However, when considering an aqueous solution mixture of both cationic forms in a 55/45 [AmO2]2+ /Am3+ ratio, the theoretical-experimental agreement is recovered. The EXAFS and XANES spectra which would correspond to a pure [AmO2]2+ aqueous solution are proposed. In the XANES case, the main features characterizing the simulated spectrum are consistent with those previously found in the experimental XANES spectra of stable [UO2]2+ and [PuO2]2+ in water. Introduction Americium is formed as a byproduct of the irradiation of uranium by neutrons as well as of Pu decay.1It is of great interest since it is one of the main responsible of the high radiotoxicity of minor actinoids in nuclear fuel waste.2–4 To engineer such a waste recycling process, to understand the biological hazard of americium or to study its geochemical speciation, detailed knowledge of its complex solution chemistry is necessary. Unlike lanthanoids, the aqueous chemistry of actinoids is very rich since each element does have a distinct behaviour with different stable oxidation states and species.5The An(VI) solution chemistry of lighter actinoids, as U, Pu and Np, is dominated by the actinyl species, [AnO2]2+, which is stable and plays an important role in nuclear fuel technology. However, the behavior of Am in solution tends to stabilize lower oxidation states, mainly the Am(III), adopting the monatomic form, Am3+. The stabilization of americyl species requires the presence of highly oxidizing media.6 The ability to examine these higher oxidation states solution chemistry would yield relevant information on how these forms could be incorporated in new separation processes of minor actinoids in nuclear fuel recycling.1–4 Structural information on aqueous americium ionic solutions has been obtained by means of Extended X-Ray Absorption Fine Structure Spectroscopy (EXAFS).6–9 This technique provides accurate short-range structural information around metal ions10,11 at concentrations as low as 10−4M, typical values found in actinoid solutions.12 Unfortunately, its ra- 2 diotoxicity makes the experimental information about americium highly-resource demanding and scarce. Extracting information and interpreting the spectra becomes an involved task, since experimental structural information obtained from other techniques is limited, so that structural models to help in the EXAFS fitting are lacking. This leads in many cases to assumptions about the model system based on analogies with other lighter actinoids12 or to find out synergies with theoretical tools.13–17 The most stable species of Am in solution is its aqua ion, [Am(H2O)n]3+, whose EXAFS spectrum has been measured by Allen et al.9and Stumpf et al.7,8 These studies established Am3+ hydration numbers in the range of 7.4-10 with an Am-O distance range of 2.477- 2.494 Å. Riddle et al.6have very recently obtained the first EXAFS spectrum of Am(VI), [AmO2]2+ , in a 1M HNO3aqueous solution containing NaBiO3. These authors reported that the sample solution must be an Am(VI)/Am(III) mixture having a 70/30 ratio on the basis of UV-vis spectroscopy, although radiolysis degradation and sample stability may affect these values during the sample preparation, transportation and measuring.6Because of the low americium concentration (below 0.015M) in the sample we can assume that the recorded spectrum can be envisaged as a weighted sum of the two individual and independent contributions provided the absence of significant interactions among the americium ions. Thus, Riddle et al.6obtained structural parameters of the present species, assuming a complex model including contributions from the two oxidation states, Am(III) and Am(VI). For the americyl, these authors reported Am-Oyl and Am-OIdistances of 1.69 Å and 2.44 Å, respectively. These values were obtained from a fitting of a mixture solution composed by 70% of Am(VI), ∼29% of Am(III) and a marginal presence of Am(V) (∼1%). A first-shell coordination number of 1.3 for Oyl, which results from the main contribution of two Oyl back-scatterers weighted by the percentage of americyl in the sample (∼70%), and 4.1±0.8 which roughly results from the backscattering contribution of five oxygen atoms of the first hydration shell weighted by the same percentage of americyl in the sample were obtained. Debye-Waller factors, which reflect the structural and dynamical disorder of a given shell 3 of back-scatterers surrounding the metal cation, surprisingly shows smaller stiffness for the Am=Oyl covalent bonds than for the Am-OH2intermolecular bonds, 0.006 Å2and 0.002 Å2, respectively, a fact difficult to rationalize. Because the EXAFS spectrum corresponds to a mixture of oxidation states as opposed to the most common EXAFS fitting scenario involving only one well-defined species as absrober, this is a challenging experiment which needs additional information to fully resolve the hydration structure of the cation mixture present in the sample. Computational modeling of actinoids has become an interesting alternative or complement to experiment and is currently a well-established tool.18 Nevertheless, published theoretical studies15,17,19–22 on americium compounds in solution are scarce compared to other actinoids, although the experience from other actinoids should be partially transferable. Computation of EXAFS spectra from the structural information obtained by statistical simulations can contribute, as previously shown, to the interpretation of the experimental spectra.13,15,17,23–25 Conversely, the agreement with experimental results supports the interaction potentials employed as well as the physicochemical properties derived from the simulation. The aim of this work is to rationalize the experimental XAS spectra6of the [AmO2]2+ /Am3+ mixture in solution from a theoretical view, identifying individual contributions. For this purpose, XAS spectra will be computed from the structural information provided by MD simulations. In this case, both cations form stable aqua ions in solution, [AmO2(H2O)n]2+ and [Am(H2O)m]3+, with n=5 and m=8-9. The first task is the development of specific ab initio interaction potentials of these cations with water. To fulfill this requirement we use our Hydrated Ion Model,26 which is based on the assumption that the cationic reference species in solution is the aqua ion, i.e. the hydrated ion.13,26,27 Recently we have published16 a classical interaction potential, based on first-principles calculations, to describe uranyl in water considering the hydrated uranyl, [UO2(H2O)5]2+ , instead of the naked ion [UO2]2+ in aqueous solution. From classical MD simulations, we were able to successfully simulate its EXAFS spectrum and investigate the hydration dynamics, structure and thermodynamics of 4 aqueous uranyl. The performance of this potential has been shown recently in the description of uranyl in the liquid vein of a clay.28 Computational Methods The Hydrated ion model is a classical electrochemical concept developed to better describe the physicochemical properties of highly-charged metal cations.29,30 This model recognizes the hydrated ion as the representative solute instead of the naked ion. This is necessary because the polarization exerted by the ion on its first shell is such that first-shell water molecules must be considered to be different molecular species than bulk water molecules.31 Therefore, first-shell water molecules must have a different description than bulk water molecules and their charges and interaction potential parameters must be developed specifically for them. This imposes a restriction: the model is only applicable to hydrated ions whose first-shell residence times are longer than the simulation time. This restriction is due to the fact first-shell water release would lead to an arrangement where one or several water molecules with fitted parameters to be in the first shell are in the bulk. To get the ab initio interaction energies, all QM calculations are carried out in the presence of the full first hydration sphere. This procedure has two main advantages: the first one is completely general, it avoids the overpolarization that would occur if only one water molecule was considered to describe the ion-water interaction; the second one concerns the case of open-shell metal cations, the ion is in the correct electronic state in solution due to the presence of the ligand field provided by the hydration shell. Additional information about the Hydrated Ion Model and its extension allowing the first-shell water molecule exchange on the basis of a polarizable can be found elsewhere.13,16,32–35 B3LYP36,37 reproduces reasonably well the properties of actinyl cations38,39 treating the multireference problem in an effective way. In order to be consistent with the previous uranyl potential, the level of theory chosen for [AmO2(H2O)5]2+ was B3LYP employing the 5 semi-relativistic Stuttgart pseudopotential40 with its recommended basis for americium and aug-cc-PVDZ for light atoms.41 The americyl-pentahydrate is a quartet and B3LYP wavefunction shows low spin contamination, <S(S+1)> = 3.84 instead of 3.75, the pure-spin eigenvalue. To check the structural and energy results we have carried out a NEVPT242–44 computation of the americyl-hydrate including 9 electrons and 10 orbitals in the active space. Basis sets ma-def2-TZVP45 on oxygen atom, def2-SVP46 on hydrogen atom and DEF-TZVP47/SD(60,MWB) pseudopotential for Am40 were used. Our B3LYP optimizations produce Am-Oyl and Am-OIbondlengths of 1.71 Å and 2.48 Å, respectively, and an interaction energy of -226.2 kcal mol−1including counterpoise correction. NEVPT2 as implemented in the ORCA program48 provides very similar results: optimized distances of 1.75 Å and 2.42 Å for Am-Oyl and Am-OIbondlengths, respectively, and an interaction energy of -239.4 kcal mol−1including counterpoise correction. If the structural disagreement was substantial the simulated EXAFS, which will be presented later on, would deviate strongly from experiment. Since such is not the case, B3LYP seems an adequate computational level for the purposes of this article, i.e. to understand the nature of the Am(VI)/Am(III) mixture. The optimized geometries and interaction energies of Am(VI) at both levels of series are included in Table S1 of the Supporting Information (SI). The method used to develop the interaction potential of uranyl,16 which is a general strategy for any hydrated actinyl cation, was applied on the americyl-pentahydrate. The total interaction potential of an actinyl in solution within the Hydrated Ion Model, as developed for the uranyl case,16 is the result of combining three interaction potentials: one to describe the flexibility of the actinyl molecular cation, (O=An=O), a second intermolecular potential between the actinyl and the water molecules forming the hydrate, ([AnO2]2+−H2O), which defines the intrinsic dynamic of the pentahydrated actinyl, ([AnO2(H2O)5]2+) and a third intermolecular potential to describe the interaction of bulk water molecules with the actinyl hydrate, ([AnO2(H2O)5]2+ – H2O). This implies that the model considers two types of water molecules in solution: first-shell and bulk water molecules. The intramolecular interaction 6 potential within the [AmO2]2+ hydrated ion was developed specifically for this species. The non-electrostatic interaction contribution between bulk water molecules and the hydrated ion is the same as for the uranyl case since it was found to be fairly unspecific to the actinoid (Figure S1 of the SI). The interaction potential parameters can be found in Tables S2-S6 of SI. The level of theory chosen to develop the potential for [Am(H2O)8]3+ was MP2 using the semi-relativistic pseudopotential of Sttugart,49 which was specifically developed to treat the Am3+ cation, with its recommended basis set on americium and cc-PVTZ41,50–52 on light atoms. All quantum-mechanical calculations were run with Gaussian09.53 Dolg and col.49 developed a series of specific pseudopotentials for the trivalent cations of the actinoid series, including the f-electrons in the pseudopotential. In this way the challenging issue of the f-electron open-shell is not longer found. This series of pseudopotentials is particularly convenient to deal with the actinoid hydration, as Dolg. and col.54 showed by studying the actinoid hydration of the whole An3+ series with the semicontinuum model. Am3+, like most large trivalent cations, must have a first-shell water mean residence time in the range 2-10 ns.17,55 The aim of the Am3+-H2O interaction potential is to get a reasonable representation of the solution structure validated by realistic simulated EXAFS and XANES spectra. Therefore, for Am3+ the coordination number of the first shell was kept fixed at a given hydration number following our Hydrated Ion model.26,34 Because the coordination number along the actinoid series is known to change from 9 to 8, we have considered both Am3+ hydrates, [Am(H2O)m]3+ for m=8 and 9, to develop an intermolecular potential with water. This gives us a reasonable range of possible Am3+ hydrates in solution. The gas phase optimized Am-O distances for the octahydrate and enneahydrate are 2.51 Å and 2.56 Å, respectively. The optimized geometries and interaction energies of Am(III) with both coordination numbers are included in Table S1 of the SI. The first-shell water molecule geometry was that of the gas phase QM optimization of [Am(H2O)m]3+. RESP56 effective charges obtained from PCM57,58 polarized wavefunction were assigned to the first-shell water molecules. The 7 employed water models are not polarizable and therefore polarization of the charges of the hydrated ion by the bulk solvent is included in a mean field sense with the PCM method. All water-water van der Waals interactions were described by the TIP4P59 Lennard-Jones parameters. Additionally angular harmonic interaction potentials (O-Am-O) for first-shell oxygen atoms were incorporated. The first-shell water oxygen-metal interaction was modeled as a harmonic bonding potential. The angular and bonding harmonic interaction terms are included in order to preserve the first-shell coordination integrity. The interaction potential parameters can be found in Tables S2, S3, S7 and S8 of SI. 5 ns molecular dynamics simulations were run for each of the americium species. Simulation boxes contained the ion and 1500 water molecules including first-shell water molecules. The TIP4P59 model was employed for bulk water molecules. The simulation ensemble was NPT at 300 K and 1 atm. The time step chosen was 1 fs and all water molecules were kept rigid. Electrostatic interactions were calculated using the Ewald Summation method with cutoff radius of 14 Å and 10−6for relative error. The program used was DL_POLY CLASSIC.60 Details of the simulation can be found elsewhere.16 Results and Discussion Table 1 summarizes the main results of our simulations and the published experimental works. In the rest of the section we detail the our results. Figure 1 displays the Am-O and Am-H RDFs for the simulations of Am3+ in water (top), which includes the results derived from the intermolecular potentials assuming the Am(III) octahydrate (solid lines) and ennea-hydrate (dashed lines), and [AmO2]2+ (bottom). The Am(III) hydration structure derived from the [Am(H2O)8]3+−H2O potential is dominated by a well-defined first hydration shell with two sharp peaks centered at 2.47 Å for Am-O and 3.15 Å for Am-H pairs. The second shell is formed by 19-20 water molecules with Am-O and Am-H peaks centered at 4.64 Å and 5.21 Å, respectively. When the [Am(H2O)9]3+−H2O 8 Table 1: Simulation and experimental results of Am(III) and Am(VI) in solution. The properties displayed are translational self diffusion coefficient (Di),61 hydration enthalpy (∆Hhydr),30 metal-oxygen distances (r),7–9 coordination numbers (CN)7–9 and Debye-Waller factors (DW).7–9 [AmO2·(H2O)5]2+ [Am·(H2O)n]3+ MD Exp. MD n= 8 MD n= 9 Exp. DiÂů105(cm2s−1) 1.3 ±0.1 ∼0.7 ±0.2 0.8 ±0.2 0.625 ±0.003 ∆Hhydr, (kcal mol−1) -342 ∼-325,∼-401 -870 -886 -823 r(Am-Oyl), (Å) 1.72 1.69 - - - CN(Am-Oyl) 2∗1.3 - - - DW(Am-Oyl), (Å2) 0.007 0.006 - - - r(Am-OI), (Å) 2.44 2.48 2.47 2.53 2.473 CN(Am-OI) 5∗4.1 8∗9∗7.4, 8.4, 10 r(Am-OII), (Å) 4.64 - 4.66 4.72 2.473 CN(Am-OII) 29∗- 19 19 - DW(Am-OI), (Å2) 0.007 0.002 0.006 0.007 0.007, 0.009 ∗Fixed. 9 spectra is not satisfactory. As suggested by Riddle et al.6the relative instability of the Am(VI) oxidation state leads to a mixture of Am(VI) and Am(III) in the solution and a marginal presence of Am(V). Figure 2 (bottom) shows the comparison of the experimental (dashed black line) with the theoretical spectra (blue and magenta lines) obtained from two different weighted sums of the Am3+ and [AmO2]2+ simulated spectra shown in the two upper plots. The much better agreement gives evidence for the presence of an Am cationic mixture in the aqueous solutions recorded by Riddle et al.6In our case, the best [AmO2]2+ /Am3+ ratio matching as much as possible the experimental signal was achieved with a 55/45 value (blue line), whereas the suggested experimental value was 70/30 (magenta line). Although both ratios give simulated spectra similar to the experimental one, our proposed ratio performs better. The observed discrepancy of ratios is probably a consequence of the uncertainty in both the theoretical procedure and the experimental determination of the Am species concentration in the measured sample. The MD simulation used for the Am3+ contribution to the simulated spectrum was that using the [Am(H2O)8]3+ hydrate. The comparison of the experimental results with the simulated spectrum obtained using the Am(III) ennea-hydrate MD simulation is plotted in Figure S4 of SI (bottom). It is clear the better theoretical-experimental agreement derived from the use of the Am(III) octahydrate EXAFS spectrum. Two conclusions can be derived from the analysis of Figure 2 and Figure S4 of SI. The first one is that, even though the high oxidizing character of the medium, the experimental spectrum clearly is the combination of the two Am cationic species regardless the cation ratio applied and the Am(III) hydrate selected to combine. The second conlusion is that the best agreement obtained with a different cation ratio to that experimentally proposed, but not too far, suggests some uncertainty in the individual Am species population really present in the measured sample. As a consequence of this analysis we can establish that the simulated EXAFS spectrum (red line) in the middle of Figure 2 is our theoretical proposal for the spectrum of an [AmO2]2+ dilute aqueous solution. An additional test on the cationic mixture aqueous solution can be made by the theoretical- 16 Figure 3: Experimental (dashed lines) vs. simulated (solid lines) Am L3-edge k2-weighted XANES spectra. Top: experimental (ref.[ 7]) and simulated Am3+ aqueous solution; Middle: experimental Am ionic mixture (ref.[ 6]) and simulated spectrum of [AmO2]2+ (red line); Bottom: experimental Am ionic mixture (ref.[ 6] and simulated spectra with the 70/30 (magenta line) and a 55/45 (blue line) ([AmO2]2+ /Am3+) ratios. 17 experimental comparison of the XANES spectra of the same samples. Figure 3 displays such test for the Am(III) sample (left part) and Am(VI) sample (middle part). The agreement between the experiment (black dashed line) and the theoretical model using the [Am(H2O)8]3+ (green line) is remarkable. On the contrary, there is a poor agreement of the experimental XANES (black dashed line in the center part) of the sample containing the Am(VI) oxidation state and the simulated spectrum derived from the pure [AmO2]2+ MD simulation. However, when the spectrum is computed considering the [AmO2]2+ /Am3+ mixture, the agreement is clearly improved. Again the 55/45 ratio (blue line in the right part of Figure 3) performs better than the 70/30 experimental proposal (magenta line). If we accept that this satisfactory result is not a consequence of fortuitous error cancellations, the model suggests that the experimental ratio of the mixture might be reconsidered. Furthermore, as in the EXAFS case, we propose the red line in the center of Figure 3 as the XANES spectrum of a pure sample of [AmO2]2+ in water, something that up to the present has not been measured. The larger stability of higher oxidation states in water for [UO2]2+ and [PuO2]2+ allows the experimental determination of the XANES for aqueous solutions only containing those actinyls.68,69 The energy gap between the white line and the first resonance is ∼18 and ∼21 eV for [UO2]2+ and [PuO2]2+ XANES spectra, respectively. In the case of our simulated pure [AmO2]2+ aqueous solution this gap is 23 eV, what corroborates the global consistency among experimental and theoretical results. It is interesting to point out that this gap in the experimental spectrum of the mixture is ∼30 eV. The energy position of the edge in XANES has long been used as a semi-quantitative index on the absorber atom oxidation state. For systems with O atoms in the first hydration shell there is a linear correlation between the oxidation state and the edge energy, with an increase of ∼1.5 eV per formal oxidation state.70 FEFF9.6 computations of the XANES spectra of aqueous solutions of the Am(III) and Am(VI) oxidation states (left:green line, and middle:red line in Figure 3) predicts 4.0 eV to compare with the empirically estimated of 4.5 eV. However, when analyzing the experimental XANES spectra (dashed lines on Figure 3, left:Am(III), middle or 18 right:Am(VI)) the difference is 2.0 eV whereas the value from the theoretical XANES spectra of Am(III) (left:green line) and the cationic mixture (right:blue line) is 2.5 eV. In conclusion, this study has established a spectroscopical-theoretical procedure to deal with an aqueous solution mixture of two americium cations, Am3+ and [AmO2]2+ which contribute separately to a sole X-ray absorption spectrum. From this global experimental signal, the individual contributions to the XANES and EXAFS spectra corresponding to the pure aqueous solutions of each individual ionic species is achieved based on reliable intermolecular potentials and ab initio methods of multiple scattering theory. Discarding error cancellations, if the sum of the two contributions leads to a fair agreement with experiment, we could assume as correct the individual contributions and consequently to support the simulated EXAFS and XANES spectra of [AmO2]2+ as the expected Am L3-edge spectra of a pure [AmO2]2+ aqueous solution. This may be envisaged as a challenge for new experiments intended to design appropriate medium conditions to stabilize selectively the [AmO2]2+ species in water. The combined experimental-theoretical strategy opens possibilities to the analysis of complex mixtures solutions of species containing a particular absorber atom present in different oxidation states. Acknowledgement We would like to thank Prof. Steven Conradson for providing us some of the experimental spectra. This work was financially supported by Consejeria de Economia, Innovacion, Ciencia y Desarrollo, Junta de Andalucía of Spain (Proyecto de Excelencia, P11-FQM7607) and the Spanish Ministry of Education, Culture and Sports for the PhD scholarship FPU (Formación de Profesorado Universitario). 19 Supporting Information Available The Supporting Information is available free of charge on the ACS Publications website at DOI: Cartesian coordinates of optimized structures of [Am(H2O)8]3+, [Am(H2O)9]3+ and AmO 2+ 2, EXAFS and XANES FEFF input files, Figure comparing interaction energy scans of a bulk water molecule with [AmO2(H2O)5]2+ given by DFT and the developed classical potential, gas phase quantum-mechanical optimized geometries of the Am3+ and [AmO2]2+ hydrates, interaction potential parameters, angle-solved RDFs, theoretical-experimental comparison of the EXAFS spectrum corresponding to: (Top) simulated spectra built from [Am(H2O)8]3+ and [Am(H2O)9]3+ MD simulations, (Bottom) simulated spectra built with the 55/45 ratio using for Am(III) the [Am(H2O)9]3+ MD simulation. References (1) Runde, R. W.; Schulz, W. W. In The Chemistry of the Actinide and Transactinide Elements, 3rd ed.; Morss, L. R., Edelstein, N., Fuger, J., Katz, J. J., Eds.; Springer, 2008; Vol. 2; Chapter 8, pp 1265–1395. (2) Morimoto, K.; Kato, M.; Uno, H.; Hanari, A.; Tamura, T.; Sugata, H.; Sunaoshi, T.; Kono, S. Preparation and Characterization of (Pu, U, Np, Am, simulated FP)O2−x.J. Phys. Chem. Solids 2005,66, 634–638. (3) Yoshimochi, H.; Nemoto, M.; Mondo, K.; Koyama, S.; Namekawa, T. Fabrication Technology for MOX Fuel Containing AmO2by an in-Cell Remote Process. J. Nucl. Sci. Technol. 2004,41, 850–856. (4) Prieur, D.; Carvajal-Nunez, U.; Vitova, T.; Somers, J. Local and Electronic Structure of Americium-Bearing PuO2. Eur. J. Inorg. Chem. 2013,2013, 1518–1524. 20 (5) Choppin, G. R.; Jensen, M. P. In The Chemistry of the Actinide and Transactinide Elements, 3rd ed.; Morss, L. R., Edelstein, N., Fuger, J., Katz, J. J., Eds.; Springer, 2008; Vol. 4; Chapter 23, pp 2524–2621. (6) Riddle, C.; Czerwinski, K.; Kim, E.; Paviet, P.; Weck, P.; Poineau, F.; Conradson, S. Characterization of Pentavalent and Hexavalent Americium Complexes in Nitric Acid Using X-Ray Absorption Fine Structure Spectroscopy and First-Principles Modeling. J. Radioanal. Nucl. Chem. 2016,309, 1087–1095. (7) Stumpf, S.; Stumpf, T.; Dardenne, K.; Hennig, C.; Foerstendorf, H.; Klenze, R.; Fanghänel, T. Sorption of Am (III) Onto 6-Line-Ferrihydrite and its Alteration Products: Investigations by EXAFS. Environ. Sci. Techn. 2006,40, 3522–3528. (8) Stumpf, T.; Hennig, C.; Bauer, A.; Denecke, M. A.; Fanghänel, T. An EXAFS and TRLFS Study of the Sorption of Trivalent Actinides Onto Smectite and Kaolinite. Radiochim. Acta 2004,92, 133–138. (9) Allen, P.; Bucher, J.; Shuh, D.; Edelstein, N.; Craig, I. Coordination Chemistry of Trivalent Lanthanide and Actinide Ions in Dilute and Concentrated Chloride Solutions. Inorg. Chem. 2000,39, 595–601. (10) Koningsberger, D.; Prins, R. X-ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS, and XANES; John Wiley and Sons: New York, NY, 1988. (11) Muñoz-Páez, A.; Sánchez Marcos, E. In Comprehensive Inorganic Chemistry II: From Elements to Applications, 2nd ed.; Reedijk, J., Poeppelmeier, K., Eds.; Elsevier, 2013; Vol. 9; pp 133–159. (12) Denecke, M. A. Actinide Speciation Using X-Ray Absorption Fine Structure Spectroscopy. Coord. Chem. Rev. 2006,250, 730–754. 21 (13) Galbis, E.; Hernández-Cobos, J.; den Auwer, C.; Le Naour, C.; Guillaumont, D.; Simoni, E.; Pappalardo, R. R.; Sánchez Marcos, E. Solving the Hydration Structure of the Heaviest Actinide Aqua Ion Known: The Californium (III) Case. Angew. Chem. Int. Ed. 2010,122, 3899–3903. (14) Atta-Fyn, R.; Bylaska, E. J.; Schenter, G. K.; de Jong, W. A. Hydration Shell Structure and Dynamics of Curium(III) in Aqueous Solution: First Principles and Empirical Studies. J. Phys. Chem. A 2011,115, 4665–4677. (15) D’Angelo, P.; Martelli, F.; Spezia, R.; Filipponi, A.; Denecke, M. A. Hydration Properties and Ionic Radii of Actinide (III) Ions in Aqueous Solution. Inorg. Chem. 2013, 52, 10318–10324. (16) Pérez-Conesa, S.; Torrico, F.; Martínez, J. M.; Pappalardo, R. R.; Sánchez Marcos, E. a Hydrated Ion Model of [UO2]2+ in Water: Structure, Dynamics, and Spectroscopy from Classical Molecular Dynamics. J. Chem. Phys. 2016,145, 224502. (17) Spezia, R.; Migliorati, V.; D’Angelo, P. On the Development of Polarizable and Lennard-Jones Force Fields to Study Hydration Structure and Dynamics of Actinide(III) Ions Based on Effective Ionic Radii. J. Chem. Phys. 2017,147, 161707. (18) Dolg, M. Computational Methods in Lanthanide and Actinide Chemistry; John Wiley & Sons: University of Cologne, Germany, 2015. (19) Vallet, V.; Privalov, T.; Wahlgren, U.; Grenthe, I. The Mechanism of Water Exchange in AmO2(H2O)2+ 5and in the Isoelectronic UO2(H2O)2+ 5and NpO2(H2O)2+ 5Complexes as Studied by Quantum Chemical Methods. J. Am. Chem. Soc. 2004,126, 7766–7767. (20) Pomogaev, V.; Tiwari, S. P.; Rai, N.; Goff, G. S.; Runde, W.; Schneider, W. F.; Maginn, E. J. Development and Application of Effective Pairwise Potentials for UOn+ 2, NpOn+ 2, PuOn+ 2, and AmOn+ 2(n= 1, 2) Ions With Water. Phys. Chem. Chem. Phys. 2013,15, 15954–15963. 22 (21) Tiwari, S. P.; Rai, N.; Maginn, E. J. Dynamics of Actinyl Ions in Water: A Molecular Dynamics Simulation Study. Phys. Chem. Chem. Phys. 2014,16, 8060–8069. (22) Fabrizio, A.; Rotzinger, F. P. Quantum Chemical Study of the Water Exchange Mechanism of the Americyl (VI) Aqua Ion. Inorg. Chem. 2016,55, 11147–11152. (23) Filipponi, A.; D’Angelo, P.; Pavel, N. V.; Di Cicco, A. Triplet Correlations in the Hydration Shell of Aquaions. Chem. Phys. Lett. 1994,225, 150–155. (24) Palmer, B. J.; Pfund, D. M.; Fulton, J. L. Direct Modeling of EXAFS Spectra from Molecular Dynamics Simulations. J. Phys. Chem. 1996,100, 13393–13398. (25) Merkling, P. J.; Muñoz-Páez, A.; Sánchez Marcos, E. Exploring the Capabilities of X- Ray Absorption Spectroscopy for Determining the Structure of Electrolyte Solutions: Computed Spectra for Cr3+ or Rh3+ in Water Based on Molecular Dynamics. J. Am. Chem. Soc. 2002,124, 10911–10920. (26) Martínez, J. M.; Pappalardo, R. R.; Sánchez Marcos, E. First-Principles Ion-Water Interaction Potentials for Highly Charged Monoatomic Cations.Computer Simulations of Al3+, Mg2+ and Be2+.J. Am. Chem. Soc. 1999,121, 3175–3184. (27) Morales Negrito, N. Estudio Teórico de Propiedades Fisicoquímicas de Cationes Metálicos En Disolución: Evolución en el Grupo de los Alcalinos y en la Serie de los Lantánidos. Ph.D. thesis, Universidad de Sevilla, Spain, 2015. (28) Pérez-Conesa, S.; Martínez, J. M.; Sánchez Marcos, E. Hydration and Diffusion Mechanism of Uranyl in Montmorillonite Clay: Molecular Dynamics Using an Ab Initio Potential. J. Phys. Chem. C 2017,121, 27437–27444. (29) Frank, H. S.; Evans, M. W. Free Volume and Entropy in Condensed Systems III. J. Chem. Phys. 1945,13, 507–532. 23 (30) Marcus, Y. Ions in Solution and their Solvation; John Wiley and Sons, 2015; pp 107– 116. (31) Taube, H. Use of Oxygen Isotope Effects in Study of Hydration of Ions. J. Phys. Chem. 1954,58, 523–528. (32) Pappalardo, R. R.; Sánchez Marcos, E. Recovering the concept of the hydrated ion for modeling ionic solutions: a Monte Carlo study of zinc(2+) in water. J. Phys. Chem. 1993,97, 4500–4504. (33) Pappalardo, R. R.; Martínez, J. M.; Sánchez Marcos, E. Application of the Hydrated Ion Concept for Modeling Aqueous Solutions Containing Highly Charged Ions: A Monte Carlo Simulation of Cr3+ in Water Using an ab Initio Intermolecular Potential. J. Phys. Chem. 1996,100, 11748–11754. (34) Martínez, J. M.; Pappalardo, R. R.; Sánchez Marcos, E.; Refson, K.; Díaz-Moreno, S.; A., M. Dynamics of a Highly Charged Ion in Aqueous Solutions : Molecular Dynamics Simulations of Dilute CrCl3Aqueous Solutions Using Interaction Potentials Based on the Hydrated Ion Concept. J Phys Chem B 1998,102, 3272–3282. (35) Galbis, E.; Hernández-Cobos, J.; Pappalardo, R.; Sanchez Marcos, E. Collecting Highorder Interactions in an Effective Pairwise Intermolecular Potential Using the Hydrated Ion Concept: The Hydration of Cf(III). J. Chem. Phys. 2014,140, 214104–(1)–(11). (36) Becke, A. D. Density-Functional Thermochemistry. III. The Role of Exact Exchange. J. Chem. Phys. 1993,98, 5648–5652. (37) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. Ab Initio Calculation of Vibrational Absorption and Circular Dichroism Spectra Using Density Functional Force Fields. J. Phys. Chem. 1994,98, 11623–11627. 24 (38) Clavaguéra-Sarrio, C.; Vallet, V.; Maynau, D.; Marsden, C. J. Can Density Functional Methods Be Used for Open-Shell Actinide Molecules? Comparison With Multiconfigurational Spin-Orbit Studies. J. Chem. Phys. 2004,121, 5312–5321. (39) Averkiev, B. B.; Mantina, M.; Valero, R.; Infante, I.; Kovacs, A.; Truhlar, D. G.; Gagliardi, L. How Accurate are Electronic Structure Methods for Actinoid Chemistry? Theor. Chem. Acc. 2011,129, 657–666. (40) Küchle, W.; Dolg, M.; Stoll, H.; Preuss, J. Energy-Adjusted Pseudopotentials for the Actinides. Parameter Sets and Test Calculations for Thorium and Thorium Monoxide. J. Chem. Phys. 1994,100, 7535–7542. (41) Kendall, R. A.; Dunning Jr., T. H.; Harrison, R. J. Electron Affinities of the First-Row Atoms Revisited. Systematic Basis Sets and Wave Functions. J. Chem. Phys. 1992, 96, 6796–6806. (42) Angeli, C.; Cimiraglia, R.; Malrieu, J.-P. N-electron valence state perturbation theory: a fast implementation of the strongly contracted variant. Chem. Phys. Lett. 2001,350, 297–305. (43) Angeli, C.; Cimiraglia, R.; Malrieu, J.-P. n-electron valence state perturbation theory: A spinless formulation and an efficient implementation of the strongly contracted and of the partially contracted variants. J. Chem. Phys. 2002,117, 9138–9153. (44) Jung, J.; Atanasov, M.; Neese, F. Ab Initio Ligand-Field Theory Analysis and Covalency Trends in Actinide and Lanthanide Free Ions and Octahedral Complexes. Inorg. Chem. 2017,56, 8802–8816. (45) Zheng, J.; Xu, X.; Truhlar, D. G. Minimally augmented Karlsruhe basis sets. Theoretical Chemistry Accounts 2011,128, 295–305. 25