Session T23: Actinides and Transport Magnetism in Metals

A quantum Monte Carlo study of thorium halide molecules

We present electronic structure calculations of thorium halide molecules, namely, a series of $ ThCl_n $ and $ ThBr_n $ (n=1,2,3,4) systems. We calculate the bond dissociation energies for the sequence of the following dissociation reactions: $ ThCl_n -> ThCl_{n-1} + Cl $ and $ ThBr_n -> ThBr_{n-1}+Br $. We apply both large core and small core energy-consistent pseudopotentials for thorium atom and we employ DFT GGA and hybrid functionals for energies and geometry calculations. For high accuracy energies we use the fixed-node diffusion Monte Carlo (DMC) method. In DMC calculations we employ both single- and multi-reference trial wave functions in order to test the quality of the nodal surfaces. We study the dissociation energies with regard to different multiplicities of the $ ThX_{n} $ molecules and the different behavior of the bromides with respect to chlorides. We compare our results with hybrid DFT methods and experimental values. Supported by DOE and NSF.   

Electron binding energy of uranium-ligand and uranyl-ligand anions

Electron binding energies of the early actinide element uranium in gas-phase anion complexes are calculated by relativistic density functional theory (DFT) with two different exchange-correlation functions (RPBE and B3LYP) and also in the Hartree-Fock (HF) approximation\footnote{ADF2010.02, SCM.com}. Scalar and spin-orbit calculations are performed, and the calculated energies are compared to available experimental measurements and shown to disagree by energies of order 1 eV. Strong correlations that are poorly treated in DFT and HF can be included by a hybrid approach in which a generalized Anderson impurity model is numerically diagonalized. Reduction-oxidation (redox) potentials of aqueous actinide ions show improved agreement with measured values in the hybrid approach\footnote{S. E. Horowitz and J. B. Marston, J. Chem. Phys {\bf 134} 064510 (2011).}. We test whether or not similar improvements are found in the gas-phase.   

Electronic structure of Pu metal and intermetallics as determined by ARPES and XPS

We exploit our unique capability in angle resolved photoemission (ARPES) measurement for plutonium materials to compare the electronic structure of delta Pu, PuCoIn5 and PuCoGa5. We present a systematic study for polycrystalline delta Pu metal as well as the first photoemission measurements on single crystals of the superconductor PuCoIn5. These first results for PuCoIn5 are compared to results for PuCoGa5 and variations in the electronic structure are attributed to differences in 5f hybridization with conduction states based on differences in lattice values. Laser ablation was used to clean the surface of Pu materials and cleanliness was monitored through O 1s and Pu 4f core levels as well as valence band features. Observation of these spectra provides better insight into the differentiation of contaminant features versus the strong correlation effects within Pu metal.   

Elastic moduli of nearly pure polycrystalline plutonium

We measure elastic moduli of microalloyed poly-crystalline cylindrical specimen of Pu-239. We observe $\alpha \rightarrow \beta \rightarrow \gamma \rightarrow \delta$ phase transitions and find that the elastic moduli of nearly pure plutonium are the same as those of Ga-stabilized plutonium.   

Magnetization at 1 Mbar Near the Tricritical Point in the Ce(0.9-$x)$La($x)$Th(0.10) System

The gamma to alpha isostructural transition in the Ce(0.9-$x)$La($x)$Th(0.10) system is measured as a function of La alloying and external pressure up to 1 Mbar using magnetic susceptibility. We probe a line of discontinuous transitions, as indicated by the change in volume, decreasing exponentially from 118 K to close to 0 K with increasing La doping, and the transition changes from being first order to continuous at a critical concentration close to $x$ = 0.14. At the tricritical point, the magnetic susceptibility increase rapidly near the critical concentration and approaches large values at $x$= 0.35 signifying that a heavy Fermi-liquid state evolves at large doping near the critical concentration. The Wilson ratio reaches a value above two for a narrow range of concentrations where the specific heat and susceptibility vary most rapidly with the doping concentration.   

Actinide oxides under pressure

We use the self-interaction corrected local spin density approximation to investigate the oxidation of actinide dioxides under pressure. The methodology enables us to determine the ground state valency configuration of the actinide 5f electrons and to study the localization/delocalization transition that occurs under pressure. We argue that this delocalization facilitates the oxidation of the actinide dioxides and present results for the estimated transition pressures.   

Dephasing due to electron interactions in inhomogeneous systems

At sufficiently low temperatures, the dephasing time $\tau_\varphi$ of mesoscopic samples is governed by so-called Johnson-Nyquist (electronic) noise. We study the spatial dependence of the corresponding noise correlation function (NFC) in inhomogeneous systems. Using the fluctuation-dissipation theorem and the random-phase approximation, we derive a real-space integro-differential equation for the NCF and show that it reduces to a diffusion equation in the case of strong screening. In particular, using a method based on the spectral determinant, we evaluate the NCF for arbitrary networks of quasi-1D disordered wires with boundary conditions. As an application, we construct a realistic quantum dot model via a set of parallel wires connected at contacts to leads, and calculate the temperature dependence of $\tau_\varphi$ as well as the quantum corrections to the conductance. Furthermore, we analyze the observability of the elusive 0D regime (reached at $T < E_{\rm Thouless}$ with the characteristic $\tau_\varphi \propto T^{-2}$ behavior) in such systems, and discuss alternative scenarios of its observation.   

Does aluminum conduct better than copper at the nanoscale? A first-principles study of metallic nanowires

From first-principles, we present a theoretical and comparative investigation of the role of quantum confinement in altering the electronic, transport, and phonon properties of linear, single-atom thick chains, i.e. nanowires, of metallic (Au, Ag, Cu, and Al) atoms. Our results for the ballistic quantum transport properties and electronic structure are in perfect agreement with those previously published. Motivated by this, we also consider electron-phonon interactions in such devices, where we report an order of magnitude reduction in the electron-phonon coupling constant for Al, whereas an enhancement is predicted for Au, Ag, and Cu.   

Quantum oscillations and low-T resistivity of the 5s delafossite PdCoO$_2$

We report dHvA torque and resistivity data on the highly conductive delafossite PdCoO$_2$. Quantum oscillations confirm electronic structure calculations in showing a single, highly 2D Fermi surface with almost exactly half filling. However the cyclotron masses and the details of the warping are consistent with PdCoO$_2$ being a 5s metal, rather than the 4d metal indicated in most electronic structure papers. The room temperature resistivity is $\rho_{ab}=2.6 \pm 0.2$~$\mu\Omega$-cm, lower than all elemental metals apart from the noble metals. The low-T residual resistivity of our samples is 0.008~$\mu\Omega$-cm, corresponding to an extremely long mean free path of 20~$\mu$m, surprising for a flux-grown material. The temperature dependence of the electron-phonon contribution to the resistivity is exponential rather than $T^5$, indicating phonon drag, and the only observation of phonon drag in resistivity outside the alkali metals.   

Dielectric function of Ni-Pt alloys from 0.6 to 6.6 eV by spectroscopic ellipsometry

The complex dielectric function of different Ni-Pt alloys (10{\%} to 25{\%} Pt concentration, 10nm thickness) was determined using spectroscopic ellipsometry over a broad photon energy range from 0.6 to 6.6eV. The films were deposited on a thick SiO$_{2}$ layers using Si as a substrate. The data were fitted using previously determined optical constants for Si and SiO$_{2}$. Optical constants of the Ni-Pt alloys were described using a Drude model (free carrier term), a pole due to d-intraband transition, and 2 to 3 Lorentz oscillators due to interband transitions. Data were compared with the calculated band structure of Nickel and Platinum. Results showed only small changes with the variation of Pt concentration or with annealing at 500\r{ }C for 30s.   

The Role of $d$-Orbitals in the Rashba Splitting on Au(111) and Ag(111)

We investigate the Rashba-type spin splitting in $sp$-derived Shockley surface states on (111) surfaces of noble metals, such as Au(111) and Ag(111), based on first-principles calculations including the spin-orbit interaction. By turning on and off $l$-dependent spin-orbit coupling one by one, we find that although the surface states on Au(111) have predominantly $p$-orbital character, the spin splitting in energy originates mainly from $d$-orbital character of the surface states. We also demonstrate that the spin splitting in surface states of both metallic surfaces of Au(111) and Ag(111) can be controlled by varying the sizes of $d$-orbital parts of the surface-state wave functions. These results show that in addition to difference in the atomic spin-orbit strength in Au and Ag, difference in $d$-orbital contributions to the surface states makes substantial difference in the sizes of the Rashba-type spin splitting in their surface electronic structures. This work was supported by the NRF of Korea (Grant Nos. 2009-0081204 and 2011-0018306) and KISTI Supercomputing Center (Project No. KSC-2011-C2-04).   

Effects of Co doping on the metamagnetic states in fcc Fe$_{1-x}$Co$_{x}$

It is well known that fcc-Fe have shows metamagnetism, with a low-spin state (LS) at small volume and and a high-spin state (HS) at large volume in the total-energy vs volume curve. In this work, we have studied the evolution of the metamagnetic states in the Fe$_{1-x}$Co$_{x}$ alloy as a function of Co concentration by means of first principles calculations. The ground state properties were obtained using the Full-Potential Linear Augmented Plane Waves method and the Generalized Gradient Approximation for the exchange-correlation functional. The alloying was modeled using the self-consistent virtual crystal approximation. The magnetic states are obtained from the total-energy as a function of the spin moment calculations, obtained using the Fixed Spin Moment methodology. For fcc-Fe, we found that the ground state corresponds to the LS state. Increasing the Co concentration the HS state decrease in energy. Thus, for $x=0.05$ the energy of the LS and HS states is practically the same, corresponding to a spin-glass state. The LS state is substituted by a paramagnetic state for $x>0.3$ of Co concentration. Interestingly, for the alloy with $x\sim0.35$ the total-energy vs volume curve shows ``effective symmetry,'' which is expected to exhibit invar behavior.   

Investigation of Possible Phase Separation in Rapidly Quenched Fe$_{1-x}$Co$_x$ Alloys with Cluster Expansion Model

Recently, experiments observed phase separation in rapidly quenched Fe$_{1-x}$Co$_x$ alloy in the ordered $\alpha'$ phase [ J. Alloys Compd. 424, 145 (2006) ]. It is also believed that this is not an equilibrium result because of phase rule violation in the published phase diagram. To clarify this situation, we calculate the phase diagram of the system using cluster expansion in combination with a genetic algorithm. We calculated free energy of the system using super-cells up to 32 atoms with compositions ranging from 35 to 100 at. \% Fe. Possible explanations for the experimental observations will be discussed.   

Atomic Structures and Magnetic Properties of Fe-rich Fe$_{1-x}$Co$_x$ Alloys: A Genetic Algorithm Search

Using genetic algorithm with first-principles calculations, we performed a broad global search for low-energy crystal structures of Fe-rich Fe$_{1-x}$Co$_x$ alloys. We found that Fe-rich Fe$_{1-x}$Co$_x$ alloys are highly configurationally degenerate and there are many additional off-stoichiometric stable structures to the well-known stoichiometric FeCo - B$_2$ structure, giving a possibility for atomistic manipulation of the alloys. The Co-Co nearest-neighbor pair is strongly unfavorable in Fe-rich Fe$_{1-x}$Co$_x$ alloys. The magnetic moment of Fe atom is increasing with Co concentration while that of Co atom is almost constant, inducing a Slater-Pauling curve for magnetic moment per atom. The magnetic moment of Fe atom is strongly dependent on the number of Co nearest-neighbors and it increases with this number.