Magnetism in Yb- and Ce-based heavy-fermion metals under pressure

For more than 30 years the investigation of heavy-fermion (HF) metals has been one of the most fascinating and interesting fields in condensed matter physics both experimentally and theoretically. The HF phenomenon is observed in compounds containing rare-earth elements such as, e.g., Ce or Yb. The ground-state properties of these systems is considered to result from a competition between the Kondo effect and the Ruderman-Kittel-Kasuya-Yosida (RKKY) interaction. The Kondo effect induces a screening of the local 4f moments by the conduction electrons and, thus, favors a nonmagnetic ground state. By contrast, the indirect RKKY interaction promotes a magnetic ordering of the local moments. Therefore, depending on the strength of these interactions, the ground state of a HF compound ranges from paramagnetic (PM) to magnetically ordered. Even interplay between magnetism and superconductivity is observed in some materials. High-pressure experiments are of particular interest because for a given system the ground state can be tuned by external pressure (p). In the case of Yb-based Kondo-lattice systems, pressure stabilizes the magnetic state, while for Ce-based Kondo-lattice compounds pressure favors a non-magnetic ground state. Therefore, pressure is an ideal tool to tune a HF material through a zero-temperature magnetic instability in order to study quantum critical behavior in its vicinity.

The present thesis addresses the effect of pressure, magnetic field, and temperature on the properties of the rare-earth compounds Yb(Rh0.94Ir0.06)2Si2, YbIr2Si2, and CeRuPO. Electrical resistivity measurements under pressure have been performed in order to investigate the evolution of the ground-state properties and to search for the putative QCP supposed to exist in the studied HF materials.

At atmospheric pressure, Yb(Rh0.94Ir0.06)2Si2 does not order down to 20mK. Applying a small pressure, AFM order is observed at low temperatures. Upon further increasing pressure, the magnetic transition temperature increases, and a second magnetic transition appears inside the ordered state. The electrical resistivity measurements under pressure suggest that, at ambient pressure, Yb(Rh0.94Ir0.06)2Si2 orders slightly below 20 mK. The T- p phase diagram of Yb(Rh0.94Ir0.06)2Si2 and of YbRh2Si2 can be superposed by shifting the pressure axis of Yb(Rh0.94Ir0.06)2Si2 by ?p = -0.06 GPa. The electrical resistivity studies indicate that in Yb(Rh0.94Ir0.06)2Si2, Ir substitution acts primarily as negative chemical pressure and disorder effects play only a minor role. The results point at the existence of a pressure (volume) controlled quantum critical point (QCP) at a hypothetical negative critical pressure of pc = -0.25GPa.

The ambient-pressure thermodynamic and transport properties of YbRh2Si2 reveal a PM ground state, in contrast to the Rh-homolog which possesses a magnetic ground state being situated in the direct vicinity to an AFM QCP. Application of pressure on YbRh2Si2 is expected to tune the system through a QCP, providing a unique opportunity to investigate the physical properties at and around a pressure (volume) controlled magnetic QCP in a clean stoichiometric Yb system. The Landau Fermi-liquid (LFL) state observed at low temperatures at ambient pressure survives in the pressure range up to pLFLc ÷ 3 GPa. With further increasing pressure, the resistivity shows a temperature dependence weaker than quadratic. This, so-called, non Fermi-liquid (NFL) behavior observed down to the lowest temperatures extends up to the critical pressure, pc ÷ 8 GPa, where magnetic order sets in. This broad region of NFL behavior might reveal the existence of a novel type of metallic phase. The magnetic order develops suddenly hinting at a first-order transition at pc. With further increasing pressure, the magnetic state is stabilized. The high-T resistivity studies allow to determine the pressure evolution of the Kondo and crystalline electric field (CEF) energy scales. In the PM region, the CEF splitting is independent of p, while TK decreases exponentially in the same pressure range. At about the critical pressure, a low-lying Kondo scale and the excited CEF levels at higher temperatures can be clearly resolved.

At ambient pressure CeRuPO is a FM Kondo-lattice system with TC = 14 K and TK ÷ 10 K. So

far, the behavior at a FM QCP in a Kondo-lattice system is not settled. Therefore, pressure studies on CeRuPO offer the great opportunity to investigate the suppression of FM order in a Ce-based Kondo-lattice system. Upon applying pressure the magnetic ordering temperature in CeRuPO shifts toward lower temperatures. Furthermore, the ground state of CeRuPO changes from FM to AFM order at p* ÷ (0.87 ö 1.01) GPa. Our results indicate a critical pressure of pc ÷ 3 GPa where the magnetic transition temperature is suppressed to zero temperature in a first-order like way. Therefore, we come to the conclusion that a magnetic QCP does not exist in CeRuPO. Beyond pc, LFL behavior was observed at low temperatures, in support of our previous conclusion. The pressure evolution of the high-T electrical resistivity cannot be understood in a simple picture for a Ce-based HF metal considering a dominant Kondo energy scale. At low pressures, the temperature dependence of the resistivity above the magnetic transition is strongly affected by the Kondo effect, magnetic fluctuations, and CEF splitting. At higher pressures, the contribution to ?(T) from Kondo scattering on the ground state and on excited CEF levels can be separated. Our study indicates that CeRuPO is a further example of a FM system in which application of pressure suppresses the magnetic order, but also destabilizes FM order in favour of an AFM one, preventing the appearance of a FM QCP.