UCLA

Following the discovery of superconductivity in an Fe pnictide compound (LaFeAsO) in 2008, many other Fe-based superconductors were discovered with Tc as high as 56 K. Among them, the most extensive studies were carried out on the so called 122 system (AeFe2As2, Ae=Ba, Sr and Ca), because of the availability of large, high-quality single crystals. A recurring theme from these works is the rich interplay between antiferromagnetic and su- perconducting orders, and their relationship to structure and so-called electronic nematic instabilities. The research described here exploits two compounds with new and nontrivial crystal structures to open a new perspective on the important physics of the FBS.

Two newly discovered members of the FBS family, Ca1−xRExFeAs2 (CaRE112) and Ca10Pt3As8(Fe2As2)5 (10-3-8), possess intriguing structural characteristics and electronic band properties. As one of the most anisotropic FBSs, the critical temperature in the 10- 3-8 family can be induced up to 35 K by appropriate doping/external pressure while the CaRE112 compound can be doped into superconductors up to 47 K. They possess nontriv- ial structural and chemical characteristics. Firstly, structurally, unlike the other intensively studied pnictides which crystalize in tetragonal structure, 10-3-8 crystallizes in a triclinic structure. However, the “magic” FeAs layer, which is proposed to play a crucial role in mediating superconductivity, still maintains the local C4 rotational symmetry. On the other hand, CaRE112 has monoclinic structure at room temperature. However, owing to a unique spacer layer consisting of zigzag As chains in the crystal structure, FeAs layer loses its local C4 rotational symmetry even at room temperature, which is very unique in all FBSs. Sec- ondly, the nature of the spacer layers in 10-3-8 and CaRE112 are quite different from each other. The 10-3-8 family has the skutterudite Pt3As8 layer as the spacer layer, which can be assigned with integer number of oxidization states, thus the spacer layer will not con- tribute density of states at the Fermi level. On the other hand, the CaRE112 family has the zigzag chains as its spacer layer, which can not be assigned with integer oxidization states and contributes significant amount of density of states at the Fermi level. The fact that these two systems have similar FeAs interlayer distance but quite distinct characteristics of their spacer layers make them great systems to study the effect of the interlayer coupling on competing orders in FBSs.

In this thesis, I present a systematic experimental study, from synthesis to characteriza- tion, for both materials. I make combined transport, thermodynamic, neutron scattering, and muon spin relaxation measurements to investigate the interplay of competing orders and elaborate the role of the interlayer coupling on these competing orders. It has been reported previously that Ca10Pt3As8(Fe2As2)5, the parent compound of the 10-3-8 family, shows structural/magnetic instabilities. By substituting Co on Fe sites in the 10-3-8 family, the structural/magnetic phase transitions are suppressed and superconductivity up to 13.5 K is stabilized in an extended dome-like region in the temperature–dopant concentration phase diagram. More importantly, we demonstrate that within our experimental resolu- tion, no phase coexistence of antiferromagnetism and superconductivity exists in Co doped 10-3-8. Our research on the CaRE112 (RE = La, Ce, Pr, Nd) system is pioneering. Our refined synthesis recipes make us the first group to grow sizable CaRE112 single crystals with controlled doping. We identify the parent phase of the CaLa112 system, in which a monoclinic to triclinic structural phase transition and a paramagnetic to stripe-like antifer- romagnetic phase transitions are clearly evidenced. In addition, the metallic nature of its spacer layers is demonstrated. By Co doping on the Fe sites in CaLa112, we suppress the structural/magnetic phase transitions and induce superconductivity up to 20 K in a dome shaped region in the temperature-dopant concentration phase diagram. Our measurements of the superconducting and magnetic volume fractions show that these two phases coexist microscopically in the underdoped region, in contrast to the Co doped 10-3-8 compound, where coexistence is absent. Supported by model calculations, we discuss the differences in the phase diagrams of the 112 and 10-3-8 compounds in terms of the FeAs interlayer coupling, whose strength is affected by the character of the spacer layer, which is metallic in the 112 and insulating in the 10-3-8. Finally, we extend the discussion from CaLa112 to CaRE112 (RE = Ce, Pr, Nd). The structural and magnetic phase transitions of the FeAs layer are revealed in Ca0.71RE0.29FeAs2 (RE = Ce, Pr, Nd). Using Ca0.71Ce0.29FeAs2 as a representative, we demonstrate that an antiferromagentic ordering of Ce strongly entangled with the Fe moments develops at low temperatures. When Co is doped on the Fe sites in Ca0.71Ce0.29FeAs2, we show although Co doping suppresses the magnetic/structural ordering of the FeAs layer, it has little effect on the Ce ordering. We argue the lack of bulk super- conductivity in Co-doped Ca0.71Ce0.29FeAs2 arises from the excess electron doping of FeAs layer.