Density Functional Studies of Actinyl Aquo Complexes Studied Using Small-Core Effective Core Potentials and a Scalar Four-Component Relativistic Method

Theoretical actinide molecular science

Interest in the chemistry of the early actinide elements (notably uranium through americium) usually results either from the nuclear waste problem or the unique chemistry of these elements that result from 5f contributions to bonding. Computational actinide chemistry provides one useful tool for studying these processes. Theoretical actinide chemistry is challenging because three principal axes of approximation have to be optimized. These are the model chemistry (the choice of approximate electron-electron correlation method and basis sets), the approximate relativistic method, and a method for modeling solvent (condensed phase) effects. In this Account, we arrange these approximations in a three-dimensional diagram, implying that they are relatively independent of each other. A fourth level of approximation concerns the choice of suitable models for situations too complex to treat in their entirety. We discuss test cases for each of these approximations. Gas-phase data for uranium fluorides and oxofluorides such as UF(6) and UO(2)F(2) show that GGA functionals provide accurate geometries and frequencies while hybrid density functional theory (DFT) functionals are superior for energetics. MP2 is seen to be somewhat erratic for this set of compounds, and CCSD(T) gives the most accurate results. Three different relativistic methods, small-core effective core potentials (SC-ECP), ZORA, and all-electron scalar, provide comparable results. The older large-core ECP (LC-ECP) approach is consistently worse and should not be used. We confirmed these conclusions through studies of the actinyl aquo complexes [AnO(2)(OH(2))(5)](n+), (An = U, Np, or Pu and n = 1 or 2) that are also used to test solvation models. As long as the first coordination sphere of the metal is included explicitly, continuum solvation models are reliable, and we found no clear advantage for the (costly) explicit treatment of the second coordination sphere. Spin-orbit effects must be included to reproduce the correct trend in An(VI)/An(V) reduction potentials. We propose a multistep mechanism for the experimentally observed oxygen exchange of UO(2)(2+) cations in highly alkaline solutions present in tank wastes. This process involves an equilibrium between [UO(2)(OH)(4)](2-) and [UO(2)(OH)(5)](3-), followed by formation of the stable [UO(3)(OH)(3)](3-) intermediate that forms from [UO(2)(OH)(5)](3-) through intramolecular water elimination. The [UO(3)(OH)(3)](3-) intermediate facilitates oxygen exchange through proton shuttling. We explain the experimentally observed stabilization of the pentavalent oxidation state of actinyl ions by macrocyclic ligands (such as 18-crown-6) as an effect of solvation: the large macrocycle screens the positive charge of the ion from the polarizable solvent. Alkyl-substituted isoamethyrin complexes are bent despite being aromatic because of steric factors, rather than fit/misfit criteria regarding the actinyl ion. By application of an efficient DFT code, actinide molecules with more than 100 atoms can now be studied routinely. "Real" chemical questions can be answered as long as we take great care to apply methods that are accurate with respect to the three axes of approximation described above. While the exclusive focus of this Account has been on the early actinide elements, these conclusions also apply elsewhere in the periodic table.

Neptunium(VII) in high-ionic-strength alkaline solutions — [NpO2(OH)4]1– or [NpO4(OH)2]3–?

Relativistic density functional theory (ZORA-PBE, COSMO solvation) is used to address the title question, based on comparison with recent experimental data ( 1 ). Structural data (bond lengths), vibrational frequencies, and 17O NMR chemical shifts are used to prove that [NpO4(OH)2]3– is the predominant species in high-ionic-strength alkaline solutions of NpVII. Neptunium(VII) complexes have stronger bonds than their formally isoelectronic uranium(VI) analogues. The experimentally observed 300 ppm shift in 17O chemical shifts between the known [UO2(OH)4]2– and NpVII solution is shown to be partly a function of the central metal (NpVII vs. UVI) and not of the coordination environment (tetraoxo vs. dioxo). Comparing, for a given An (UVI or NpVII), actinyl complexes [AnO2X4]2–/1–, X = Cl, F, OH, a decreasing strength of the axial actinyl bond is observed that is traced to electronic factors (equatorial π-competition).

Structure of early actinides(V) in acidic solutions

Protactinium occupies a key position in the actinide series between thorium and uranium. In aqueous acidic solution, it is stable at oxidation state (V), occurring either as an oxocation or as a naked ion, depending on the media. Very few structural information on the hydration sphere of Pa(V) in acidic medium is available, in particular in hydrofluoric acid. Combined EXAFS and theoretical calculations have been used in this work to characterize the protactinium coordination sphere at various HF concentrations. The correlation of the XAFS data with quantum chemical calculations provides complementary structural and electronic models from ab initio techniques. At HF concentrations from 0.5 to 0.05 M, both theoretical calculations and EXAFS data suggest that the protactinium coordination sphere is mainly composed of fluoride ions. At the lowest HF concentration, the occurrence of a monooxo bond is observed with EXAFS, in agreement with the literature. A comparison of these data with related neptunium(V) and plutonium(V) diooxocations in perchloric acid is also presented.

Direct spectroscopic characterization of aqueous actinyl(VI) species: a comparative study of Np and U

The hydrolysis reactions of Np(VI) were investigated under an ambient atmosphere by attenuated total reflection Fourier transform infrared (ATR FT-IR) spectroscopy, NIR absorption spectroscopy, and speciation modeling applying the updated NEA thermodynamic database. For the first time, spectroscopic results of Np(VI) hydrolysis reactions are provided in the submillimolar concentration range and at pH values up to 5.3. The calculated speciation pattern and the results from FT-IR spectroscopy are comparatively discussed with results obtained from the U(VI) system under identical conditions. For both actinides, the formation of similar species can be derived from infrared spectroscopic results at pH values < or = 4, namely, the free cation AnO(2)(2+) (An = U, Np) and monomeric hydrolysis products. At higher pH, the infrared spectra evidence structurally different species contributing to the speciation of both actinides. At pH 5, the formation of a carbonate-containing dimeric complex, that is, (NpO(2))(2)CO(3)(OH)(3)(-), probably occurs during the hydrolysis reactions of neptunium, which is supported by the calculated speciation and results from NIR spectroscopy. For uranium, the presence of additional hydroxo complexes is assumed in this pH range. However, an unequivocal assignment of the spectral features to distinct species remains difficult. In particular, in the concentration range (0.5 mM) that constitutes the lower limit for the spectroscopic investigations of Np(VI) in the present work, monomeric and polymeric species obviously contribute to the U(VI) speciation considerably increasing the complexity of the spectral data.

Quantum mechanical calculation of aqueuous uranium complexes: carbonate, phosphate, organic and biomolecular species

BACKGROUND

Quantum mechanical calculations were performed on a variety of uranium species representing U(VI), U(V), U(IV), U-carbonates, U-phosphates, U-oxalates, U-catecholates, U-phosphodiesters, U-phosphorylated N-acetyl-glucosamine (NAG), and U-2-Keto-3-doxyoctanoate (KDO) with explicit solvation by H2O molecules. These models represent major U species in natural waters and complexes on bacterial surfaces. The model results are compared to observed EXAFS, IR, Raman and NMR spectra.

RESULTS

Agreement between experiment and theory is acceptable in most cases, and the reasons for discrepancies are discussed. Calculated Gibbs free energies are used to constrain which configurations are most likely to be stable under circumneutral pH conditions. Reduction of U(VI) to U(IV) is examined for the U-carbonate and U-catechol complexes.

CONCLUSION

Results on the potential energy differences between U(V)- and U(IV)-carbonate complexes suggest that the cause of slower disproportionation in this system is electrostatic repulsion between UO2 [CO3]3(5-) ions that must approach one another to form U(VI) and U(IV) rather than a change in thermodynamic stability. Calculations on U-catechol species are consistent with the observation that UO2(2+) can oxidize catechol and form quinone-like species. In addition, outer-sphere complexation is predicted to be the most stable for U-catechol interactions based on calculated energies and comparison to 13C NMR spectra. Outer-sphere complexes (i.e., ion pairs bridged by water molecules) are predicted to be comparable in Gibbs free energy to inner-sphere complexes for a model carboxylic acid. Complexation of uranyl to phosphorus-containing groups in extracellular polymeric substances is predicted to favor phosphonate groups, such as that found in phosphorylated NAG, rather than phosphodiesters, such as those in nucleic acids.

Lanthanide(III) and actinide(III) complexes [M(BH4)2 (THF)5][BPh4] and [M(BH4)2(18-crown-6)][BPh4] (M = Nd, Ce, U): synthesis, crystal structure, and density functional theory investigation of the covalent contribution to metal-borohydride bonding

Treatment of [M(BH4)3(THF)3] with NEt3HBPh4 in THF afforded the cationic complexes [M(BH4)2(THF)5][BPh4] [M = U (1), Nd (2), Ce (3)] which were transformed into [M(BH4)2(18-crown-6)][BPh4] [M = U (4), Nd (5), Ce (6)] in the presence of 18-crown-6; [U(BH4)2(18-thiacrown-6)][BPh4] (7) was obtained from 1 and 18-thiacrown-6 in tetrahydro-thiophene. Compounds 1, 3.C4H8S, 4.THF, 5, and 6.THF exhibit a penta- or hexagonal bipyramidal crystal structure with the two terdentate borohydride ligands in apical positions; the BH4 groups in the crystals of 7.C4H8S are in relative cis positions, and the thiacrown-ether presents a saddle shape, with two diametrically opposite sulfur atoms bound to uranium in trans positions. The crystal structures of these complexes, as well as those of previously reported [M(BH4)2(THF)5]+ cations, do not reveal any clear-cut lanthanide(III)/actinide(III) differentiation. The structural data obtained for [M(BH4)2(18-crown-6)]+ (M = U, Ce) by relativistic density functional theory (DFT) calculations are indicative of a small shortening of the U...B with respect to the Ce...B distance, which is accompanied by a lengthening of the U-Hb bonds and an opening of the Hb-B-Hb angle (Hb = bridging hydrogen atom of the eta3-BH4 ligand). The Mulliken population analysis and the natural bond orbital analysis indicate that the BH4 -->M(III) donation is greater for M = U than for M = Ce, as well as the overlap population of the M-Hb bond, thus showing a better interaction between the uranium 5f orbitals and the Hb atoms. The more covalent character of the B-H-U three-center two-electron bond was confirmed by the molecular orbital (MO) analysis. Three MOs represent the pi bonding interactions between U(III) and the three Hb atoms with significant 6d and 5f orbital contributions. These MOs in the cerium(III) complex exhibit a much lesser metallic weight with practically no participation of the 4f orbitals.

Catalytic cycles for the reduction of [UO2]2+ by cytochrome c7 proteins proposed from DFT calculations

The mechanism of the reduction of the hydrated uranyl cation, [UO2](2+), by the cytochromes G. sulfurreducens and D. acetoxidans has been studied using density functional theory calculations. We propose that the initial electron transfer step from the heme is to a cation-cation complex in the case of D. acetoxidans, but for G. sulfurreducens, it is to a single uranyl cation, which then forms a U(V)-U(VI) complex with a second uranyl cation. For both enzymes, the subsequent catalytic pathways are very similar. A U(V)-U(V) complex is formed, which then undergoes disproportionation via two successive protonation steps of one uranyl group, to give a U(VI)-U(IV) complex which dissociates to individual U(VI) and U(IV) species, the former being bound at the enzyme active site. Intermediate structures along the catalytic pathway are consistent with EXAFS data.

Relativistic density functional theory study of dioxoactinide(VI) and -(V) complexation with alaskaphyrin and related Schiff-base macrocyclic ligands

Formation of complexes of alaskaphyrin 1, bi-pyen 2 and bi-tpmd 3 ligands with actinyl ions AnO2(n+), An = U, Np, Pu and n = 1, 2, was studied using density functional theory (DFT) within the scalar relativistic four-component approximation. The alaskaphyrin complexes of the uranyl are predicted to have a bent conformation, in contrast to the experimentally available X-ray data. This deviation is likely due to crystal packing effects. Apart from these conformational differences, calculated geometry parameters and vibrational frequencies are in agreement with the available experimental data. The character of bonding in the complexes is investigated using bond order analysis and extended transition states (ETS) energy decomposition. Metal-to-ligand bonds can be described as primarily ionic although substantial charge transfer is observed as well. Based on ETS analysis, it is shown that steric and/or fit/misfit requirements of actinyl cations to the ligand cavities, among the studied complexes, are the most favorable for the bi-pyen ligand 2, because its flexibility allows for optimal metal-to-donor-atom distances. Planarity of the equatorial coordination sphere of the actinide atom is found to be less important than the ability of a ligand to provide optimal uranium-to-nitrogen bond lengths. Experimental differences in demetalation rates between similar alaskaphyrin, bi-pyen and bi-tpmd uranyl complexes are explained as a result of easier protonation of the Schiff-base nitrogen of the latter. Reduction potentials calculated for the uranium complexes show a good agreement with the experiment, both in relative and in absolute terms.

Electronic structure and bonding in actinyl ions and their analogs

This Feature Article seeks to present the current state of knowledge, both experimental and theoretical, of the electronic structure and bonding in actinyl ions and related species, such as the isoelectronic imido compounds as well as in linear triatomic actinide molecules of the type X-An-Y.

Hydration and Water-Exchange Mechanism of the UO22+ Ion Revisited: The Validity of the “n + 1” Model

Hydration and water-exchange mechanism of the UO2(2+) ion was studied by the B3LYP calculations. Relative Gibbs energies in aqueous phase of the 4-, 5-, and 6-fold uranyl(VI) hydrates were compared. A model with a complete first hydration shell and one water in the second shell was used (which is called "n + 1" model) to compare the energy of the UO2(2+) ion with different hydration numbers. The n + 1 model tends to overestimate the overall stability of the complex, and this type of model should be carefully used for the determination of the coordination number or the coordination mode such as unidentate or bidentate. A stable 5-fold uranyl(VI) hydrate goes through a very rapid water-exchange process via an associative (A-) mechanism keeping the 5-fold uranyl(VI) the dominant species.

Comparative Study of Uranyl(VI) and -(V) Carbonato Complexes in an Aqueous Solution

Electrochemical, complexation, and electronic properties of uranyl(VI) and -(V) carbonato complexes in an aqueous Na2CO3 solution have been investigated to define the appropriate conditions for preparing pure uranyl(V) samples and to understand the difference in coordination character between UO22+ and UO2+. Cyclic voltammetry using three different working electrodes of platinum, gold, and glassy carbon has suggested that the electrochemical reaction of uranyl(VI) carbonate species proceeds quasi-reversibly. Electrolysis of UO22+ has been performed in Na2CO3 solutions of more than 0.8 M with a limited pH range of 11.7 < pH < 12.0 using a platinum mesh electrode. It produces a high purity of the uranyl(V) carbonate solution, which has been confirmed to be stable for at least 2 weeks in a sealed glass cuvette. Extended X-ray absorption fine structure (EXAFS) measurements revealed the structural arrangement of uranyl(VI) and -(V) tricarbonato complexes, [UO2(CO3)3]n- [n = 4 for uranyl(VI), 5 for uranyl(V)]. The bond distances of U-Oax, U-Oeq, U-C, and U-Odist are determined to be 1.81, 2.44, 2.92, and 4.17 A for the uranyl(VI) complex and 1.91, 2.50, 2.93, and 4.23 A for the uranyl(V) complex, respectively. The validity of the structural parameters obtained from EXAFS has been supported by quantum chemical calculations for the uranyl(VI) complex. The uranium LI- and LIII-edge X-ray absorption near-edge structure spectra have been interpreted in terms of electron transitions and multiple-scattering features.

Interactions of 1-Methylimidazole with UO2(CH3CO2)2 and UO2(NO3)2: Structural, Spectroscopic, and Theoretical Evidence for Imidazole Binding to the Uranyl Ion

The first definitive high-resolution single-crystal X-ray structure for the coordination of the 1-methylimidazole (Meimid) ligand to UO2(Ac)2 (Ac = CH3CO2) is reported. The crystal structure evidence is confirmed by IR, Raman, and UV-vis spectroscopic data. Direct participation of the nitrogen atom of the Meimid ligand in binding to the uranium center is confirmed. Structural analysis at the DFT (B3LYP) level of theory showed a conformational difference of the Meimid ligand in the free gas-phase complex versus the solid state due to small energetic differences and crystal packing effects. Energetic analysis at the MP2 level in the gas phase supported stronger Meimid binding over H2O binding to both UO2(Ac)2 and UO2(NO3)2. In addition, self-consistent reaction field COSMO calculations were used to assess the aqueous phase energetics of combination and displacement reactions involving H2O and Meimid ligands to UO2R2 (R = Ac, NO3). For both UO2(NO3)2 and UO2(Ac)2, the displacement of H2O by Meimid was predicted to be energetically favorable, consistent with experimental results that suggest Meimid may bind uranyl at physiological pH. Also, log(Knitrate/KAc) calculations supported experimental evidence that the binding stoichiometry of the Meimid ligand is dependent upon the nature of the reactant uranyl complex. These results clearly demonstrate that imidazole binds to uranyl and suggest that binding of histidine residues to uranyl could occur under normal biological conditions.

Dimensional Dependence of Electronic Structure of Fullerene Polymers

Fullerene polymers made of C(60) are systematically investigated by means of a first-principles pseudopotential approach within the local density approximation of the density functional theory. We assume 10 different structures of fullerene polymers. The first three are C(60) polymer networks cross-linked by [2+2] cycloadditional four-membered rings, and the other seven are composed of peanut-shaped fused C(60) polymer chains cross-linked by either seven-membered rings or eight-membered rings. Owing to the overlap of wave functions as well as the hybrid networks of sp(2)-like (3-fold coordinated) and sp(3)-like (4-fold coordinated) carbon atoms, the electronic structure is considerably different from each other. We find that the resulting electronic structure is either semiconductor or semimetal depending on the spatial dimensionality of materials.

Quantum Chemical Calculations of Reduction Potentials of AnO22+/AnO2+ (An = U, Np, Pu, Am) and Fe3+/Fe2+ Couples

The reduction potentials of the AnO(2)(H(2)O)(5)(2+)/AnO(2)(H(2)O)(5)(+) couple (An = U, Np, Pu, and Am) and Fe(H(2)O)(6)(3+) to Fe(H(2)O)(6)(2+) in aqueous solution were calculated at MP2, CASPT2, and CCSD(T) levels of theory. Spin-orbit effects for all species were estimated at the CASSCF level. Solvation of the hydrated metal cations was modeled both by polarizable conductor model (PCM) calculation and by solvating the solutes with over one thousand TIP3P water molecules in the QM/MM framework. The redox reaction energy calculated by QM/MM method agreed well with the PCM method after corrections using the classical Born formula for the contribution from the rest of the solvation sphere and correction for dynamic response of solvent polarization in the MM region. Calculated reduction potentials inclusive of spin-orbit effect, zero-point energy, thermal corrections, entropy effect, and PCM solvation energy were found to be comparable with experimental data. The difference between CASPT2 calculated and experimental reduction energies were less than 35 kJ/mol in all cases, which ensures that CASPT2 (and CCSD(T)) calculations provide reasonable estimates of the thermochemistry of these reactions.

Experimental and Theoretical Investigations of IR Spectra and Electronic Structures of the U(OH)2, UO2(OH), and UO2(OH)2 Molecules

Reactions of laser-ablated U atoms and H2O2 molecules produce UO2, H2UO2, and UO2(OH)2 as major products and U(OH)2 and HU(O)OH as minor products. Complementary information is obtained from similar reactions of U atoms with D2O2, with H2 + O2 mixtures, and with H2O in excess Ar. Through extensive relativistic density functional theory calculations, we have determined the geometry structures and ground states of these U species with a variety of oxidation states U(II), U(IV), U(V), and U(VI). The calculated vibrational frequencies, IR intensities, and isotopic frequency ratios are in good agreement with the experimental values, thus supporting assignments of the observed matrix IR spectra. We propose that the reactions proceed by forming an energized [U(OH)4] intermediate from reactions of the excited U atom with two H2O2 molecules. Because of the special stability of the U(VI) oxidation state, this intermediate decomposes to the UO2(OH)2 molecule, which reveals a distinctive difference between the chemistries of U and Th, where the major product in analogous Th reactions is the tetrahedral Th(OH)4 molecule owing to the stable Th(IV) oxidation state.

Vibrational Spectroscopy of Mass-Selected [UO2(ligand)n]2+ Complexes in the Gas Phase: Comparison with Theory

The gas-phase infrared spectra of discrete uranyl ([UO2]2+) complexes ligated with acetone and/or acetonitrile were used to evaluate systematic trends of ligation on the position of the O=U=O stretch and to enable rigorous comparison with the results of computational studies. Ionic uranyl complexes isolated in a Fourier transform ion cyclotron resonance mass spectrometer were fragmented via infrared multiphoton dissociation using a free electron laser scanned over the mid-IR wavelengths. The asymmetric O=U=O stretching frequency was measured at 1017 cm(-1) for [UO2(CH3COCH3)2]2+ and was systematically red shifted to 1000 and 988 cm(-1) by the addition of a third and fourth acetone ligand, respectively, which was consistent with increased donation of electron density to the uranium center in complexes with higher coordination number. The values generated computationally using LDA, B3LYP, and ZORA-PW91 were in good agreement with experimental measurements. In contrast to the uranyl frequency shifts, the carbonyl frequencies of the acetone ligands were progressively blue shifted as the number of ligands increased from two to four and approached that of free acetone. This observation was consistent with the formation of weaker noncovalent bonds between uranium and the carbonyl oxygen as the extent of ligation increases. Similar trends were observed for [UO2(CH3CN)n]2+ complexes, although the uranyl asymmetric stretching frequencies were greater than those measured for acetone complexes having equivalent coordination, which is consistent with the fact that acetonitrile is a weaker nucleophile than is acetone. This conclusion was confirmed by the uranyl stretching frequencies measured for mixed acetone/acetonitrile complexes, which showed that substitution of one acetone for one acetonitrile produced a modest red shift of 3-6 cm(-1).