Electron spectroscopy of strongly correlated 3d and 5f alloy systems.

Abstract

This thesis investigates correlated electron alloys using photoemission spectroscopy (PES) and inverse photoemission spectroscopy (BIS). These materials display a range of ground state phenomena such as antiferromagnetic insulator to metallic transitions, heavy-fermion properties, and high temperature superconductivity. The strong Coulomb intrasite repulsion of the 3d or 5f orbital, spin correlations, and the hybridization with the valence band are responsible for the unusual properties. The spectral weight is essentially the imaginary part of the single particle Green's function, which links the spectroscopic measurements to many-body Hamiltonians. Nd$\sb{2-x}{\rm Ce}\sb{x}{\rm CuO}\sb{4-y}$ is an antiferromagnetic insulator for x = 0 and a high temperature superconductor (HTS) for x = 0.15. Using PES, I have found that the insulator to metal transition occurs by a transfer of spectral weight into the insulating gap rather than by a shift of the chemical potential into the conduction band. This gap weight is found by angle-resolved photoemission spectroscopy (ARPES) to display a Fermi surface that obeys Luttinger's theorem. Ironically, the Luttinger theorem is consistent with Fermi liquid theory, but much experimental evidence supports non Fermi liquid properties for the normal state. The spectral lineshapes for the HTS's have almost linear onsets and tails extending to high binding energies. This suggests that the imaginary part of the self energy should have a functional dependence of the form $Im\Sigma({\bf k},\omega)=\Delta\vert\omega/\Delta\vert\sp\mu$ with a high energy crossover at $\omega=\Delta$ to 1/$\omega$ behavior. Such a scale invariant self energy has been used to fit ARPES spectra for the HTS's $\rm Bi\sb2Sr\sb2CaCu\sb2O\sb8,$ $\rm YBa\sb2Cu\sb3O\sb{6.9},$ and $\rm Nd\sb{1.85}Ce\sb{0.15}CuO\sb{4-\it y}.$. The alloy Y$\sb{1-x}{\rm U}\sb{x}{\rm Pd}\sb3$ resembles somewhat the Ce compounds, for which the the Anderson impurity Hamiltonian (AIH) describes the ground state properties and spectroscopy. As x decreases, all peaks shift relative to $E\sb{F}.$ This Fermi level tuning arises because the number of conduction electrons decreases as U$\sp{4+}$ is replaced by Y$\sp{3+}.$ The emergence of a peak near $E\sb{F}$ for small x signals a new ground state as expected in the AIH. To substantiate the results for Y$\sb{1-x}{\rm U}\sb{x}{\rm Pd}\sb3$ I have measured the related alloys Zr$\sb{1-x}{\rm U}\sb{x}{\rm Pd}\sb3,\ {\rm Th}\sb{1-x}{\rm U}\sb{x}{\rm Pd}\sb3,$ and Y$\sb{1-x}{\rm Pr}\sb{x}{\rm Pd}\sb3.$ The absence of tuning in these is consistent with the spectroscopic evidence that Zr and Th are tetravalent and Pr trivalent.