Structure and Dynamics of Uranyl(VI) and Plutonyl(VI) Cations in Ionic

Subscriber access provided by ANDREWS UNIV

Article

Structure and Dynamics of Uranyl(VI) and Plutonyl(VI) Cations in Ionic Liquid/Water Mixtures Via Molecular Dynamics Simulations Katie Maerzke, George S Goff, Wolfgang Runde, William F. Schneider, and Edward J. Maginn J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/jp405473b • Publication Date (Web): 21 Aug 2013 Downloaded from http://pubs.acs.org on August 25, 2013

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

submitted for publication to J. Phys. Chem. B August 15, 2013

Structure and Dynamics of Uranyl(VI) and Plutonyl(VI) Cations in Ionic Liquid/Water Mixtures via Molecular Dynamics Simulations Katie A. Maerzke,† George S. Goff,‡ Wolfgang H. Runde,‡ William F. Schneider†,§ and Edward J. Maginn†,∗ Department of Chemical and Biomolecular Engineering, University of Notre Dame, Notre Dame, IN 46556 USA Chemistry Division, Los Alamos National Laboratory, Los Alamos, NM 87545 USA Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN 46556 USA Abstract A fundamental understanding of the behavior of actinides in ionic liquids is required to develop advanced separation technologies. Spectroscopic measurements indicate a change in the coordination of uranyl in the hydrophobic ionic liquid 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([EMIM][Tf2 N]) as water is added to the system. Molecular dy2+ namics simulation of dilute uranyl (UO2+ 2 ) and plutonyl (PuO2 ) solutions in [EMIM][Tf2 N]/water mixtures have been performed in order to examine the molecular-level coordination and dynamics of the actinyl cation (AnO2+ 2 ; An = U, Pu) as the amount of water in the system changes. The simulations show that the actinyl cation has a strong preference for a first solvation shell with five oxygen atoms, although a higher coordination number is possible in mixtures with little or no water. Water is a much stronger ligand for the actinyl cation than Tf2 N, with even very small amounts of water displacing Tf2 N from the first solvation shell. When enough water is present, the inner coordination sphere of each actinyl cation contains five water molecules without any Tf2 N. Water also populates the second solvation shell, although it does not completely displace the Tf2 N. At high water concentrations a significant fraction of the water is found in the bulk ionic liquid, where it primarily coordinates with the Tf2 N anion. Potential of mean force simulations show that the progressive addition of up to five water molecules to uranyl is very favorable, with ∆G ranging from −52.3 kJ/mol for the addition of the first water molecule to −37.6 kJ/mol for the addition of the fifth. Uranyl and plutonyl dimers formed via bridging Tf2 N ligands are found in [EMIM][Tf2 N] and in mixtures with very small amounts of water. Potential of mean force calculations confirm that the dimeric complexes are stable, with relative free energies of up to −9 kJ/mol in pure [EMIM][Tf2 N]. We find that the self-diffusion coefficients for all the components in the mixture increase as the water content increases, with the largest increase for water and the smallest increase for the ionic liquid cation and anion. The velocity autocorrelation functions also indicate changes in structure and dynamics as the water content changes.

Keywords: actinides, separations, aqueous, association †

University of Notre Dame, Department of Chemical and Biomolecular Engineering Los Alamos National Laboratory § University of Notre Dame, Department of Chemistry and Biochemistry ∗ Corresponding author. Address: 182A Fitzpatrick Hall, University of Notre Dame, Notre Dame, IN 46556. Telephone: 574-631-5687. Email: [email protected] ‡

1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 42

1 Introduction The renewed interest in the expansion of nuclear energy to meet future energy demands includes the search for novel technologies for the separation of actinide elements in advanced nuclear fuel cycles. Currently, the only technology deployed on a large scale for the reprocessing of used nuclear fuel (UNF) is the PUREX process (Plutonium-Uranium REduction Extraction) which involves the dissolution of the UNF into concentrated nitric acid followed by multiple liquid-liquid extraction processes for partitioning various fission products. 1,2 Ionic liquids (ILs) have recently been identified as a novel class of media with potential application to nuclear separations. ILs are low-melting salts, often containing an asymmetrical organic cation and an organic or inorganic anion. They are of particular interest because they offer a wide range of tunable physical properties, including viscosity, hydrophobicity, conductivity, and liquidus range, which can be modified by choice of the cations and anions. ILs have negligible vapor pressure, are often non-flammable and of low corrosivity, can have high thermal stabilities and often exhibit wide electrochemical windows of up to 6 V. 3 All of these properties make them attractive for use in separation processes relevant to the nuclear industry such as solvent extraction, ion exchange, dissolution, crystallization, and electrorefining. 4 While the current understanding of actinide chemistry in ILs has been discussed in several recent review articles, 4–11 many questions still remain to be explored. Water is a common impurity in IL systems, which can serve as a coordinating ligand to complex actinide (An) ions, particularly in ILs containing weakly complexing anions such as bis(trifluoromethylsulfonyl)imide (Tf2 N). Water is a deterimental impurity for electrochemical separations processes, and plays an integral role during liquid-liquid solvent extraction. Understanding the chemical interactions between a solute and solvent can provide a molecular-level understanding of the solvent’s ability to stabilize solution species. ILs are profoundly different from molecular solvents due to their inherently high ionic strength and nanostructural ordering, 12–14 which significantly impacts both the chemical potential of dissolved species and the local coordination environment. Molecular dynamics (MD) simulations can provide valuable information about the interactions between An solutes and ILs which can be difficult to obtain experimentally. Atomistic-level simulations have been successful at interrogating the properties of ILs in both pure 15 and aqueous 16,17 environments. A range of different computational studies have also been conducted on actinides. Quantum mechanical calculations 18 as well as first principle and classical MD simulations 19 have been performed on small actinyl complexes. Although uranyl(VI) is the most widely studied of the actinyl cations, a limited number of quantum calculations have also been performed on the plutonyl cation. 20–26 These types of simulations can provide exceptionally detailed pictures of the coordination enviromment, dynamics, and energetics of different species in


Page 3 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solution. Given the difficulties associated with studying actindes in the laboratory, computational studies of these systems are an important complement to experimental investigations. The first MD simulation study of a uranyl(VI) ion in an ionic liquid was performed by Chaumont et al. in 2003. They examined uranyl nitrates and chlorides in the ionic liquid 1-nbutyl-3-methylimidazolium hexafluorophosphate (or [BMIM][PF6 ] for short). 27 It was found that the uranyl cations and neutral uranyl nitrate or chloride complexes are surrounded by PF6 anions, followed by a second solvation shell of BMIM cations. This work was followed by studies of uranyl and europium cations in an ionic liquid comprised of 1-ethyl-3-methylimidazolium ([EMIM]) cations paired with a mixture of AlCl4 and Cl anions. 28 It was found that uranyl has a strong tendency to form UO2 Cl2− 4 complexes. Hydrated europium complexes were able to exchange water molecules with chloride anions, though the number of coordinated water molecules depended on the ionic liquid anion, suggesting that trace amounts of water and the identity of the ionic liquid anions can have an impact on the coordination environment. Uranyl cations and uranyl chloride complexes were simulated in dry and “humid” [BMIM][PF6 ] to examine the effect of water on the coordination enviroment. 29 It was found that the “naked” uranyl cations (UO2+ 2 ) coordinate with the PF6 anions in the dry IL but rapidly coordinate with five water molecules in the wet IL. In contrast, the uranyl chloride complexes remain stable in both the wet and dry IL, indicating that chloride is a stronger ligand for uranyl than water. Another MD simulation study of uranyl complexes in dry and wet [BMIM][PF6 ] examined uranyl coordination with nitrate and octyl(phenyl)-N, N -diisobutylmethylcarbamoyl phosphine oxide (CMPO). In wet IL solutions the water molecules displace the IL anions. In dry IL the NO3 in the [(UO2 )(NO3 )(CMPO)]+ complex coordinates in a bidentate manner while in a wet IL the coordination is monodentate. 30 Other MD studies have examined the solvation of UCln− complexes in the ILs methyl-tributylammonium 6 bis(trifluoromethylsulfonyl)imide ([MeBu3 N][Tf2 N]) and [BMIM][Tf2 N] under “humid” and dry conditions. 31 It was found that the [BMIM] cation solvates the UCln− better than the MeBu3 N 6 cation, but the effect of water content on solvation was minimal. A joint experimental and computational study showed that the solubility and dissociation of uranyl salts depends on the identity of the IL anion, and that the presence of chloride ions can enhance the solubility of the uranyl salts. 32 Potential of mean force simulations were also performed for the coordination of chloride with uranyl cations in the ionic liquids [BMIM][Tf2 N] and [MeBu3 N][Tf2 N]. 33 The simulations revealed that chloride complexation is extremely favorable, with a free energy of complexation of more than 100 kJ/mol. The solvation around the different chloro-complexes evolves from purely anionic around the uranyl cation to purely cationic around the negatively charged UO2 Cl2− 4 complex. The second solvation shell contains oppositely charged ions, causing “onion-type” layers of alternating charge. Another joint experimental and computational study found that nitrate and chloride are stronger ligands for uranyl than the Tf2 N anion. 34



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 42

Simulations were recently conducted to examine the complexation of perrhenate (ReO− 4 ) anions with uranyl cations in traditional solvents such as water and acetonitrile as well as in three different ILs ([BMIM][Tf2 N], [Bu3 MeN][Tf2 N] and trimethyl-butylammonium bis(trifluoromethylsulfonyl)imide [Me3 BuN][Tf2 N]). 35 ReO− 4 was found to be a weak ligand for uranyl in water, but a strong ligand in acetonitrile. It was also a strong ligand in the ILs, due to the very weak complexing ability of the Tf2 N anion. Potential of mean force calculations in the different ILs show that complexation by three to five perrhenate ions is favorable, with five-coordinate favorable in [BMIM][Tf2 N] and three or four-coordinate favorable in [Bu3 MeN][Tf2 N]. Additionally, the formation of uranyl oligomers cannot be ignored in the IL solutions. Although dimerization was not observed over the course of a 5 ns MD simulation of two UO2 (ReO4 )2− 4 complexes in [BMIM][Tf2 N], PMF calculations suggest that the dimeric complex is more stable than the individual complexes by approximately −15.5 kJ/mol. Because the associated and dissociated states are separated by a relatively large barrier of 33.5 kJ/mol, the timescales for observing the formation of dimers is quite long and difficult to observe over short MD trajectories. While these previous studies have helped shed light on the behavior of uranyl ions in a handful of different ILs, many questions still remain. Sampling limitations associated with relatively short MD trajectories make it difficult to determine equilibrium distributions of strongly coordinating species. For example, uranyl complexes initialized as associated UO2 X2 species exhibited different coordination environments than when the system was initalized as dissociated ions (UO2+ 2 + 2X− ). 32 System size limitations have also resulted in the use of concentrations of water and uranyl ions in simulations that are difficult (if not impossible) to observe experimentally. For example, previous MD simulations of uranyl cations in mixtures of [BMIM][PF6 ] and water were carried out at a water mole fraction of 0.50 29,30 , although the experimental saturation limit of water occurs at a mole fraction of 0.26. 36 It is unlikely that the presence of low concentrations of hydrophilic uranyl cations can alter the solubility limit this much. In addition, previous simulations have only looked at extreme conditions of totally “dry” IL-actinyl systems, or extremely “wet” systems where a large amount of water is present. It is more likley that small amounts of water will be present in these systems, due to the hygroscopic nature of ILs. Very little information is available on what happens at intermediate water concentrations. Finally, all previous simulations of actinide complexes in ionic liquids have been limited to the uranium-containing species; how do other actinide anions such as plutonyl behave in IL-water systems? Here we present MD simulation results that examine the liquid structure, coordination environment and dynamics of uranyl(VI) and plutonyl(VI) cations in [EMIM][Tf2 N]/water mixtures. Water concentrations range from 0.0 − 0.29 mole fraction, where the upper concentration is near the saturation limit. 37 Long simulations are carried out at elevated temperatures to overcome the sampling difficulties noted above. In addition, potential of mean force calculations are con-


Page 5 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ducted to determine free energy profiles for the progressive addition of water to actinyl cations in [EMIM][Tf2 N] as well as for the association between two actinyl cations in [EMIM][Tf2 N] at varying water concentrations.

2 Simulation Details Classical MD simulations were performed to study uranyl(VI) (UO2+ 2 ) and plutonyl(VI) (PuO2+ 2 ) in solutions of the hydrophobic IL [EMIM][Tf2 N] with small amounts of added water at 400 K. Structures of the ions can be found in Figure 1. The simulations were performed at a somewhat elevated temperature of 400 K due to the very slow dynamics at 298 K (see Figure S1 in Supporting Information). Actinide concentrations of 0.1 M were chosen for the simulations as being representative of typical experimental concentrations for spectroscopic characterization. To obtain reasonable statistics, 12 actinyl cations were simulated, resulting in a relatively large system size of 480 [EMIM][Tf2 N] ion pairs for the pure ionic liquid solution. The actinyl cations were added to the solution as AnO2 (Tf2 N)2 (An = U, Pu) resulting in an additional 24 Tf2 N anions in each mixture. To keep the molarity of the solution roughly constant while adding small amounts of water, two [EMIM][Tf2 N] ion pairs were removed and alternatively 27 and 28 water molecules were added. The mole fraction in the mixtures, defined by assuming that each IL ion pair is a “molecule,” ranged from 1.0 for pure [EMIM][Tf2 N] to 0.71 for the systems with the most water, which is at roughly the experimental solubility limit of water in [EMIM][Tf2 N] at 298 K. 37 The solubility of water in [EMIM][Tf2 N] increases slightly with increasing temperature, so given that the simulations are performed at 400 K in the presence of hydrophilic actinyl cations suggests that the simulations are well within the miscibility window. Quality force fields are required to enable accurate prediction of thermodynamic, dynamic, and structural properties via molecular simulation. In this work, we have used the united-atom IL force field of Zhong et al., 38,39 which gives accurate densities and heats of vaporization for pure ILs, and also gives accurate excess molar volumes, excess enthalpies, viscosities, and self-diffusion coefficients for IL/water mixtures. 39 Although the force field yields self-diffusivities of pure ILs that are slightly lower than experiment, the overall performance of this force field is among the best of non-polarizable “class I” IL force fields currently available. This fixed charge, united atom model is computationally efficient, which is important due to the large systems and long simulation times required in the present study. The uranyl(VI) and plutonyl(VI) cations are modeled with a force field developed by our group for for hydrated actinyl(V/VI) cations. 40,41 High-level quantum calculations of actinyl-water potential energy surfaces were used to parameterize the force field. The mean squared error in the 5 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

H

EMIM+ CH3

C

H

Page 6 of 42

C

O

PuO22+

N Pu N

CH3

C O

H

CH2

Tf2N– O O

O

N S

UO22+ U

S CF3

CF3 O

O

O

Figure 1: Structures of the 1-ethyl,3-methylimidazolium (EMIM) cation, bis(trifluoromethylsulfonyl)imide (Tf2 N) anion, and actinyl cations used in this work.

Table 1: Mole fraction [EMIM][Tf2 N] (IL), number of water molecules, number of [EMIM][Tf2 N] ion pairs, ratio of the number of water molecules to actinyl ions used in the present study.

xIL

Nwater

NIL

Nwater /NAn

1.00

0

480

0.00

0.95

28

478

2.33

0.90

55

476

4.58

0.85

83

474

6.92

0.81

110

472

9.17

0.77

138

470

11.50

0.74

165

468

13.75

0.71

193

466

16.08

classical fit to the quantum mechanical potential energy surface, as reported by Pomogaev et al. is 25.6 and 27.3 kJ/mol for uranyl(VI) and plutonyl(VI), respectively. Ignoring the larger errors at very short water-actinyl separations, which are irrelevant configurations for MD simulations under near-ambient conditions, the mean unsigned percentage error in the fit for both actinyl cations is approximately 20%. This approach has been shown to accurately predict liquid structure, selfdiffusion coefficients and hydration free energies for actinyl(V/VI)/water systems for U, Np, Pu, and Am. The force field assigns the actinyl(VI) cations a formal charge of +2 and an equilibrium 6 ACS Paragon Plus Environment

Page 7 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


O=An=O bond angle of 180 degrees. The nominal An=O bond length in the force field is somewhat longer for uranyl (0.1761 nm) than plutonyl (0.1700 nm), and the force field O=An=O angle is more flexible for uranyl than plutonyl, with harmonic force constants of kθ = 198 kJ/mol/rad2 and 602 kJ/mol/rad2 , respectively. While the actinyl cations have a formal charge of +2, the Tf2 N anion in our simulations has a scaled charge of −0.8, resulting in a net charge of +0.4 for each AnO2 (Tf2 N)2 . The use of scaled charges in IL force fields is a common way to account for charge transfer (and improve the dynamics) without explicitly including expensive polarizability terms. 42,43 To maintain charge neutrality in the system, the extra +4.8 charge is compensated for through a uniform background charge of −4.8. Water is represented using the well-known extended simple point charge (SPC/E) model. 44 These fixed-charge models were chosen for their simplicity and efficiency; in addition, the actinyl model does not include polarizability terms, and force field parameterization was not a goal of this work. The molecular dynamics simulation package GROMACS 4.5.5 45–48 was used for all MD simulations. All properties were calculated from 15 ns production runs in the microcanonical (N V E) ensemble and averaged over multiple independent simulations. Umbrella sampling 49,50 simulations were run to calculate the potentials of mean force 51 using the weighted histogram analysis method (WHAM) 52–54 as implemented in the GROMACS analysis routine g wham. 55 Additional simulation details may be found in the Supporting Information.

3 Results and Discussion 3.1 Coordination Environment To examine the coordination enviroment in the equatorial plane of the actinyl(VI) cations, we calculated radial distribution functions (RDFs) and number integrals (NIs) for U and Pu with the oxygen atoms in water and Tf2 N. Results are shown in Figure 2. The position of the maximum of the first peak for the An-O(H2 O) RDF ranges from 0.243 nm for uranyl and 0.245 nm for plutonyl in the 0.95 mole fraction IL mixture to 0.247 nm (for both actinyls) in mixtures with a higher water content. The peak position is at slightly longer distances of 0.251 − 0.257 nm for the An-O(Tf2 N) RDFs (see Table S2 in Supporting Information). The An-O(H2 O) peak positions for mixtures with a high water content are similar to the value of 0.246 nm reported for both uranyl and plutonyl modeled with this force field in SPC/E water. 41 The results are also similar to other values for actinyl complexes in aqueous solution reported in other computational work, 21,23,56–63 though slightly longer than the results of previous quantum mechanical calculations 22,25,26,60,64,65 and the distances of 0.240 − 0.242 nm measured via various experimental techniques. 66–71 The An-O(Tf2 N) peak position is slightly shorter than the previously reported value of 0.26 nm for 7 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 42

uranyl in [BMIM][Tf2 N]. 32 This small difference is likely due to the use of different force fields for both the actinyl cation and IL. The first minimum after the first peak of the An-O(H2 O) and AnO(Tf2 N) RDFs is used to define the first solvation shell (and likewise the first minimum after the second peak for the second solvation shell), resulting in a 0.32 nm boundary for the first solvation shell and a 0.60 nm boundary for the second solvation shell for both uranyl and plutonyl. Given the very small dependence of the location of the minima on the identity of the actinyl and the water content, the same boundaries for the solvation shells were used for all mixtures. Although the first solvation shell is well-defined, the second minimum used to determine the boundary for the second solvation shell is harder to define, especially for Tf2 N, as can be seen from the RDFs and NIs in Figure 2. The problem is even more pronounced if Tf2 N anions are counted based on the positions of the nitrogen atoms, which are roughly the center of mass, rather than oxygen atoms. The elogated nature of the Tf2 N anion makes it difficult to clearly distinguish which anions are in the first and second solvation shells. In addition, a single Tf2 N anion can have oxygen atoms in both the first and second solvation shell, resulting in a split peak in the An-N(Tf2 N) RDF with a first minimum greater than one (see Figure S2). Thus, we focus on the distribution of the oxygen atoms, although for comparison purposes we have also counted the number of Tf2 N nitrogen atoms up to 0.80 nm from an actinide (the minimum of the An-N(Tf2 N) RDF). The large, sharp first peak in the An-O(H2 O) RDF (see Figure 2) indicates a strong preference for the water to be coordinated with the actinide, which is more pronounced at lower concentrations of water. As the water content increases, the height of the first peak decreases. A smaller second peak begins to grow at 0.85 mole fraction IL (0.15 mole fraction water), indicating that water begins to populate the second solvation shell. The height of the second peak increases from 0.85 to 0.81 mole fraction IL, but then changes only very slightly as the water content is further increased. The first peak in the An-O(Tf2 N) RDF, though smaller than the An-O(H2 O) peak, is still quite large for mixtures with little or no water. This indicates reasonably strong association between the actinide and Tf2 N oxygen atoms in cases where very little water is present. For the mixtures with 0.85 mole fraction IL and less, (i.e., those with water concentrations of more than five water molecules per actinyl) the Tf2 N no longer coordinates with the actinide, as can be seen by the disappearance of the first peak in the An-O(Tf2 N) RDF. The second peak in the An-O(Tf2 N) RDF becomes somewhat more broad and slightly shorter with increasing water content as water replaces some of the Tf2 N in the second solvation shell. Average coordination numbers can be found using the value of the number integral at the boundary of the solvation shells (see Figure 2 and Table 2). We have also performed a more detailed analysis of the simulation trajectories to calculate the distribution of the coordination numbers and determine the amount of Tf2 N that coordinates bidentate (through oxygen atoms on opposite sulfer atoms). In the pure ionic liquid, we find coordination numbers larger than five due to some of the


Page 9 of 42

150

15 1500

100

U - O(H2O)

1000

10

500

50

5

0 0.2 0.25 0.3

0

0 U - O(Tf2N)

20

10

0

0 1500

100

20

Pu - O(H2O)

1000

NI (r)

10

g(r)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10

500

50

5

0 0.2 0.25 0.3

0

0 Pu - O(Tf2N)

20 10 0 0.2

20 10

0.4

0.6

0.2

0.4

0.6

0 0.8

r [nm]

Figure 2: Radial distribution functions (left) and number integrals (right) for uranium (top two rows) and plutonium (bottom two rows) with oxygen in IL/water mixtures. Insets show the height of the first AnO(H2 O) peak. 1.0 mole fraction IL (black), 0.95 (red), 0.90 (green), 0.85 (blue), 0.81 (magenta), 0.77 (orange), 0.74 (violet), and 0.71 (cyan).

actinyl cations coordinating with six Tf2 N anions and a small fraction of Tf2 N anions coordinating bidentate (see Table 2), and even a small amount of uranyl coordinated with six Tf2 N anions, one of which is bidentate, resulting in a coordination number of seven (see Figure 3). The coordination number is larger for uranyl (5.61) than plutonyl (5.21) due to a larger fraction coordinated with six Tf2 N anions and/or with bidentate coordinated Tf2 N; i.e., 42% of the uranyl cations have a coordination number of six, compared to only 22% of the plutonyl cations. Uranyl has a slightly longer An=O bond length than plutonyl and a more flexible O=An=O angle (see Table S3), which makes it easier for the somewhat bulky Tf2 N anion to fit in the area surrounding the actinide. We also find that the uranyl cations are more bent in the pure IL than in mixtures with water, with an average O=U=O angle of 166.2 degrees in pure [EMIM][Tf2 N], rather than 171.2 degrees in the 0.71 mole fraction IL mixture. The same is true to a lesser extent for the plutonyl cations, with angles of 173.6 and 174.5 degrees in the pure IL and 0.71 mole fraction IL mixture, respectively. We should note here that the intramolecular bond stretching and angle bending parameters were determined from gas phase quantum mechanical calculations at the B3LYP level of theory. 41 9 ACS Paragon Plus Environment


1.2

0

2

4

6

0

2

4

6

xIL = 1.0

0.8 0.4

0.4 2+

0.8

0.0 0.95 (28 H2O)

0.8

0.8 0.4

0.0

0.0 0.90 (55 H2O)

0.8 0.4

0.8 0.4

0.0

0.0 0.85 (83 H2O)

0.8 0.4

0.8 0.4

0.0

0.0 0.71 (193 H2O)

0.8 0.4 0.0

8 1.2

0.4

0.0

Fraction AnO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 42

0

0.8 0.4

2

4

6

0

# H2O

2

4

6

0.0 8

# Tf2N

Figure 3: Distribution of the coordination numbers of water (left) and Tf2 N (right) oxygen atoms with 2+ UO2+ 2 (red) and PuO2 (blue) in the first solvation shell. The distributions at intermediate mole fractions are identical to those at 0.85 and 0.71 mole fraction IL and are thus not shown. The error bars indicate the standard deviation.

Prior simulation results for uranyl in [BMIM][Tf2 N] using different force fields find an even higher anion coordination number of 6.3. These simulations also found that the position of the first peak in the An-O(Tf2 N) RDF was shifted to a somewhat longer distance of 0.26 nm 32 as compared to the present results. The higher coordination number and difference in peak position is likely due to their use of an IL force field with unit charges for the ions, rather than the scaled charges of ±0.8 used in this work. Unit charges should lead to significantly stronger interactions between the actinyl cation and the Tf2 N anion. In addition, differences in the intramolecular bond stretching and angle bending terms used in the two studies may lead to differences in coordination tendencies. Recent simulations using the uranyl force field of Gaillard et al. 32 find 5.7 Tf2 N oxygen atoms coordinated to uranyl in [BMIM][Tf2 N], with the position of the first peak at 0.254 nm and a slightly larger coordination number of 6.1 in [MeBu3 N][Tf2 N] due to a significant fraction of bidentate Tf2 N. 35 These results are generally consistent with the results found in the present study. The coordination environment of the actinyl cations changes with the addition of water. For the mixture having an IL mole fraction of 0.95, (where the water to actinyl ratio is 2.33), we observe that all of the water coordinates with an actinyl cation. The first solvation shell is filled in with Tf2 N anions to reach a total oxygen atom coordination number of at least five. A small fraction 10 ACS Paragon Plus Environment

0W4A 0W5A 0W6A 0W7A 1W3A 1W4A 1W5A 1W6A 2W2A 2W3A 2W4A 2W5A 3W1A 3W2A 3W3A 3W4A 4W0A 4W1A 4W2A 5W0A 5W1A 6W0A

1.2 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0

xIL = 0.95 (28 H2O)

xIL = 0.90 (55 H2O)

xIL = 0.85 (83 H2O)

0W4A 0W5A 0W6A 0W7A 1W3A 1W4A 1W5A 1W6A 2W2A 2W3A 2W4A 2W5A 3W1A 3W2A 3W3A 3W4A 4W0A 4W1A 4W2A 5W0A 5W1A 6W0A

Fraction AnO2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


2+

Page 11 of 42

Figure 4: Distribution of the overall composition of the coordination environment of water and Tf2 N with 2+ UO2+ 2 (red) and PuO2 (blue). The x-axis indicates the composition of the first solvation shell, where ‘W’ stands for water and ‘A’ stands for the ionic liquid anion (Tf2 N). For example, “1W4A” indicates a first solvation shell comprised of 1 water molecule and 4 Tf2 N anions. Only the mixtures with the smallest amount of water are shown; the distributions in the mixtures with a higher water content are identical to the distribution at 0.85 mole fraction IL. The error bars indicate the standard deviation.

of the actinyl cations have a coordination number of six due to the presence of an additional Tf2 N anion or Tf2 N anion coordinating bidentate. As the mole fraction of IL is decreased to 0.90 (4.58 water:actinyl ratio), we again find all of the water in the first actinyl solvation shell, with on average less than one Tf2 N anion remaining in the first solvation shell. For these mixtures with less than five water molecules per actinyl cation, the water is very strongly bound to the actinyl cations, resulting in a broad distribution of coordination numbers (see Figure 3). For the 0.95 mole fraction IL mixtures, we find actinyl cations with 1 − 5 water molecules and 1 − 6 Tf2 N oxygen atoms in the first solvation shell. This distribution narrows somewhat for the 0.90 mole fraction IL mixture, although the coordination environment still varies considerably. Though we see a broad range of compositions of the inner coordination sphere, we find a strong preference for the first solvation shell of the actinyl cation to contain a total of five oxygen atoms, though six are possible at very low concentrations of water when Tf2 N bidentate coordination is more likely. Figure 4 shows the actual probabilities of different first solvation shell compositions for different water concentrations, while Figure 5 shows representative snapshots of the water and ions that coordinate with UO2+ 2 . Representative snapshots with the Tf2 N anions removed for clarity can be found in Figure S3 of the Supporting Information.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

0 H2O

1 H2O

2 H2O

3 H2O

Page 12 of 42

4 H2O

5 H2O

Figure 5: Simulation snapshots showing the first solvation shell of uranyl for coordination environments with 0–5 water molecules. Uranium (orange), oxygen (red), hydrogen (white), nitrogen (blue), sulfur (yellow), and CF3 (cyan).

For mixtures with at least a 5 : 1 ratio of water to actinyl (0.85 mole fraction IL and below), the actinyl ion shows a strong preference for a first coordination sphere consisting of five water molecules without any Tf2 N, as the results in Figures 3 and 4 as well as the data in Table 2 suggest. It has been well established through experiment 66–72 and theory 20,22,56,57,60–64,73,74 that the coordination number of uranyl in aqueous solution is five, though a small fraction of the actinyl cations may be coordinated with 4 or 6 water molecules. 24,69–71,75 Additionally, a smaller number of theoretical calculations indicate the coordination number of plutonyl in aqueous solution is also five. 20,22,24 Experiments show that Tf2 N does not bind to uranyl in aqueous solution 76 or in the ILs 1hexyl3methylimidazolium bis(trifluoromethylsulfonyl)imide ([HMIM][Tf2 N]) and [EMIM][Tf2 N] when water is present in at least a 6 : 1 ratio. 77,78 Our simulations indicate that the same is true for any mixture with enough water to fully solvate all the actinyl cations with five water molecules. As the IL content decreases from 0.85 to 0.71 mole fraction, water molecules begin to populate the second coordination shell, though they do not entirely displace the Tf2 N anions. Rather, a significant fraction (up to 41%) of the water molecules are uncoordinated with an actinyl cation and instead are dispersed in the ionic liquid (see Table 3). Using distance criteria of 0.22 nm for the H(H2 O)-O(H2 O) first solvation shell radius and 0.26 nm for the H(H2 O)-O(Tf2 N) first solvation shell radius (see RDFs in Figure S4), we have calculated the fraction of water uncoordinated to an actinyl cation that is coordinated to either another water molecule or a Tf2 N anion. We find that most of the water (85 − 92%) is coordinated with the Tf2 N anion, although a small amount


Page 13 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(1 − 9%) is coordinated with another water molecule. This is consistent with prior experimental 79 and simulation results. 16,80,81 We see no evidence for strong hydrogen bonds between water and the axial oxygen atoms in uranyl and plutonyl, as the O(An)-H(H2 O) RDF is negligible up to about 0.25 nm and the first peak does not occur until 0.32 nm (see Figure S5). This is in agreement with previous simulations of actinyl cations in aqueous solution 25,40,64,82 though some have found very weak short-lived hydrogen bonds 61,62,65



Table 2: The average coordination numbers in the first and second solvation shells for the oxygen atoms in water and Tf2 N coordinated with the actinyl cation, the fraction of bidentate Tf2 N in the first solvation shell, and the number of Tf2 N nitrogen atoms (roughly equivalent to the center-of-mass) up to 0.80 nm.a

first solvation shell UO2+ 2

14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Page 14 of 42

fraction bidentate

PuO2+ 2

UO2+ 2

PuO2+ 2

second solvation shell UO2+ 2

PuO2+ 2

N atoms UO2+ 2

PuO2+ 2

xIL

water

Tf2 N

water

Tf2 N

Tf2 N

Tf2 N

water

Tf2 N

water

Tf2 N

Tf2 N

Tf2 N

1.00

–

5.612

–

5.215

0.0907

0.0634

–

10.213

–

10.769

6.21

6.22

0.95

2.3331

2.832

2.3323

2.731

0.0357

0.0267

0.0096

12.433

0.0137

12.624

6.696

6.747

0.90

4.563

0.453

4.562

0.442

0.0137

0.0101

0.021

14.384

0.031

14.424

7.314

7.324

0.85

5.0031

0.0091

4.9952

0.0083

0.0055

0.0055

0.951

13.871

0.922

13.922

7.251

7.252

0.81

5.0091

0.0071

5.0011

0.0021

0.0

0.0

2.069

12.837

2.016

12.898

6.944

6.961

0.77

5.0131

0.0061

5.0001

0.0031

0.0

0.0

2.999

12.007

2.901

12.081

6.692

6.451

0.74

5.0161

0.0051

5.0011

0.0021

0.0

0.0

3.761

11.291

3.804

11.305

6.461

6.451

0.71

5.0191

0.0051

5.0011

0.0021

0.0

0.0

4.554

10.685

4.475

10.735

6.253

6.262

a

Subscripts indicate uncertainties in the final digit.

ACS Paragon Plus Environment

Page 15 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 3: The fraction of the total water in the system uncoordinated to an actinyl cation and the fraction of water uncoordinated to an actinyl cation that is coordinated to either a Tf2 N anion or to another water molecule. Note that a water molecule can be coordinated to both a Tf2 N anion and another water molecule, or to neither; hence, the rows do not necessarily sum to one.a

uncoordinated with AnO2+ 2

coordinated with Tf2 N coordinated with water

xIL

UO2+ 2

PuO2+ 2

UO2+ 2

PuO2+ 2

UO2+ 2

PuO2+ 2

0.95

0.000

0.000

–

–

–

–

0.90

0.0023

0.0022

0.9178

0.9259

0.0

0.0

0.85

0.1431

0.1492

0.9042

0.9064

0.0161

0.0141

0.81

0.2329

0.2396

0.8962

0.8931

0.0341

0.0351

0.77

0.3088

0.3151

0.8802

0.8801

0.0531

0.0581

0.74

0.3651

0.3633

0.8651

0.8661

0.0731

0.0731

0.71

0.4092

0.4143

0.8511

0.8502

0.0911

0.0941

a


The distribution of coordination environments in the second solvation shell is much broader than in the first solvation shell, as can be seen in Figure S6. Representative snapshots from the simulations showing the location of the water molecules in the first and second solvation shells can be found in Figure S7. Molecules in the second solvation shell are only weakly coordinated to the actinyl cation and thus the boundary between the second solvation shell and the bulk solution is less well-defined than the boundary between the first and second solvation shells. As shown below, this results in molecules being able to move more freely between the second solvation shell and the bulk liquid. For mixtures with low water content (1.0 − 0.90 mole fraction IL), the second solvation shell is made up almost entirely of Tf2 N anions.

3.2 Actinyl Dimeric Complex Formation To determine whether any actinyl(VI) aggregation occurs, we calculated the An-An RDFs, 2+ as seen in Figure 6. For both UO2+ 2 and PuO2 , we observe a large, somewhat broad peak centered

at approximately 0.8 nm in the pure IL mixture, indicating that there is clearly some aggregation of the actinyl cations taking place. As the amount of water in the system increases, this peak becomes smaller and shifts to longer distances, eventually almost completely flattening out. Due to the dilute nature of these mixtures (12 actinyl cations), there is a significant amount of uncertainty in the degree of aggregation, as can be seen in Figure S8. Nevertheless, the simulations suggest that clustering of actinyl cations occurs; in pure [EMIM][Tf2 N] an average 2.5 ± 1 of the 12 actinyl cations participate in what we refer to as “dimeric complexes.” These complexes are created by 15 ACS Paragon Plus Environment


Tf2 N anion(s) forming a bridge between two cations. Potential of mean force calculations (reported in the next section) are consistent with this interpretation. As the water content increases and the water displaces Tf2 N anions from the first solvation shell, there is a rapid reduction in the number of actinyl dimeric complexes, from approximately 2.5 in pure [EMIM][Tf2N] to 1.5 ± 1 after the first addition of water (0.95 mole fraction IL).

15 U-U 10 5

g(r)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 42

0 Pu - Pu 10 5 0 0.4

0.8

1.2

1.6

2

r [nm] Figure 6: U-U (top) and Pu-Pu (bottom) radial distribution functions. For clarity, only select mole fractions are shown. 1.0 mole fraction IL (black), 0.95 (red), 0.90 (green), 0.85 (blue), and 0.71 (cyan). The simulation snapshots show uranyl and plutonyl dimers in pure [EMIM][Tf2 N].

This suggest that the appropriate ligand can serve to bridge two actinyl cations, leading to the formation of dimeric complexes in solution. We observe that uranyl(VI) and plutonyl(VI) dimers form in the pure IL via two bridging Tf2 N ligands, with the oxygen atoms on opposite sulfur atoms coordinating to the different actinide cations. This results in an approximately parallel arrangement of the actinyl cations, though there is not a strong angular preference. The formation of uranyl dimeric complexes in solution has been observed previously in potential of mean force − simulations of UO2+ 2 with perrhenate (ReO4 ) in [BMIM][Tf2 N] and mixtures of [BMIM][Tf2 N]

with acetonitrile. 35 In this case, the association occurs through two bridging ReO4 ligands, with the uranyl cations more or less parallel to each other, much like the arrangement found here. The uranyl perrhenate dimeric complex in [BMIM][Tf2 N] was not observed in UV-vis and EXAFS experiments, although the solutions used in the experiments had a very dilute concentration of uranyl (0.01 − 0.001 M). 35 Experiments have shown that perrhenate and oxalate ligands can bridge 16 ACS Paragon Plus Environment

Page 17 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to form uranyl dimeric complexes under the correct conditions. 83–86 Carboxylate and carbonate ligands can also cause dimerization. 87–89 The well-studied hydrolysis of uranyl shows dimers and trimers with bridging OH ligands exist in solution along with monomers. 90–98 The separation between the uranium atoms in these dimers is generally 0.38−0.49 nm, 85,88,89,91,92,94,95,98 which is considerably shorter than the distance observed here, indicating that Tf2 N is a much weaker coordinating ligand. The exception is perrhenate-bridged dimers, which have U-U distances of 0.66 − 0.69 nm. However, the intramolecular oxygen-oxygen distances in oxalate, carboxylate, and carbonate ligands are much shorter (≈ 0.19−0.22 nm) than those found in perrhenate (≈ 0.28 nm) and Tf2 N (0.22 − 0.52 nm, see Figure S9), and in the case of hydroxo-bridged dimers, the uranium atoms actually coordinate to the same oxygen, resulting in very small U-U separations. Recent work on uranium in ionic liquids has yielded some crystal structures of uranyl dimers where one of the ionic liquid components is able to bridge between the metal centers, with U-U distances of 0.58 and 0.56 nm. 99,100 However, in these cases the ionic liquid incorporates a carboxylic acid group, with relatively short intramolecular oxygen-oxygen distances (≈ 0.22 nm), which forms O=C=O bridges. Crystal structures for uranyl with other, somewhat more unconventional and elongated ligands, such as bis(diphenylphosphino)methane dixoide, 101 , carbamoyl methyl sulfoxide, 102 1,4-di(butylsulfinyl)butane, 103 and citric acid and tricarballylatic acid (under certain conditions) 104 show the formation of dimers with larger U-U separations of 0.65 − 1.35 nm. 101–104 The carbamoyl methyl sulfoxide ligand is most structurally similar to Tf2 N and results in a similar U-U distance of 0.79 nm. 102 As far as we are aware, there is no UO2 (Tf2 N)2 crystal structure or solution-phase measurement available which could confirm the U-U distance of approximately 0.8 nm found in the present work. However, crystal structures have been determined for a number of lanthanide-Tf2 N complexes. Tf2 N generally coordinates lanthanides in a bidentate manner, with the oxygen atoms on opposite sulfur atoms coordinating the metal 105–108 although in some cases one of the Tf2 N anions coordinates monodentate. 106,107,109,110 Given the lack of axial oxygens for lanthanide cations, it is not surprising that a larger amount of bidentate Tf2 N coordination is found due to the larger available area around the metal cation. Many other experimental and simulation studies have found evidence of Tf2 N anions facilitating dimeric complex formation of metals, including europium, iron, titanium, solicon, ruthenium, barium and lithium 107,110–120 . Though previous simulations of uranyl ions in ILs found that the uranyl cation was surrounded by a layer of anions followed by a layer of cations, 27,29,30,33 we do not see a strong layering effect. Tf2 N anions are indeed found in the first and second solvation shells, but the actinide-EMIM RDF (calculated with respect to the nitrogen atoms in the imidazolium ring) exhibits only a very small, broad first peak at relatively long distances from the actinyl cation, indicating that the EMIM cations are not very structured around the actinyl cations (see Figure S2). The Tf2 N-Tf2 N, EMIMTf2 N, and EMIM-EMIM RDFs reveal that the structure of the bulk ionic liquid does not change 17 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 42

significantly with increasing water content (see Figure S10). Most of the added water coordinates with the actinyl cations, leaving at most 41% of the water dispersed throughout the bulk ionic liquid, for an effective mole fraction of water in [EMIM][Tf2 N] of approximately 0.17 (79 water molecules/466 ion pairs). Prior studies of water/IL mixtures indicate that although large amounts of water can have a significant effect on the structure of the IL, the relatively small amounts of water used in the present study have a neglible effect on the IL structure. 39,81,121,122

3.3 Potentials of Mean Force To obtain a more detailed picture of the association tendencies of various species, umbrella sampling was used to compute potentials of mean force (PMFs) for systems consisting of 1 UO2 (Tf2 N)2 in 80 [EMIM][Tf2 N] ion pairs with the progressive addition of 1 − 6 water molecules 2+ UO2+ 2 · xH2 O + H2 O → UO2 · (x + 1)H2 O

(1)

We also computed PMFs for the addition of a Tf2 N anion to UO2+ 2 · 5 H2 O 2+ UO2+ 2 · 5H2 O + Tf2 N → UO2 · (xH2 O + Tf2 N) + yH2 O

(2)

where x is three or four and y is one or two, depending on whether the Tf2 N coordinates mono or bidentate, as well as for the addition of a one water molecule to a fully hydrated PuO2+ 2 ion 2+ PuO2+ 2 · 5H2 O + H2 O → PuO2 · 6H2 O

(3)

Figure 7 shows the PMFs, listed as G(r), for the addition of water to UO2+ 2 . The deep potential minimum at around 0.25 nm indicates that water binds strongly to the uranyl(VI) cation at this distance, consistent with our earlier MD results. The free energy difference between the associated and dissociated states ranges from −52.3 kJ/mol for the addition of the first water molecule to −37.6 kJ/mol for the addition of the fifth water. Numerical values for the association free energies along with the location of the minima are listed in Table 4. The PMFs clearly show why conformation sampling with conventional MD is difficult; once a water molecule associates with the uranyl cation, the free energy barrier for leaving the first solvation shell is quite large. Figure 7 also shows the free energy profile for the addition of a sixth water molecule to UO2+ 2 · 5 H2 O and PO2 · 5 H2 O. Addition of a sixth water molecule to uranyl is energetically unfavorable, with ∆G = 12.1 kJ/mol and a barrier of 18.4 kJ/mol between the first and second solvation shells. We note that these free energies are in reasonable agreement with prior calculations for the associative and dissociative pathways of water exchange in aqueous solution, 56,58,60,63,123,124 though the surrounding solvent in


Page 19 of 42

40 Tf2N 6 H2 O

20

G(r) [kJ/mol]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0

-20 5 H2 O

-40

1 H2 O -60 0.2

0.4

0.6

0.8

1

r [nm] Figure 7: Potentials of mean force for UO2+ 2 · x H2 O + H2 O for x = 0 (black), x = 1 (red), x = 2 (green), 2+ x = 3 (blue), x = 4 (magenta), x = 5 (orange), PuO2+ 2 · 5 H2 O + H2 O (dashed orange), and UO2 · 5 H2 O + Tf2 N (violet). The simulation snapshots show UO2+ 2 · 5 H2 O + H2 O with the sixth water in the first and second solvation shells.

our case is [EMIM][Tf2 N], not water. Adding a sixth water molecule to the first solvation shell of plutonyl is even more unfavorable, with ∆G = 30.7 kJ/mol and a barrier of 36.6 kJ/mol between the first and second solvation shells. To accomodate the sixth water molecule, the O=An=O angle distorts from an average of 171.6 degrees for five-coordinate uranyl or 174.5 degrees for plutonyl to 155.3 degrees for six-coordinate uranyl and plutonyl. It appears that the higher free energy barrier associated with six-coordinate plutonyl is due to the fact that the O=Pu=O angle has a larger force constant than the O=U=O angle, 41 thus requiring more energy to become distorted than the uranyl ion. We should note that the distortion in the six-coordinate actinyl complexes may be an artifact of the parameterization of the force field, though others have found that structures optimized using density functional theory for six-coordinate uranyl are significantly distored, with four water molecules at distances of approximately 0.25 nm, two at longer distances of 0.265 nm, and a O=U=O angle of 171 degrees. 56 Though this structure is less distorted than ours, we should recall that we have forced all six water molecules to be at a distance of 0.25 nm from the uranyl cation, which could cause the O=U=O angle to change. Figure 7 shows that a shallow potential minimum at around 0.44 nm appears after the addition of the first water molecule. This corresponds to hydrogen bonding between the approaching water molecule and water molecules already in the first solvation shell of the actinyl ion. This secondary



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 42

Table 4: Relative free energies (in kJ/mol) and position of the minima (in nm) for the progressive addition of water molecules or a Tf2 N anion to the first and second solvation shells of an actinyl(VI) cation in [EMIM][Tf2 N].a

first solvation shell

second solvation shell

position [nm]

∆G [kJ/mol]

position [nm]

∆G [kJ/mol]

UO2+ 2 + H2 O

0.2491

−52.31

–

–

UO2+ 2 · H2 O + H2 O UO2+ 2 · 2 H2 O + H2 O UO2+ 2 · 3 H2 O + H2 O UO2+ 2 · 4 H2 O + H2 O UO2+ 2 · 5 H2 O + H2 O PuO2+ 2 · 5H2 O + H2 O 2+ UO2 · 5 H2 O + Tf2 N

0.2491

−49.26

0.4371

−3.37

0.2511

−47.43

0.4331

−6.73

0.2511

−44.16

0.4413

−7.93

0.2531

−37.63

0.4392

−8.91

0.2551

12.11

0.4452

−9.42

0.2571

30.73

0.4432

−9.41

0.6016

−4.11

reaction

a

—

–


well becomes deeper with the addition of more water, ranging from −3.3 kJ/mol for the second water molecule to approximately −9 kJ/mol for the fifth and sixth water molecules. We also calculated the PMF for the addition of a Tf2 N anion to UO2+ 2 · 5 H2 O. In this case, the Tf2 N anion can displace one or two water molecules from the first solvation shell of the uranyl 2+ cation, resulting in UO2+ 2 · (4 H2 O + Tf2 N) or UO2 ·(3 H2 O) + Tf2 N), depending on whether Tf2 N

coordinates in a monodentate or bidentate manner. The addition of Tf2 N into the second solvation shell is somewhat favorable, as seen from the broad, relatively shallow potential well of −4 kJ/mol at from approximately 0.55 − 0.7 nm. Note that the distance constrained in the PMF calculation is between the centers of mass of the two ions, which for Tf2 N is significantly different from the positions of the oxygen atoms that actually associate with UO2+ 2 . This accounts for the shifting of the potential well to longer distances. As the Tf2 N anion begins to approach the first solvation shell, G(r) becomes positive. At distances of approximately 0.46 − 0.36 nm, the Tf2 N anion coordinates monodentate to the uranyl cation and displaces one water molecule to the second solvation shell, with a free energy increase of approximately 9 − 53 kJ/mol. At distances of less than 0.36 nm, the Tf2 N coordination changes from monodentate to bidenate, and two water molecules are displaced. This is even more unfavorable, with free energy changes ranging from 53 − 193 kJ/mol. This is consistent with our finding that even small amounts of water displace Tf2 N anions from the first solvation shell of actinyl cations. Previous PMF simulations of uranyl in [BMIM][Tf2 N] find that the addition of chloride anions results in large, negative free energy changes on the order of −103 to −162 kJ/mol 33 , 20 ACS Paragon Plus Environment

Page 21 of 42

30 8 H2 O

15 0 -15

G(r) [kJ/mol]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


4 H2 O

15 0 -15

0 H2 O

15 0 -15 0.5

0.7

0.9

1.1

r [nm]

2+ Figure 8: Potentials of mean force for AnO2+ 2 –AnO2 in pure ionic liquid and for systems with 4 and 8 2+ 2+ 2+ 2+ water molecules. UO2 –UO2 (red) PuO2 –PuO2 (blue) in pure [EMIM][Tf2 N] (bottom), with 4 water molecules (middle), and with 8 water molecules (top). The error bars indicate the standard deviation calculated from the average of the two independent simulations. The simulation snapshots show associated uranyl complexes.

consistent with the observation that chloride is a much stronger ligand for uranyl than Tf2 N or water. 29,32,34,67,125–127 The addition of perrhenate (ReO− 4 ) to the same system results in slightly smaller free energy changes of −23 to −34 kJ/mol. 35 To further examine the actinyl dimer complexes, first seen in the An-An RDFs (see Figure 6), we calculated the potential of mean force between two uranyl(VI) and two plutonyl(VI) cations (U-U and Pu-Pu) in pure [EMIM][Tf2 N] and in [EMIM][Tf2 N] with the addition of 4 or 8 water molecules (similar water to actinyl ratios as in the 0.95 and 0.90 mole fraction IL simulations). The results are shown in Figure 8. In the pure IL association takes place from 0.66 to 0.94 nm, with a minimum for uranyl-uranyl interactions of approximately −7 ± 1 kJ/mol and for plutonyl-plutonyl interactions of −9 ± 2 kJ/mol. The relatively broad range of distances over which the relative free energy is negative is due to the flexibility of the Tf2 N anion, which results in a wide range of intramolecular oxygen-oxygen distances (≈ 0.27 − 0.52 nm, see Figure S9). For both uranyl and plutonyl, a small barrier of approximately +3 kJ/mol at 1.0 nm separates the associated and dissociated states.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 42

These results are qualitatively similar to the findings of Chaumont and Wipff 35 , in which uranyl–uranyl dimerization was observed to be promoted by perrhenate anions dissolved in [BMIM][Tf2 N]. There are key quantitative differences between the two cases, however. In the perrhenate simulations, a much larger energy barrier between the associated and dissociated states of approximately 33.5 kJ/mol was observed at 0.83 nm. The minimum potential of −15.5 kJ/mol was observed at 0.65 nm. The position of both the barrier and the minimum occur at shorter distances than we found in [EMIM][Tf2 N], likely due to the more compact structure of the bridging perrhenate ligand as compared to Tf2 N. As water is added to the system, the favorable relative free energy for dimer formation decreases and the minimum of the potential well shifts to longer distances, as can be seen in Figure 8. When enough water is present, the relative binding strength of Tf2 N decreases such that instead of bridging directly between two of the metal centers, the Tf2 N coordinates with one of the metal centers and then hydrogen bonds with a water molecule in the first solvation shell of the other actinyl. Presumably, in mixtures with an even higher water content, the Tf2 N anions would hydrogen bond to water molecules in the first solvation shell of two different actinyl cations, resulting in a very weak association at long distances. The PMF results are all consistent with the changes in the An-An RDFs with increasing water content shown in Figure 6.

3.4 Residence Time Correlation To better understand the coordination of water and Tf2 N with the actinyl(VI) cations, we calculated the following residence time correlation function 128 R(t) =

*

N0 1 X θi (0)θi (t) N0 i=1

+

(4)

where N0 is the number of molecules in the first (or second) solvation shell at time 0 and θi (t) is the Heaviside function, which is 1 if the ith molecule is in the solvation shell at time t and 0 otherwise. Thus if R(t) = 1, the identity of the molecules in the first solvation shell of the actinyl cation are unchanged. As molecules leave the first solvation shell and are replaced, R(t) decays to zero. The residence time of molecules in the first solvation shell, τ , is calculated as Z ∞ R(t)dt τ=

(5)

0

To perform the integration, R(t) over the first 10 ns was fit to two exponential functions which were then integrated to obtain τ . Figure 9 shows that for mixtures with 0.95 and 0.90 mole fraction IL (a water to actinyl ratio of less than five), water in the first solvation shell is completely correlated with the actinyl 22 ACS Paragon Plus Environment

2+

2+

UO2

1.0

PuO2

1.0

H2 O

0.8

R(t)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.8

0.6

0.6

0.4

0.4

0.2

0.2

1.0

1.0

Tf2N

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0

0

5

10

0

5

10

R(t)

Page 23 of 42

0.0 15

time [ns] Figure 9: Residence time correlation for the actinyl cation with water (top) and Tf2 N (bottom) in the first solvation shell for uranyl (left) and plutonyl (right). 1.0 mole fraction IL (black), 0.95 (red), 0.90 (green), 0.85 (blue), 0.81 (magenta), 0.77 (orange), 0.74 (violet), and 0.71 (cyan).

cation over the length of the entire simulation, making estimation of the residence time difficult. To improve the statistics, we performed eight independent simulations for the 0.95 and 0.90 mole fraction IL mixtures, resulting in the broad distributions discussed in Section 3.1. As the water content increases, the residence time correlation function decays more rapidly, indicating increased mobility of the coordinated water molecules (see Figure 9 and Table 5). We find that the residence times for water are longer for the mixtures with plutonyl than with uranyl. Previous simulation results find that the water exchange process for aqueous uranyl proceeds through an associative process with an extremely short-lived six-coordinate intermediate. 56,58–60,64,124 Given that the relative free energy for plutonyl coordinated with six water molecules is considerably higher than for uranyl (by ≈ 18 kJ/mol), it is perhaps unsurprising that the water exchange rate is slower for plutonyl than uranyl. The residence times for the Tf2 N anion are slightly shorter for plutonyl than uranyl. For the mixtures with 0.85 mole fraction of IL and below, there is not a statistically significant amount of Tf2 N in the first solvation shell to determine residence times with any accuracy.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 42

Table 5: Residence times (in ns) for water and Tf2 N around the actinyl calculated from the ion pair correlation function.a

first solvation shell UO2+ 2

second solvation shell

PuO2+ 2

xIL

water

Tf2 N

water

1.00

–

2.76

–

0.95

–

0.72

0.90

–

0.85

UO2+ 2 water

Tf2 N

water

32

–

4.07

–

41

–

0.789

–

1.73

–

1.92

0.192

–

0.216

–

1.238

–

1.21

7.12

–

257

–

0.9510

0.811

0.92

0.864

0.81

3.363

–

92

–

0.9512

0.604

1.46

0.605

0.77

1.921

–

4.96

–

0.783

0.441

1.01

0.491

0.74

1.52

–

3.62

–

0.7116

0.361

0.988

0.351

0.71

1.222

–

3.12

–

0.633

0.291

0.9914

0.341

a

Tf2 N

PuO2+ 2 Tf2 N


The residence time correlation function decays much more rapidly for molecules in the second solvation shell, as shown in Figure S11. This is consistent with the small free energy difference between second solvation shell molecules and those in the bulk (see Figure 7). For the 0.95 and 0.90 mole fraction IL mixtures, there is a neglible amount of water in the second solvation shell and so it is not possible to determine a residence time. As with the first solvation shell, the residence times for water and Tf2 N in the second shell decreases with increasing water content, with little difference between uranyl and plutonyl.

3.5 Self-diffusivity To understand the dynamics of actinyl(VI) cations in [EMIM][Tf2 N] with and without added water, we calculated the self-diffusivity of each of the components in the system. The selfdiffusivity can be calculated from the mean square displacment (MSD) using an Einstein relation: D=

1 lim 6 t→∞

*

N X

|ri (t) − ri (0)|2

i=1

+

(6)

or from the velocity autocorrelation function (VACF) using a Green-Kubo relation: D=

Z

∞ 0

N 1 X hvi (t) · vi (0)idt 3N i=1


(7)

Page 25 of 42

Due to the degree of noise in the long-time tail of the VACF, the Einstein relation is generally a more reliable method for calculating the diffusion coefficient. However, the shape of the VACF at short times can tell us about the type of motion; e.g., whether the molecules (or ions) are “rattling” in cages (indicated by a negative well in the VACF) or whether the motion is more diffusive. 2+

AnO2

20

2

25 20

15

15

10

10

5

5

0

0

H2 O

100

xIL = 0.71

100

75

75

50

50

25 0

0

25

xIL = 0.85

5000

10000

2

25

PuO2

UO2

MSD [nm ]

2+

MSD [nm ]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0

5000

0 10000 15000

time [ps] Figure 10: Center of mass mean square displacement for the actinyl cation (top) and water (bottom) in systems with uranyl (left) and plutonyl (right). Note the difference in scale on the y-axis. 1.0 mole fraction IL (black), 0.95 (red), 0.90 (green), 0.85 (blue), 0.81 (magenta), 0.77 (orange), 0.74 (violet), and 0.71 (cyan).

To calculate an accurate self-diffusion coefficient using Eq. 6, the slope of the MSD must be linear. For this reason, the center of mass MSD from 1 − 7 ns was used to calculate the diffusion coefficients of the actinyl cation and water, and the center of mass MSD from 1−11 ns was used for EMIM and Tf2 N. Although there is a significant amount of noise in the uranyl and plutonyl MSDs at long times due to the small number of actinyl cations in each mixture, the MSD generally increases as water is added to the mixture, as can be seen in Figure 10. The self-diffusion coefficients for uranyl and plutonyl are the same within the uncertainty (see Table 6) and increase by approximately a factor of 2 − 3 as the water content in the system increases from 0 to 0.29 mole fraction. Although there are no experimental self-diffusion data for the actinyl cations in [EMIM][Tf2 N] at 400 K, there is a limited amount of data for uranium(VI) and uranyl chloride (UO2 Cl2− 4 ), in a variety of other ionic liquids. 129–134 For these cases, self-diffusion coefficients range from 6.64 × 10−13 25 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 42

m2 /s to 9.01 × 10−12 m2 /s, depending on the temperature and the IL. In cases where values at multiple temperatures are given, we can use the relationship D = Do exp(−Ea /RT ) to extrapolate to 400 K, resulting in values of 3.37 × 10−12 m2 /s and 8.23 × 10−11 m2 /s for uranyl in [BMIM][Cl]. 129,133 Given the range of experimental data and that the viscosity of [EMIM][Tf2 N] is significantly lower than that of [BMIM][Cl], our values of 8 × 10−11 and 7 × 10−11 m2 /s for uranyl and plutonyl, respectively, in dry [EMIM][Tf2 N] are very reasonable. Self-diffusion coefficients were also computed from velocity autocorrelation functions via Eq. 7, with results given in Table 6. In general, the self-diffusion coefficients calculated from the VACF are consistent with those calculated from the MSD. For mixtures with a very low water content (0.95 and 0.90 mole fraction IL or water to actinyl ratios of less than five), the MSD of the water is highly correlated with that of the actinyl cation, which is consistent with the high degree of correlation between water and the cations discussed in Sections 3.3 and 3.4. As more water is added to the system, the mobility of the water molecules increases by more than an order of magnitude, consistent with the reduced residence times. We note that the self-diffusion coefficient of water remains significantly lower than that of bulk SPC/E water (approximately 10.8 × 10−9 m2 /s at 400 K). The self-diffusivity of the ionic liquid cation is higher than that of the anion (see Figure S12), consistent with previous experiments 135 and simulations 15,136 for imidazolium-based cations. We also find that the diffusivity of the ionic liquid ions increases with the addition of water, which is consistent with previous experimental 135,137 and simulation 16,39,80,81,122 results. As shown in Figure 11, the VACF of the actinyl cations in pure ionic liquid exhibits a small negative well at short times indicative of cage-like “rattling” motion. As water is added to the system, the negative well first shifts to longer times and becomes slightly deeper, then shifts to still longer times and becomes more shallow. This indicates that the motion of the actinyl cations changes as the coordination environment changes from Tf2 N to water in the first solvation shell (at 0.85 mole fraction IL) and as water is added to the second solvation shell. The VACFs of uranyl and plutonyl are nearly identical. Figure 12 shows that the VACF of water decays much faster than the VACF of the ionic liquid, consistent with its larger self-diffusion coefficient. For mixtures with very low water content, the VACF of water exhibits several sharp negative wells separated by sharp peaks. This is due to the highly heterogeneous environment of the water at these low concentrations, where actinyl cations are coordinated with anywhere from one to five water molecules. As water is added to the system, the wells become less deep and the peaks less steep as the coordination environment becomes more homogeneous. At a high enough water content, the first well and peak disappear and instead a small shoulder appears in the VACF at short times followed by a relatively broad and shallow well. The VACFs of the ionic liquid cation and anion show very little change with increasing water content (see Figure S13). The VACF for


Page 27 of 42

1.0 2+

UO2

0.8 0.6 0.4 0.2 0.0

C(t)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


-0.2 2+

PuO2

0.8 0.6 0.4 0.2 0.0 -0.2 0.0

0.5

1.0

1.5

2.0

time [ps] Figure 11: Center of mass velocity autocorrelation function for uranyl (left) and plutonyl (right). For clarity, only select mole fractions are shown. 1.0 mole fraction IL (black), 0.95 (red), 0.90 (green), 0.85 (blue), and 0.71 (cyan).

EMIM contains a negative well at short times, indicating cage-like “rattling” motion. The ionic liquid anion has a much broader and shallower well in the VACF at longer times, indicating that the motion of Tf2 N is more diffusive. These results are consistent with those previously obtained for [BMIM][Tf2 N] using a different ionic liquid force field. 138

4 Conclusions Classical MD simulations of 0.1 M solutions of uranyl(VI) and plutonyl(VI) in [EMIM][Tf2 N]/water mixtures with 1.0 − 0.71 mole fraction [EMIM][Tf2 N] have been performed. The simulations reveal that the actinyl cations strongly prefer a first solvation shell of five oxygen atoms, though at low mole fractions of water the coordination number can be higher due to the presence of six Tf2 N anions or a bidentate-coordinated Tf2 N anion. Higher Tf2 N coordination numbers are found for uranyl than plutonyl, due to the more flexible O=U=O angle. Water coordinates with the actinyl 27 ACS Paragon Plus Environment


Table 6: Diffusion coefficents (10−9 m2 /s) for actinyl, water, EMIM, and Tf2 N calculated from the mean square displacement (MSD) and velocity autocorrelation function (VACF).a

actinyl

28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

Page 28 of 42

water

EMIM

Tf2 N

UO2+ 2

PuO2+ 2

UO2+ 2

PuO2+ 2

UO2+ 2

PuO2+ 2

xIL

MSD VACF

MSD VACF

MSD VACF

MSD VACF

MSD VACF

MSD VACF

MSD VACF MSD VACF

1.00

0.082

0.103

0.071

0.104

–

–

–

–

0.351

0.396

0.352

0.426

0.201

0.235

0.211

0.244

0.95

0.092

0.117

0.082

0.134

0.092

0.21

0.081

0.21

0.381

0.404

0.362

0.386

0.221

0.273

0.211

0.234

0.90

0.092

0.126

0.082

0.137

0.102

0.31

0.092

0.21

0.381

0.397

0.392

0.414

0.221

0.255

0.231

0.273

0.85

0.144

0.162

0.112

0.173

0.495

0.72

0.424

1.01

0.391

0.416

0.401

0.454

0.231

0.242

0.241

0.273

0.81

0.112

0.132

0.141

0.154

0.691

0.91

0.776

0.81

0.411

0.395

0.421

0.411

0.231

0.246

0.251

0.271

0.77

0.153

0.159

0.141

0.259

0.985

1.02

0.944

1.21

0.451

0.411

0.422

0.441

0.271

0.271

0.251

0.322

0.74

0.191

0.307

0.183

0.243

1.134

1.53

1.31

1.32

0.451

0.462

0.441

0.412

0.291

0.295

0.281

0.253

0.71

0.233

0.266

0.171

0.259

1.301

1.41

1.31

1.41

0.471

0.481

0.481

0.535

0.291

0.285

0.311

0.388

a


ACS Paragon Plus Environment

UO2+ 2

PuO2+ 2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.0 2+

UO2

0.8 0.6 0.4 0.2 0.0

C(t)

Page 29 of 42

-0.2 1.0 2+

PuO2

0.8 0.6 0.4 0.2 0.0 -0.2 0.0

0.2

0.4

0.6

time [ps] Figure 12: Center of mass velocity autocorrelation function for water in systems with uranyl (left) and plutonyl (right). Note the difference in scale on the x-axis from Figure 11. 0.95 mole fraction IL (red), 0.90 (green), 0.85 (blue), 0.81 (magenta), 0.77 (orange), 0.74 (violet), and 0.71 (cyan).

cations much more strongly than Tf2 N, with even very small amounts of water displacing the Tf2 N anions from the inner coordination sphere. Below a 5 to 1 water to actinyl ratio, all of the water molecules are found in the first solvation shell of an actinyl cation. Above this ratio, some of the water enters the second solvation shell, though it never completely displaces the Tf2 N anions. In mixtures with a high water content, a significant fraction, up to 41%, of the water molecules are uncoordinated to an actinyl cation and instead dispersed throughout the bulk ionic liquid. Most of these water molecules are coordinated with a Tf2 N anion through weak hydrogen bonds, though a small fraction are coordinated with another water molecule. The potential of mean force for the progressive addition of water molecules to uranyl in [EMIM][Tf2 N] was calculated via umbrella sampling simulations. The resulting PMF shows that the first five water molecules coordinate strongly with the uranyl cation, with ∆G ranging from −52.3 to −37.6 kJ/mol as more water is added to the system. After the addition of the first water molecule, a second smaller potential well is found at longer distances, where the approaching water is able to hydrogen bond with the water molecule(s) already coordinated with the uranyl 29 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 42

cation. The addition of a sixth water molecule is unfavorable, with ∆G = 12.1 kJ/mol, and the uranyl cation distorts significantly to accommodate the extra water molecule. Addition of a sixth water molecule to PuO2 · H2 O5 is even more unfavorable, with ∆G = 30.7 kJ/mol due to the stiffer O=Pu=O angle. Displacement of the water molecules by a Tf2 N anion is highly unfavorable, with values of ∆G over 190 kJ/mol. We find that in the neat ionic liquid, the weakly coordinating Tf2 N anions facilitate the formation of weak actinyl dimer complexes by bridging between the two actinyl ions. As water is added, these complexes become weaker and longerranged as the Tf2 N anion is displaced from the first coordiation shell. Potential of mean force calculations show that these dimer complexes are stable, with relative free energies of up to −9 kJ/mol in pure [EMIM][Tf2 N]. Calculation of the residence time correlation function clearly shows that below a 5 to 1 water to actinyl ratio, the water molecules remain coordinated with the same actinyl cation over the entire course of the simulation. Above this ratio, water molecules are able to exchange between the first and second solvation shells and between the second solvation shell and the bulk. Unsurprisingly, the exchange dynamics are much faster in the second solvation shell. Water molecules in the first solvation shell are more strongly correlated with plutonyl than with uranyl, again due to the stiffer O=Pu=O angle. As the PMF simulations indicate, the actinyl cation needs to distort to allow a water molecule to approach the actinyl cation closely enough for an exchange to take place, consistent with as associative exchange mechanism. We find that the self-diffusion coefficients of all species in the solutions increase with increasing water content. This increase is largest for water, where the self-diffusion coefficient increases by more than an order of magnitude from the lowest to the highest concentration of water. The self-diffusion coefficients of the actinyl cations increase by a factor of 2 − 3, which is significantly less than water. The short time behavior of the velocity autocorrelation function for the actinyl cations and water changes with increasing water content. The negative well in the VACF of the actinyl cations shifts to longer times and becomes shallower, indicating that the motion of the actinyl cations becomes more diffusive at higher water concentrations. The VACF of water exhibits unusual behavior for mixtures with a low water content, with several peaks and wells at short times. In these mixtures the water molecules experience a very heterogeneous environment, as the water coordination numbers of the actinyl cations range from 1 − 5. As the water content increases, the local environment becomes more homogeneous, and the VACF becomes smoother. In conclusion, large-scale classical molecular dynamics simulations are a valuable tool in understanding the coordination and dynamics of actinyl cations in solution, providing molecular-level insight into these complex systems without the need for difficult and/or dangerous experiments. In the future, more advanced models incorporating polarizability could be explored for increased accuracy in these highly charged environments. Understanding the behavior of water over a range 30 ACS Paragon Plus Environment

Page 31 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of concentrations in ionic liquid systems with actinyl cations is vital to designing improved electrochemical separation and liquid-liquid solvent extraction processes for use in advanced nuclear fuel cycles. Precisely controlling the amount of water in the system and examining the behavior of individual water molecules is much more straightforward in a simulation than an experiment. The methodology employed in this work can be applied to other systems to study the behavior of actinyl cations in ionic liquids with more strongly coordinating ligands.

Acknowledgements The authors gratefully acknowledge the Los Alamos Laboratory Directed Research and Development Program for financial support during this project. The computer resources were provided by the Center for Research Computing at the University of Notre Dame.

Supporting Information Contents The details of the simulations, along with a figure showing the MSD for uranyl at 400 and 298 K and a table with the average temperature and pressure over the course of the 15 ns N V E production run are available in the Supporting Information. Also available are a table with the position of the first maximum of the An-O(H2 O) and An-O(Tf2 N) radial distribution functions, a figure showing U-O(Tf2 N), U-N(Tf2 N), and U-N(EMIM) RDFs, a table with the average bond lengths and angles for the actinyl cations, and a figure showing simulation snapshots of the water molecules in the first solvation shell for uranyl cations with different coordination environments. Additionally, figures showing the H(H2 O)-O(Tf2 N) and H(H2 O)-O(H2 O) RDFs and the axial oxygen-water hydrogen RDFs and NIs, simulation snapshots showing the location of the water molecules in the first and second solvation shell of uranyl cations, and a figure showing the actinide-actinide RDFs for the different independent simulations, a figure showing the intramolecular oxygen-oxygen distances in Tf2 N, and a figure showing Tf2 N-Tf2 N, Tf2 N-EMIM, and EMIM-EMIM RDFs are included. Finally, figures showing the MSD and VACF of the ionic liquid cation and anion can be found in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 42

References [1] Wick, O. J. Plutonium Handbook: A Guide to the Technology; American Nuclear Society: La Grange Park, IL, 1981; Vol. 1-2. [2] Lanham, W. B.; Runion, T. C. PUREX Process for Plutonium and Uranium Recovery; ORNL-479: Oak Ridge National Laboratory, Oak Ridge, TN, 1949. [3] Forsyth, S. A.; Pringle, J. M.; MacFarlane, D. R. Ionic Liquids: An Overview. Aust. J. Chem. 2004, 57, 113–119. [4] Binnemans, K. Lanthanides and Actinides in Ionic Liquids. Chem. Rev. 2007, 107, 2592– 2614. [5] Sun, X.; Luo, H.; Dai, S. Ionic Liquids-Based Extraction: A Promising Strategy for the Advanced Nuclear Fuel Cycle. Chem. Rev. 2012, 112, 2100–2128. [6] Cocalia, V. A.; Gutowski, K. E.; Rogers, R. D. The Coordination Chemistry of Actinides in Ionic Liquids: A Review of Experiment and Simulation. Coordin. Chem. Rev. 2006, 250, 755–764. [7] Venkatesan, K. A.; Srinivasan, T. G.; Rao, P. R. V. A Review on the Electrochemical Applications of Room Temperature Ionic Liquids in Nuclear Fuel Cycle. J. Nuc. Radiochem. Sci. 2009, 10, R1–R6. [8] Takao, K.; Bell, T. J.; Ikeda, Y. Actinide Chemistry in Ionic Liquids. Inorg. Chem. 2013, 52, 3459–3472. [9] Rao, P. R. V.; Venkatesan, K. A.; Rout, A.; Srinivasan, T. G.; Nagarajan, K. Potential Applications of Room Temperature Ionic Liquids for Fission Products and Actinide Separation. Separ. Sci. Technol. 2012, 47, 204–222. [10] Venkatesan, K. A.; Rao, C. J.; Nagarajan, K.; Rao, V. P. R. Electrochemical Behavior of Actinides and Fission Products in Room Temperature Ionic Liquids. Int. J. Electrochem. 2012, 841456. [11] Visser, A. E.; Rogers, R. D. Room-Temperature Ionic Liquids: New Solvents for f-Element Separations and Associated Solution Chemistry. J. Solid State Chem. 2003, 171, 109–113. [12] Lopes, J. N. A. C.; Padua, A. A. H. Nanostructural Organization in Ionic Liquids. J. Phys. Chem. B 2006, 110, 3330–3335. [13] Chiappe, C. Nanostructural Organization of Ionic Liquids: Theoretical and Experimental Evidences of the Presence of Well Defined Local Structure in Ionic Liquids. Monotsh, Chem. 2007, 138, 1035–1043. [14] Castner, E. W.; Margulis, C. J.; Maroncelli, M.; Wishart, J. F. Ionic Liquids: Structure and Photochemical Reactions. Annu. Rev. Phys. Chem. 2011, 62, 85–105. [15] Maginn, E. J. Molecular Simulation of Ionic Liquilds: Current Status and Future Opportunities. J. Phys. Condens. Matter 2009, 21, 373101. [16] Hanke, C. G.; Lynden-Bell, R. M. A Simulation Study of Water-Dialkylimidazolium Ionic Liquid Mixtures. J. Phys. Chem. B 2003, 107, 10873–10878.


Page 33 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[17] Sieffert, N.; Wipff, G. The [BMI][Tf2N] Ionic Liquid/Water Binary System: A Molecular Dynamics Study of Phase Separaton and of the Liquid-Liquid Interface. J. Phys. Chem. B 2006, 110, 13076–13085. [18] Schreckenbach, G.; Shamov, G. A. Theoretical Actinide Molecular Science. Acc. Chem. Res. 2010, 43, 19–29. [19] Bühl, M.; Wipff, G. Insights into Uranyl Chemistry from Molecular Dynamics Simulations. Chem. Phys. Chem. 2011, 12, 3095–3105. [20] Spencer, S.; Gagliardi, L.; Handy, N. C.; Ioannou, A. G.; Skylaris, C.-K.; Willetts, A. Hy2+ dration of UO2+ 2 and PuO2 . J. Phys. Chem. A 1999, 103, 1831–1837. [21] Hay, P. J.; Martin, R. L.; Schreckenbach, G. Theoretical Studies of the Properties and So+ lution Chemistry of AnO2+ 2 and AnO2 Aquo Complexes for An = U, Np, and Pu. J. Phys. Chem. A 2000, 104, 6259–6270. [22] Tsushima, S.; Suzuki, A. Hydration Numbers of Pentavalent and Hexavalent Uranyl, Neptunyl, and Plutonyl. J. Mol. Struct.: THEOCHEM 2000, 529, 21–25. [23] Shamov, G. A.; Schreckenbach, G. Density Functional Studies of Actinyl Aquo Complexes Studied Using Small-Core Effective Core Potentials and a Scalar Four-Component Relativistic Method. J. Phys. Chem. A 2005, 109, 10961–10974. + [24] Cao, Z.; Balasubramanian, K. Theoretical Studies of UO2 (H2 O)2+ n , NpO2 (H2 O)n , and PuO+2(H2 O)2+ n Complexes (n = 4 − 6) in Aqueous Solution and Gas Phase. J. Chem. Phys. 2005, 123, 114309.

[25] Austin, J. P.; Sundararajan, M.; Vincent, M. A.; Hillier, I. H. The Geometric Stuctures, Vibrational Frequencies and Redox Properties of the Actinyl Coordination Complexes − − ([AnO2 (L)n ]m ; An = U, Pu, Np; L = H2 O, Cl− , CO2− 3 , CH3 CO2 , OH ) in Aqueous Solution, Studied by Density Functional Theory Methods. Dalton Trans. 2009, 5902–5909. [26] Odoh, S. O.; Schreckenback, G. Theoretical Study of the Structural Properties of Plutonium(IV) and (VI) Complexes. J. Phys. Chem. A 2011, 115, 144110–14119. [27] Chaumont, A.; Engler, E.; Wipff, G. Uranyl and Strontium Salt Solvation in RoomTemperature Ionic Liquids. A Molecular Dynamics Investigation. Inorg. Chem. 2003, 42, 5348–5356. [28] Chaumont, A.; Wipff, G. Solvation of Uranyl(II), Europium(III) and Europium(II) Cations in Basic Room-Temperature Ionic Liquids: A Theoretical Study. Chem. Eur. J. 2004, 10, 3919–3930. [29] Chaumont, A.; Wipff, G. Solvation of Uranyl(II) and Europium(III) Cations and Their Chloro Complexes in a Room-Temperature Ionic Liquid. A Theoretical Study of the Effect of Solvent ”Humidity”. Inorg. Chem. 2004, 43, 5891–5901. [30] Chaumont, A.; Wipff, G. Solvation of Uranyl-CMPO Complexes in Dry vs. Humid Forms of the [BMI][PF6] Ionic Liquid. A Molecular Dynamics Study. Phys. Chem. Chem. Phys. 2006, 8, 494–502. [31] Schurhammer, R.; Wipff, G. Solvation of Uranium Hexachloro Complexes in RoomTemperature Ionic Liquids. A Molecular Dynamics Investigation in Two Liquids. J. Phys. Chem. B 2007, 111, 4659–4668.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 42

[32] Gaillard, C.; Chaumont, A.; Billard, I.; Hennig, C.; Ouadi, A.; Wipff, G. Uranyl Coordination in Ionic Liquids: The Competition Between Ionic Liquid Anions, Uranyl Counterions, and Cl− Anions Investigated by Extended X-ray Absorption Fine Structure and UV-Visible Spectroscopies and Molecular Dynamics Simulations. Inorg. Chem. 2007, 46, 4815–4826. [33] Chaumont, A.; Wipff, G. Chloride Complexation by Uranyl in a Room Temperature Ionic Liquid. A Computational Study. J. Phys. Chem. B 2008, 112, 12014–12023. [34] Gaillard, C.; Chaumont, A.; Billard, I.; Hennig, C.; Ouadi, A.; Georg, S.; Wipff, G. Competitive Complexation of Nitrates and Chlorides to Uranyl in a Room Temperature Ionic Liquid. Inorg. Chem. 2010, 49, 6484–6494. [35] Chaumont, A.; Klimchuk, O.; Gaillard, C.; Billard, I.; Ouadi, A.; Hennig, C.; Wipff, G. Perrhenate Complexation by Uranyl in Traditional Solvents and in Ionic Liquids: A Joint Molecular Dynamics/Spectroscopic Study. J. Phys. Chem. B 2012, 116, 3205–3219. [36] Anthony, J. L.; Maginn, E. J.; Brennecke, J. F. Solution Thermodynamics of ImidazoliumBased Ionic Liquids and Water. J. Phys. Chem. B 2001, 105, 10942–10949. [37] Freire, M. G.; Carvalho, P. J.; Gardas, R. L.; Marrucho, I. M.; Santos, L. M. N. B. F.; Coutinho, J. A. P. Mutual Solubilities of Water and the [Cn mim][Tf2 N] Hydrophobic Ionic Liquids. J. Phys. Chem. B 2008, 112, 1604–1610. [38] Zhong, X.; Liu, Z.; Cao, D. Improved Classical United-Atom Force Field for ImidazoliumBased Ionic Liquids: Tetrafluoroborate, Hexafluorophosphate, Methylsulfate, Trifluoromethylsulfonate, Acetate, Trifluoroacetate, and Bis(trifluoromethylsulfonyl)amide. J. Phys. Chem. B 2011, 115, 10027–10040. [39] Zhong, X.; Fan, Z.; Liu, Z.; Cao, D. Local Structure Evolution and its Connection to Thermodynamics and Transport Properties of 1-Butyl-3-methylimidazolium Tetrafluoroborate and Water Mixtures by Molecular Dynamics Simulation. J. Phys. Chem. B 2012, 116, 3249– 3263. [40] Rai, N.; Tiwari, S. P.; Maginn, E. J. Force Field Development for Actinyl Ions via Quantum Mechanical Calculations: An Approach to Account for Many Body Solvation Effects. J. Phys. Chem. B 2012, 116, 10885–10897. [41] Pomogaev, V.; Tiwari, S. P.; Rai, N.; Goff, G. S.; Runde, W. H.; Batista, E. R.; 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. 2 2 2 2 Phys. DOI:10.1039/C3CP52444B. [42] Zhang, Y.; Maginn, E. J. A Simple AIMD Approach to Derive Atomic Charges for Condensed Phase Simulation of Ionic Liquids. J. Phys. Chem. B 2012, 116, 10036–10048. [43] Dommert, F.; Wendler, K.; Berger, R.; Site, L. D.; Holm, C. Force Fields for Studying the Structure and Dynamics of Ionic Liquids: A Critical Review of Recent Developments. ChemPhysChem. 2012, 13, 1625–1637. [44] Berendsen, H. J. C.; Grigera, J. R.; Straatsma, T. P. The Missing Term in Effective Pair Potentials. J. Phys. Chem. 1987, 91, 6269–6271. [45] Hess, B.; Kutzner, C.; van der Spoel, D.; Lindahl, E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J. Chem. Theory Comput. 2008, 4, 435–447. 34 ACS Paragon Plus Environment

Page 35 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[46] van der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark, A. E.; Berendsen, H. J. C. GROMACS: Fast, Flexible, and Free. J. Comp. Chem. 2005, 26, 1701–1719. [47] Lindahl, E.; Hess, B.; van der Spoel, D. GROMACS 3.0: A Package for Molecular Simulation and Trajectory Analysis. J. Mol. Mod. 2001, 7, 306–317. [48] Berendsen, H. J. C.; van der Spoel, D.; van Drunen, R. GROMACS: A Message-Passing Parallel Molecular Dynamics Implementation. Comp. Phys. Comm. 1995, 91, 43–56. [49] Torrie, G. M.; Valleau, J. P. Monte Carlo Free Energy Estimates Using Non-Boltzmann Sampling - Application to Subcritical Lennard-Jones Fluid. Chem. Phys. Lett. 1974, 28, 578–581. [50] Torrie, G. M.; Valleau, J. P. Nonphysical Sampling Distributions in Monte Carlo FreeEnergy Estimation: Umbrella Sampling. J. Comp. Phys. 1977, 23, 187–199. [51] Kirkwood, J. G. Statistical Mechanics of Fluid Mixtures. J. Chem. Phys. 1935, 3, 300–313. [52] Kumar, S.; Bouzida, D.; Swendsen, R. H.; Kollman, P. A.; Rosenberg, J. M. The Weighted Histogram Analysis Method for Free Energy Calculation on Biomolecules 1. The Method. J. Comput. Chem. 1992, 13, 1011–1021. [53] Ferrenberg, A. M.; Swendsen, R. H. New Monte Carlo Technique for Studying Phase Transitions. Phys. Rev. Lett. 1988, 61, 2635–2638. [54] Ferrenberg, A. M.; Swendsen, R. H. Optimized Monte Carlo Data Analysis. Phys. Rev. Lett. 1989, 63, 1195–1198. [55] Hub, J. S.; de Groot, B. L.; van der Spoel, D. g wham–A Free Weighted Histogram Analysis Implementation Including Robust Error and Autocorrelation Estimates. J. Chem. Theory Comput. 2010, 6, 3713–3720. [56] Vallet, V.; Wahlgren, U.; Schimmelpfennig, B.; Szabó, Z.; Grenthe, I. The Mechanism for Water Exchange in [UO2 (H2 O)5 ]2+ and [UO2 (oxalate)2 (H2 O)]2− as Studied by Quantum Chemical Methods. J. Am. Chem. Soc. 2001, 123, 11999–12008. [57] Clavaguéra-Sarrio, C.; Brenner, V.; Hoyau, S.; Marsden, C. J.; Milli/’e, P.; Dognon, J.-P. Modeling of Uranyl Cation-Water Clusters. J. Phys. Chem. B 2003, 107, 3051–3060. [58] Bühl, M.; Diss, R.; Wipff, G. Coordination Environment of Aqueous Uranyl(VI) Ion. J. Am. Chem. Soc. 2005, 127, 13506–13507. [59] Rotzinger, F. P. The Water-Exchange Mechanism of the [UO2 (OH2 )5 ]2+ Ion Revisited: The Importance of a Proper Treatment of Electron Correlation. Chem. Eur. J. 2007, 13, 800–811. [60] Tsushima, S. Hydration and Water-Exchange Mechanism of the UO2+ 2 Ion Revisited: The Validity of the ”n+1” Model. J. Phys. Chem. A 2007, 111, 3613–3617. [61] Nichols, P.; Bylaska, E. J.; Schenter, G. K.; de Jong, W. Equatorial and Apical Solvent Shells of the UO2+ 2 Ion. J. Chem. Phys. 2008, 128, 124507. [62] Frick, R. J.; Hofer, T. S.; Pribil, A. B.; Randolf, B. R.; Rode, B. M. Structure and Dynamics of the UO2+ 2 Ion in Aqueous Solution: An Ab Initio QMCF-MD Study. J. Phys. Chem. A 2009, 113, 12496–12503. [63] Kerisit, S.; Liu, C. Molecular Simulation of the Diffusion of Uranyl Carbonate Species in Aqueous Solution. Geochim. Cosmochim. Acta 2010, 74, 4937–4952. 35 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 42

[64] Hagberg, D.; Karlström, G.; Roos, B. O.; Gagliardi, L. The Coordination of Uranyl in Water: A Combined Quantum Chemical and Molecular Simulation Study. J. Am. Chem. Soc. 2005, 127, 14250–14256. [65] Bühl, M.; Kabrede, H.; Diss, R.; Wipff, G. Effect of Hydration on Coordination Properties of Uranyl(VI) Complexes. A First Principles Molecular Dynamics Study. J. Am. Chem. Soc. 2006, 128, 6357–6368. [66] Aberg, M.; Ferri, D.; Glaser, J.; Grenthe, I. Structure of the Hydrated Dioxouranium(VI) Ion in Aqueous Solution. An X-ray Diffraction and 1H NMR Study. Inorg. Chem. 1983, 22, 3986–3989. [67] Allen, P. G.; Bucher, J. J.; Shuh, D. K.; Edelstein, N. M.; Reich, T. Investigation of Aquo + 4+ and Chloro Complexes of UO2+ and Pu3+ by X-ray Absorption Fine Structure 2 , NpO2 , Np Spectroscopy. Inorg. Chem. 1997, 36, 4676–4683. [68] Wahlgren, U.; Moll, H.; Grenthe, I.; Schimmelpfennig, B.; Maron, L.; Vallet, V.; Gropen, O. Structure of Uranium(VI) in Strong Alkaline Solutions. A Combined Theoretical and Experimental Investigation. J. Phys. Chem. A 1999, 103, 8257–8264. [69] Semon, L.; Boehme, C.; Billard, I.; Hennig, C.; Lutzenkirchen, K.; Reich, T.; Rossberg, A.; Rossini, I.; Wipff, G. Do Perchlorate and Triflate anions Bind to the Uranyl Cation in an Acidic Aqueous Medium? A Combinded EXAFS and Quantum Mechanical Investigation. Chem. Phys. Chem. 2001, 2, 591–598. [70] Neuefeind, J.; Soderholm, L.; Skanthakumar, S. Experimental Coordination Enviroment of Uranyl(VI) in Aqueous Solution. J. Phys. Chem. A 2004, 108, 2733–2739. [71] Soderholm, L.; Skanthakumar, S.; Neuefeind, J. Determination of Actinide Speciation in Solution Using High-Energy X-ray Scattering. Anal. Bioanal. Chem. 2005, 383, 48–55. [72] Bardin, N.; Rubini, P.; Madic, C. Hydration of Actinyl(VI), MO2+ 2 (M = U, Np, Pu). An NMR Study. Radiochim. Acta 1998, 83, 189–194. [73] Guilbaud, P.; Wipff, G. Force Field Representation of the UO2+ Cation from Free En2 − ergy MD Simulations in Water. Tests on its 18-Crown-6 and NO3 Adducts, and on its Calix[6]arene6 - and CMPO Complexes. J. Mol. Struct.: THEOCHEM 1996, 366, 55–63. [74] Tsushima, S.; Yang, T.; Suzuki, A. Theoretical Gibbs Free Energy Study on UO2 (H2 O)2+ n and its Hydrolysis Products. Chem. Phys. Lett. 2001, 334, 365–373. [75] Ikeda, Y.; Soya, S.; Fukutomi, H.; Tomiyasu, H. Nuclear Magnetic Resonance Study of the Kinetics of the Water Exchange Process in the Equatorial Positions of Uranyl Complexes. J. Inorg. Nucl. Chem. 1979, 41, 1333–1337. [76] Gaillard, C.; Azzi, A. E.; Billard, I.; Bolvin, H.; Hennig, C. Uranyl Complexation in Fluorinated Acids (HF, HBF4 , HPF6 , HTf2N): A Combined Experimental and Theoretical Study. Inorg. Chem. 2005, 44, 852–861. [77] Nockemann, P.; Servaes, K.; van Deun, R.; van Hecke, K.; van Meervelt, L.; Binnemans, K.; Gorller-Walrand, C. Speciation of Uranyl Complexes in Ionic Liquids by Optical Spectroscopy. Inorg. Chem. 2007, 46, 11335–11344. [78] Pasilis, S. P.; Blumenfeld, A. Effect of Nitrate, Perchlorate, and Water on Uranyl(VI) Speciation in a Room-Temperature Ionic Liquid: A Spectroscopic Investigation. Inorg. Chem. 2011, 50, 8302–8307. 36 ACS Paragon Plus Environment

Page 37 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[79] Cammarata, L.; Kazarian, S. G.; Salter, P. A.; Welton, T. Molecular States of Water in Room Temperature Ionic Liquids. Phys. Chem. Chem. Phys. 2001, 3, 5192–5200. [80] Kelkar, M. S.; Maginn, E. J. Effect of Temperature and Water Content on the Shear Viscoscity of the Ionic Liquid 1-Ethyl-3-methylimidazolium Bis(trifluoromethanesulfonyl)imide as Studied by Atomistic Simulations. J. Phys. Chem. B 2007, 111, 4867–4876. [81] Chang, T. M.; Dang, L. X.; Devanathan, R.; Dupuis, M. Structure and Dynamics of N, N Diethyl-N-methylammonium Triflate Ionic Liquid, Neat and with Water, from Molecular Dynamics Simulations. J. Phys. Chem. A 2010, 114, 12764–12774. [82] Gutowski, K. E.; Dixon, D. A. Predicting the Energy of the Water Exchange Reaction and Free Energy of Solvation for the Uranyl Ion in Aqueous Solution. J. Phys. Chem. A 2006, 110, 8840–8856. [83] John, G. H.; May, I.; Sharrad, C. A.; Sutton, A. D.; Collison, K.; Helliwell, M.; Sarsfield, M. J. The Synthesis, Structural, and Spectroscopic Characterization of Uranium(IV) Perrhenate complexes. Inorg. Chem. 2005, 44, 7606–7615. [84] John, G. H.; May, I.; Sarsfield, M. J.; Collison, D.; Helliwell, M. Dimeric Uranyl Complexes with Bridging Perrhenates. Dalton Trans. 2007, 1603–1610. [85] Görller-Walrand, C.; Servaes, K. The Photochemical Behavior of the Uranyl Ion (UO2+ 2 ): A Dimer Perspective and Outlook. Helv. Chim. Acta 2009, 92, 2304–2312. [86] Tsushima, S.; Brendler, V.; Fahmy, K. Aqueous Coordination Chemistry and Photochemistry of Uranyl(VI) Oxalate Revisited: A Density Functional Theory Study. Dalton Trans. 2011, 40, 10953–10958. [87] Feldman, I.; North, C. A.; Hunter, H. B. Equilibrium Constants for the Formation of Polynuclear Tridentate 1:1 Chelates in Uranyl-Malate, -Citrate, and -Tartrate systems. J. Phys. Chem. 1930, 64, 1224–1230. [88] Allen, P. G.; Bucher, J. J.; Clark, D. L.; Edelstein, N. M.; Ekberg, S. A.; Gohdes, J. W.; Hudson, E. A.; Kaltsoyannis, N.; Lukens, W. W.; Neu, M. P. et al. Multinuclear NMR, Raman, EXAFS, and X-ray Diffraction Studies of Uranyl Carbonate Complexes in NearNeutral Aqueous Solution. X-ray Structure of [(C(NH2 )3 ][(UO2 )3 (CO3 )6 ] · 6H2 O. Inorg. Chem. 1995, 34, 4797–4807. [89] Allen, P. G.; Shuh, D. K.; Bucher, J. J.; Edelstein, N. M.; Reich, T.; Denecke, M. A.; Nitsche, H. EXAFS Determinations of Uranium Structures: The Uranyl Ion Complexed with Tartaric, Citric, and Malic Acids. Inorg. Chem. 1996, 35, 784–787. [90] Grenthe, I.; Dro˙zd˙zyński, J.; Fujino, T.; Buck, E. C.; Albrecht-Schmitt, T. E.; Wolf, S. F. Uranium. In The Chemistry of the Actinide and Transactinide Elements, 3rd ed.; Morss, L. R., Edelstein, N. M., Fuger, J., Katz, J. J., Eds.; Springer: Dordrecht, The Netherlands, 2008; Vol. 1; pp 253–698. [91] Aberg, M. On the Structures of the Predominant Hydrolysis Products of Uranyl(VI) in Solution. Acta Chem. Scand. 1970, 24, 2901–2915. [92] Perry, D. L.; Ruben, H.; Templeton, D. H.; Zalkin, A. Structure of Di-µ-aquobis(dioxobis(nitrato)uranium(VI) Diimidazole, [UO2 (NO3 )2 (H2 O)2 ] · 2C3 H4 N2 , a WaterBridged Dimer of Uranyl Nitrate. Inorg. Chem. 1980, 19, 1067–1069.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 42

[93] Toth, L. M.; Begun, G. M. Raman Spectra of Urnayl Ion and its Hydrolysis Products in Qqueous HNO3 . J. Phys. Chem. 1981, 85, 547–549. [94] Faulques, E.; Russo, R. E.; Perry, D. L. Raman Studies of Uranyl Nitrate and its Hydroxy Bridged Dimer. Spectrochimica Acta 1994, 50A, 757–763. [95] Barnes, C. E.; Shin, Y.; Saengkerdsub, S.; Dai, S. EXAFS Study of Uranyl Nitrate Dimer at High and Low Temperature. Inorg. Chem. 2008, 39, 862–864. [96] Quilés, F.; Burneau, A. Infrared and Raman Spectra of Uranyl(VI) Oxo-Hydroxo Complexes in Acid Aqueous Solutions: A Chemometric Study. Vibrational Spectroscopy 2000, 23, 231–241. [97] Fujii, T.; Fujiwara, K.; Ramana, H.; Moriyama, H. Raman Spectroscopic Determination of Formation Constant of Uranyl Hydrolysis Species (UO2 )2 (OH2 )2+ . J. Alloys Compd. 2001, 323–324, 859–863. [98] Tsushima, S.; Rossberg, A.; Ikeda, A.; Müller, K.; Scheinost, A. C. Stoichiometry and Structure of Uranyl(VI) Hydroxo Dimer and Trimer Complexes in Aqueous Solution. Inorg. Chem. 2007, 46, 10819–10826. [99] Nockemann, P.; van Deun, R.; Thijs, B.; Huys, D.; Vanecht, E.; van Hecke, K.; van Meervelt, L.; Binnemans, K. Uranyl Complexes of Carboxyl-Functionalized Ionic Liquids. Inorg. Chem. 2010, 49, 3351–3360. [100] Chen, X.-Y.; Goff, G. S.; Quiroz-Guzman, M.; D. P. Fagnant, J.; Brennecke, J. F.; Scott, B. L.; Runde, W. Directed Nucleation of Monomeric and Dimeric Uranium(VI) Complexes with a Room Temperature Carboxyl-Functionalized Phosphonium Ionic Liquid. Chem. Commun. 2013, 49, 1903–1905. [101] Kannan, S.; Rajalakshmi, N.; Chetty, K. V.; Venugopal, V.; Drew, M. G. B. Complexes of the Uranyl Ion with Bis(diphenylphosphino)methane Dioxide. Polyhedron 2004, 23, 1527– 1533. [102] Kannan, S.; Chetty, K. V.; Usman, A.; Fun, H. K. Extraction and Coordination Studies of Carbamoyl Methyl Sulfoxide (CMSO) with Uranyl Bis(beta-diketonates). Synthesis and Molecular Structure of [UO2 (DBM)22 C6 H5 CH2 SOCH2 CONHC6 H5 ]. Polyhedron 2007, 26, 1383–1387. [103] Zhu, L.; Yuan, D.; Li, B.; Li, H. Syntheses and Structures of Three Disulfoxide Uranyl Complexes. J. Coord. Chem. 2010, 63, 3006–3015. [104] Thuéry, P. Uranyl Ion Complexation by Citric and Tricarballylic Acids: Hydrothermal Synthesis and Structure of Two- and Three-Dimentional Uranium-Organic Frameworks. Chem. Commun. 2006, 853–855. [105] Bhatt, A. I.; May, I.; Volkovich, V. A.; Collison, D.; Helliwell, M.; Polovov, I. B.; Lewin, R. G. Structural Characterization of a Lanthanum Bistrilimide Complex, La(N(SO2 CF3 )2 )3 (H2 O)3 , and an Investigation of La, Sm, and Eu Electrochemistry in a Room-Temperature Ionic Liquid, [Me3 NBu][N(SO2 CF3 )2 ]. Inorg. Chem. 2005, 44, 4934– 4940. [106] Babai, A.; Mudring, A.-V. The First Homoleptic Bis(trifluoromethanesulfonyl)amide Complex Compounds of Trivalent f-Elements. Dalton Trans. 2006, 1828–1830.


Page 39 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[107] Kawamura, M.; Shimada, S. Crystal Structures of [M(NTf2)3 (H2 O)n ] (M = Eu and Yb) and a Template Effect of Linear Molecules on the 3D Hydrogen Bonding Network Structure of [Eu(NTf2)3 (H2 O)3 ]. Inorg. Chim. Acta 2007, 360, 2162–2168. [108] Babai, A.; Mudring, A.-V. The First Homoleptic Bis(trifluoromethanesulfonyl)amide Complex of Yttrium: [bmim][Y(Tf2N)4 ]. Z. Anorg. Allg. Chem. 2008, 634, 938–940. [109] Babai, A.; Mudring, A.-V. Anhydrous Praseodymium Salts in the Ionic Liquid [bmpyr][Tf2N]: Structural and Optical Properties of [bmpyr]4 [PrI6 ][Tf2N] and [bmpyr]2 [Pr(Tf2N)5 ]. Chem. Mater. 2005, 17, 6230–6238. [110] Tang, S.; Babai, A.; Mudring, A.-V. Europium-Based Ionic Liquids as Luminescent Soft Materials. Angew. Chem. Int. Ed. 2008, 47, 7631–7634. [111] Williams, D. B.; Stoll, M. E.; Scott, B. L.; Costa, D. A.; W. J. Oldham, J. Coordination Chemistry of the Bis(trifluoromethanesulfonyl)imide) Anion: Molecular Interactions in Room Temperature Ionic Liquids. Chem. Commun. 2005, 1438–1440. [112] Babai, A.; Mudring, A.-V. Homoleptic Alkaline Earth Metal Bis(trifluoromethanesulfonyl)imide Complex Compounds Obtained from an Ionic Liquid. Inorg. Chem. 2006, 45, 3249–3255. [113] Matsumoto, K.; Hagiwara, R.; Tamada, O. Coordination Enviroment Around the Lithium Cation in Solid Li2 (EMIm)(N(SO2 CF3 )2 )3 (EMIm = 1-Ethyl-3-methylimidazolium): Structural Clue of Ionic Liquid Electrolytes for Lithium Batteries. Solid State Sci. 2006, 8, 1103– 1107. [114] Zhou, Q.; Fitzgerald, K.; Boyle, P. D.; Henderson, W. A. Phase Behavior and Crystalline Phases of Ionic Liquid-Lithium Salt Mixtures with 1-Alkyl-3-methylimidazolium Salts. Chem. Mater. 2010, 22, 1203–1208. [115] Zhou, Q.; Boyle, P. D.; Malpezzi, L.; Mele, A.; Shin, J.-H.; Passerini, S.; Henderson, W. A. Phase Behavior of Ionic Liquid-LiX Mixtures: Pyrrolidinium Cations and TFSI Anions Linking Structure to Transport Properties. Chem. Mater. 2011, 23, 4331–4337. [116] Schmeisser, M.; Heinemann, F. W.; Illner, P.; Puchta, R.; Zahl, A.; van Eldik, R. Coordination of 1,10-Phenanthroline and 2,2’-Bipyridine to Li in Different Ionic Liquids. How Innocent are Ionic Liquids? Inorg. Chem. 2011, 50, 6685–6695. [117] Lasségues, J.-C.; Grondin, J.; Aupetit, C.; Johansson, P. Spectroscopic Identification of the Lithium Ion Transporting Species in LiTFSI-Doped Ionic Liquids. J. Phys. Chem. A 2009, 113, 305–314. [118] Borodin, O.; Smith, G. D. LiTFSI Structure and Transport in Ethylene Carbonate from Molecular Dynamics Simulations. J. Phys. Chem. B 2006, 110, 4971–4977. [119] Borodin, O.; Smith, G. D.; Henderson, W. A. Li Cation Environment, Transport, and Mechanical Properties of the LiTFSI Doped n-Methyl-n-alkylpyrrolidinium TFSI Ionic Liquids. J. Phys. Chem. B 2006, 110, 16879–16886. [120] Li, Z.; Smith, G. D.; Bedrov, D. Li Solvation and Transport Properties in Ionic Liquid/Lithium Salt Mixtures: A Molecular Dynamics Simulation Study. J. Phys. Chem. B 2012, 116, 12801–12809. [121] Jiang, W.; Wang, Y.; Voth, G. A. Molecular Dynamics Simulation of Nanostructural Organization in Ionic Liquid/Water Mixtures. J. Phys. Chem. B 2007, 111, 4812–4818. 39 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 42

[122] Spohr, H. V.; Patey, G. N. The Influence of Water on the Structural and Transport Properties of Model Ionic Liquids. J. Chem. Phys. 2010, 132, 234510. [123] Vallet, V.; Privalov, T.; Wahlgren, U.; Grenthe, I. The Mechanism of Water Exchange 5+ in AmO2 (H2 O)2+ and NpO2 (H2 O)2+ 5 and in the Isoelectronic UO2 (H2 O) 5 Complexes as Studied by Quantum Chemical Methods. J. Am. Chem. Soc. 2004, 126, 7766–7767. [124] Bühl, M.; Kabrede, H. Mechanism of Water Exchange in Aqueous Uranyl(VI) Ion. A Density Functional Molecular Dynamics Study. Inorg. Chem. 2006, 45, 3834–3836. [125] Sornein, M.-O.; Cannes, C.; le Naour, C.; Lagarde, G.; Simoni, E.; Barthet, J.-C. Uranyl Complexation by Chloride Ions. Formation of a Tetrachlorouranium(VI) Complex in Room Temperature Ionic Liquids [BMIM][Tf2 N] and [MeBu3 N][Tf2 N]. Inorg. Chem. 2006, 45, 10419–10421. [126] Bühl, M.; Sieffert, N.; Golubnychiy, V.; ; Wipff, G. Density Functional Theory Study of Uranium(VI) Aquo Chloro Complexes in Aqueous Solution. J. Phys. Chem. A 2008, 112, 2428–2436. [127] Soderholm, L.; Skanthakumar, S.; Wilson, R. E. Structural Correspondence Between Uranyl Chloride Complexes in Solution and Their Stability Constants. J. Phys. Chem. A 2011, 115, 4959–4967. [128] Impey, R. W.; Madden, P. A.; McDonald, I. R. Hydration of Mobility of Ions in Solution. J. Phys. Chem. 1983, 87, 5071–5083. [129] Giridhar, P.; Venkatesan, K. A.; Srinivasan, T. G.; Rao, P. R. V. Electrochemical Behavior of Uranium(VI) in 1-Butyl-3-methylimidazolium Chloride and Thermal Characterization of Uranium Oxide Deposit. Electrochimica Acta 2007, 52, 3006–3012. [130] Rao, C. J.; Venkatesan, K. A.; Nagarajan, K.; Srinivasan, T. G. Dissolution of Uranium Oxides and Electrochemical Behavior of U(VI) in Task Specific Ionic Liquid. Radiochim. Acta 2008, 96, 403–409. [131] Ikeda, Y.; Hiroe, K.; Asanuma, N.; Shirai, A. Electrochemical Studies on Uranyl(VI) Chloride Complexes in Ionic Liquid, 1-Butyl-3-methylimidazolium Chloride. J. of Nucl. Sci. and Technol. 2009, 46, 158–162. [132] Sornein, M.-O.; Cannes, C.; le Naour, C.; Hennig, C. Electrochemical Behavior of Tetrachloro and Tetrabromo Uranyl Complexes in Room Temperature Ionic Liquids. J. Electroanal. Chem. 2011, 661, 49–56. [133] Zhang, Q.; Huang, X.; Tang, H.; He, H. Electrochemical Behavior of Uranium(VI) in 1Butyl-3-methylimidazolium Chloride. Journal of Nuclear and Radiochemistry 2011, 33, 101–105. [134] Ogura, T.; Sasaki, K.; Takao, K.; Arai, T.; Ikeda, Y. Electrochemical Behavior of [UO2 Cl4 ]2− in 1-Ethyl-3-methylimidazolium Based Ionic Liquids. Sci. China Chem. 2012, 55, 1699–1704. [135] Menjoge, A.; Dixon, J.; Brennecke, J. F.; Maginn, E. J.; Vasenkov, S. Influence of Water on Diffusion in Imidazolium-Based Ionic Liquids: A Pulsed Field Gradient NMR Study. J. Phys. Chem. B 2009, 113, 6353–6359. [136] Urahata, S. M.; Ribeiro, M. C. C. Single Particle Dynamics in Ionic Liquids of 1-Alkyl-3methylimidazolium Cations. J. Chem. Phys. 2005, 122, 024511. 40 ACS Paragon Plus Environment

Page 41 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


[137] Schröder, U.; Wadhawan, J. D.; Compton, R. G.; Marken, F.; Suarez, P. A. Z.; Consorti, C. S.; de Souza, R. F.; Dupont, J. Water-Induced Accelerated Ion Diffusion: Voltammetric Studies in 1-Methyl-3-[2,6-(S)-dimethylocten-2-yl]imidazolium Tetrafluoroborate, 1-Butyl-3-methylimidazolium Tetrafluoroborate and Hexafluorophosphate Ionic Liquids. New J. Chem. 2000, 24, 1009–1015. [138] Liu, H.; Maginn, E. J. A Molecular Dynamics Investigation of the Structural and Dynamic Properties of the Ionic Liquid 1-n-Butyl-3-methylimidazolium Bis(trifluoromethanesulfonyl)imide. J. Chem. Phys. 2011, 135, 124507.



20

G(r) [kJ/mol]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 42

0 -20 -40

r 5 H2O 1 H2O

-60 0.2

0.4

0.6

0.8

r [nm]

Figure 13: For Table of Contents Only


1

Structure and Dynamics of Uranyl(VI) and Plutonyl(VI) Cations in Ionic

Recommend Documents