Doping Effects on the Kondo Lattice Materials: FeSi, CeCoin5, and YbInCu4
Three doping studies on Kondo lattices are investigated in this thesis: FeSi1-xGex, Ce1-xLaxCoIn5, and Yb1-xYxInCu4. For FeSi1-xGex, we constructed the phase diagram through the analysis of magnetic, thermal and transport measurements on single crystals. The phase diagram shows a first-order transit...
Other Authors: | |
---|---|
Format: | Others |
Language: | English English |
Published: |
Florida State University
|
Subjects: | |
Online Access: | http://purl.flvc.org/fsu/fd/FSU_migr_etd-0646 |
Summary: | Three doping studies on Kondo lattices are investigated in this thesis: FeSi1-xGex, Ce1-xLaxCoIn5, and Yb1-xYxInCu4. For FeSi1-xGex, we constructed the phase diagram through the analysis of magnetic, thermal and transport measurements on single crystals. The phase diagram shows a first-order transition from a Kondo insulator (exponentially activated properties) to a ferromagnetic metal at a critical concentration, xc ~ 0:25. The field dependence of the magnetization (M(H)) shows that the saturation moment of x = 0:27 is 10 times larger than that of x = 0:24. The spin gap of x = 0:24, 167K, is quite close to the transition temperature of x = 0:27, 150K, indicating that the characteristic energies of the two competing phases, i.e. the Curie temperature and the spin gap of the Kondo insulator, are essentially equal at the critical concentration. For x c, spin gap, transport gap and resistivity minimum systematically decrease with increasing x. Saturation moments and specific heat coefficients are almost zero for x c. The temperature dependence of magnetic susceptibility (X(T)) for x = 0:2 shows a broad maximum around 200K, indicating that the broad maximum temperature decreases with x for x c. The variable range hopping analysis suggests the existence of the localized state for this region. For x > x xc, the data break into two distinct regimes: xc» 0:5 and » 0:5 · 1. For xc 0:5, X(T) does not displays a sharp transition at Tc and M(H) increases with increasing fields. The temperature dependence of the resistivity (ρ(T)) shows metallic behavior. However, it does not have any kink at Tc. In contrast, for ~ 0:5 · 1, X(T) displays a sharp transition at Tc and M(H) saturates at H ~ 0:3T. ρ (T) has a kink at Tc. Based on the Kondo insulator picture, we can explain the specific heat coefficient y evolution with x. The transition from a Kondo insulator to a ferromagnetic metal can be explained as the consequence of the changes in hybridization between Fe 3d electrons and Si/Ge p conduction electrons in conjunction with disorder on the Si/Ge ligand site. For Ce1-xLaxCoIn5, we studied antiferromagnetic intersite correlations for the Kondo lattice by comparison with data on the single Kondo impurity. All the magnetic susceptibility per mole Ce for H || ab plane and H || c axis collapse onto one curve above 100K in Ce1-xLaxCoIn5, indicating the same high T Kondo temperature (~ 35K) for all concentrations. Further, the magnetic part of the resistivity shows the same -logT dependence above 50K for all concentrations, again indicating that the high T Kondo temperature is essentially independent of Ce concentration. The magnetic part of the heat capacity for Ce1-xLaxCoIn5 alloys has a peak around 70K, suggesting the same crystalline field splittings occurs the alloy series Ce1-xLaxCoIn5. Based on these experimental findings, the scaling laws for the susceptibility and the heat capacity reveal that the screening of the magnetic moments in this Kondo lattice involves antiferromagnetic intersite correlations and this intersite correlation has a larger energy scale compared to the Kondo impurity case. In addition, a Fermi liquid ground state appears in the La rich region while the specific heat and inelastic part of ρm show non-Fermi liquid behavior for Ce rich region. For Yb1-xYxInCu4, measurements using cantilever torque magnetometry discover the new phase above Hv for x = 0 and x = 0:1. With proper scaling of the critical fields and temperatures, data for all alloys collapse onto the same curve, representing a common phase above Hv. The magneto-resistance does not change at the new phase boundary. Due to the crystalline electric field, there is anisotropy of the valence transition in applied magnetic field in different directions. For x = 0:2, the specific heat and the resistance indicate the appearance of a spin glass state below 4K for H > 5T. Since Ytterbium occupies the corners of a tetrahedron in the F43m structure, the spin glass state is not unexpected. === A Dissertation submitted to the Department of Physics in partial fulfillment of the
requirements for the degree of Doctor of Philosophy. === Degree Awarded: Fall Semester, 2003. === Date of Defense: November 4, 2003. === Kondo Lattice, Mixed Valence State, Kondo Insulator, Heavy Fermion Superconductor === Includes bibliographical references. === Zachary Fisk, Professor Directing Dissertation; Naresh Dalal, Outside Committee Member; Stephan von Molnar, Committee Member; Nicholas Bonesteel, Committee Member; Jorge Piekarewicz, Committee Member. |
---|