David Lewis Clark

We describe the wartime challenges associated with the rapid developments in plutonium chemistry and metallurgy that were necessary to produce the core of the Trinity Device. Beginning with microgram quantities of plutonium metal late in 1943, initial measurements showed a wide and confusing variance in density and other properties. These confusing results were the first clues to the astounding complexity of plutonium. As this complexity was revealed, it introduced new challenges for the fabrication of kilogram-scale parts. In a remarkable period from January 1944 to June 1945, Manhattan Project scientists made rapid progress in understanding plutonium chemistry and metallurgy. By early 1945, they had discovered five of the six ambient-pressure phases of unalloyed plutonium and reported the density of these phases to within a value of 0.1 g/cm 3 of those accepted today. They solved the stability problem introduced by these phases with a rapid alloy development program that ultimately identified gallium as the preferred element to stabilize the δ-phase, producing a plutonium alloy still of scientific and technical interest today. We conclude with a description of postwar developments in these areas, including applications of wartime plutonium metallurgy to civilian applications in nuclear reactors. We dedicate this paper to the memory of Ed Hammel, the Manhattan Project plutonium metallurgist whose previous description and documentation of plutonium history during the war has been essential in our research.

The synthesis and characterization of transuranium elements played an important and in the history of the periodic table. Prior to their discovery, elements up to uranium were thought to be 6d elements, and in the late 1930s, they were placed in the periodic table as such. The discovery of neptunium and plutonium and the determination of their chemical properties suggested that valence electrons were entering 5f orbitals and represented the emergence of a new series in the periodic table. The original characterization of americium and curium failed, until it was realized that they may behave more like uranium and plutonium than transition elements. Glenn Seaborg introduced the actinide concept and proposed rearranging the periodic table to create a new 5f actinide series akin to the 4f lanthanide series. Due to imperfect screening of the 5f electrons, and the subsequent changes in the energetics of both the 5f and 6d orbitals as one progresses along the series, the actinides experience an actinide contraction and display a fascinating set of periodic trends. Moving from left to right, the valence 5f electrons contract and lose their ability to form chemical bonds. The crossover from bonding (itinerant) to ionic (magnetic) behavior gives rise to many exotic and interesting chemical and physical properties and has challenged modern approaches to electronic structure both in theory and experiment. The multiplicity of oxidation states, coupled with the hydrolysis behavior of the aqueous ions, makes the chemical behavior of protactinium through americium among the most complex of the elements in the periodic table.

Understanding the nature of covalent (band-like) vs ionic (atomic-like) electrons in metal oxides continues to be at the forefront of research in the physical sciences. In particular, the development of a coherent and quantitative model of bonding and electronic structure for the lanthanide dioxides, LnO(2) (Ln = Ce, Pr, and Tb), has remained a considerable challenge for both experiment and theory. Herein, relative changes in mixing between the O 2p orbitals and the Ln 4f and 5d orbitals in LnO(2) are evaluated quantitatively using O K-edge X-ray absorption spectroscopy (XAS) obtained with a scanning transmission X-ray microscope and density functional theory (DFT) calculations. For each LnO(2), the results reveal significant amounts of Ln 5d and O 2p mixing in the orbitals of t(2)g (sigma-bonding) and e(g) (pi-bonding) symmetry. The remarkable agreement between experiment and theory also shows that significant mixing with the O 2p orbitals occurs in a band derived from the 4f orbitals of a(2u) symmetry (sigma-bonding) for each compound. However, a large increase in orbital mixing is observed for PrO2 that is ascribed to a unique interaction derived from the 4f orbitals of t(1u) symmetry (sigma- and pi-bonding). O K-edge XAS and DFT results are compared with complementary L-3-edge and M-5,M-4-edge XAS measurements and configuration interaction calculations, which shows that each spectroscopic approach provides evidence for ground state O 2p and Ln 4f orbital mixing despite inducing very different core-hole potentials in the final state.

Mixed valence O-doped UO2+x. and photoexcited UO2 containing transitory U3+ and U5+ host a coherent polaronic quantum phase (CPQP) that exhibits the characteristics of a Frohlich-type, nonequilibrium, phononcoupled Bose-Einstein condensate whose stability and coherence are amplified by collective, anharmonic motions of atoms and charges. Complementary to the available, detailed, real space information from scattering and EXAFS, an outstanding question is the electronic structure. Mapping the Mott gap in UO2, U4O9, and U3O7 with O XAS and NIXS and UM5 RIXS shows that O doping raises the peak of the U5f states of the valence band by similar to 0.4 eV relative to a calculated value of 0.25 eV. However, it lowers the edge of the conduction band by 1.5 eV vs the calculated 0.6 eV, a difference much larger than the experimental error. This 1.9 eV reduction in the gap width constitutes most of the 2-2.2 eV gap measured by optical absorption. In addition, the XAS spectra show a tail that will intersect the occupied U5f states and give a continuous density-of-states that increases rapidly above its constricted intersection. Femtosecond-resolved photoemission measurements of UO2, coincident with the excitation pulse with 4.7 eV excitation, show the unoccupied U5f states of UO2 and no hot electrons. 3.1 eV excitation, however, complements the O-doping results by giving a continuous population of electrons for several eV above the Fermi level. The CPQP in photoexcited UO2 therefore fulfills the criteria for a nonequilibrium condensate. The electron distributions resulting from both excitations persist for 5-10 ps, indicating that they are the final state that therefore forms without passing through the initial continuous distribution of nonthermal electrons observed for other materials. Three exceptional findings are: (1) the direct formation of both of these long lived (> 3-10 ps) excited states without the short lived nonthermal intermediate; (2) the superthermal metallic state is as or more stable than typical photoinduced metallic phases; and (3) the absence of hot electrons accompanying the insulating UO2 excited state. This heterogeneous, nonequilibrium, Frohlich BEC stabilized by a Fano-Feshbach resonance therefore continues to exhibit unique properties.

Actinyl-tricarbonato anions [(AnO(2))(CO3)(3)](4-) (An = U-Cm) in various environments were investigated using theoretical approaches of quantum-mechanics, molecular-mechanics and cluster-models. Cations and solvent molecules in the 2nd coordination sphere affect the equatorial An <- O-eq bonds more than the axial An = O-ax bonds. Common actinide contraction is found for calculated and experimental axial bond lengths of U-92 to Pu-94, though no longer for Pu-94 to Cm-96. The tendency of U to Pu forming actinyl(VI) species dwindles away toward Cm, which already features the preferred An(III)/Ln(III) oxidation state of the later actinides and all lanthanides. The well known change from d-type to typical U-Pu-Cm type and then to f-type behavior is labeled as the plutonium turn, a phenomenon that is caused by f-orbital energy-decrease and f-orbital localization with increase of both nuclear charge and oxidation state, and a non-linear variation of effective f-electron population across the actinide series. Both orbital and configuration mixing and occupation of antibonding 5f type orbitals increase, weakening the AnO(ax) bonds and reducing the highest possible oxidation states of the later actinides.

The synthesis, electronic structure, and characterization via single-crystal X-ray diffraction, nuclear magnetic resonance (NMR) spectroscopy, and magnetic susceptibility of (Me4N)(2)PuCl6 are reported. NMR measurements were performed to both search for the direct Pu-239 resonance and to obtain local magnetic and electronic information at the Cl site through Cl-35 and Cl-37 spectra. No signature of Pu-239 NMR was observed. The temperature dependence of the Cl spectra was simulated by diagonalizing the Zeeman and quadrupolar Hamiltonians for Cl-35, Cl-37, and N-14 isotopes. Electronic structure calculations predict a magnetic Gamma(5) triplet ground state of Pu(IV) in the crystalline electric field of the undistorted PuCl6 octahedron. A tetragonal distortion would result in a very small splitting (similar to 20 cm(-1)) of the triplet ground state into a nonmagnetic singlet and a doublet state. The Cl shifts have an inflection point at T approximate to 15 K, differing from the bulk susceptibility, indicating a nonmagnetic crystal field ground state. The Cl spin-lattice relaxation time is constant to T = 15 K, below which it rapidly increases, also supporting the nonmagnetic crystal field ground state.

Hypervalent UO2, UO2(+x) formed by both addition of excess O and photoexcitation, exhibits a number of unusual or often unique properties that point to it hosting a polaronic Bose-Einstein(-Mott) condensate. A more thorough analysis of the O X-ray absorption spectra of UO2, U4O9, and U3O7 shows that the anomalous increase in the width of the spectral features assigned to predominantly U 5f and 6d final states that points to increased dispersion of these bands occurs on the low energy side corresponding to the upper edge of the gap bordered by the conduction or upper Hubbard band. The closing of the gap by 1.5 eV is more than twice as much as predicted by calculations, consistent with the dynamical polaron found by structural measurements. In addition to fostering the excitation that is the proposed mechanism for the coherence, the likely mirroring of this effect on the occupied, valence side of the gap below the Fermi level points to increased complexity of the electronic structure that could be associated with the Fermi topology of BEC-BCS crossover and two band superconductivity. (C) 2015 Published by Elsevier B.V.