| Although high-temperature superconductivity in layered copper oxides (cuprates) was discovered more than twenty years ago, the nature of the high temperature superconductivity still remains elusive. The discovery of the Fe-based superconductors with the maximum transition temperature of 56K has broken the cuprates monopoly in the physics of high temperature superconducting compounds and opened a new avenue for investigating the superconducting mechanism. In this dissertation I will focus on iron chalcogenide, which is an important member of Fe-based superconductors. Similar to other classes of unconventional superconductors, the superconductivity in Fe-based superconductors can be achieved by suppressing the long-range antiferromagnetic order of the parent compounds through either pressure or charge carrier doping. Although the superconducting pairing mechanism of Fe-based supercondcutors has not yet been identified, a great deal of experiments has shown that the superconducting pairing in these materials is associated with magnetic spin fluctuations. Further clarification of the relationship between magnetism and superconductivity in these materials is among the central topics in the field of superconductivity.;For iron pnictides, such as LaO1-xF xFeAs and Ba1-xK xFe2As2, the antiferromagnetism of undoped parent compounds is widely believed to be driven by a spin density wave instability arising from the nesting of two Fermi surface (FS) pockets by a vector Q =(pi, pi) (in the units of the inverse tetragonal lattice parameters.). The FS nesting vector corresponds to the wavevector of the AFM order. However, this itinerant model is not applicable to the antiferromagnetism of the parent compound of iron chalcogenides, since its antiferromagnetic wavevector is Q AF = (pi, 0), 45° rotated relative to the FS nesting vector. Although the parent compound of iron chalcogenides possesses an antiferromagnetic state distinct from that of iron pnictides, the superconducting state of the optimally-doped iron chalcogeinde exhibits spin resonance in magnetic excitation spectra, similar to that seen in optimally-doped iron pnictide superconductors; spin resonance is observed at the same wavevector (pi, pi) in both types of materials. This suggests that both classes of materials might have the same magnetic origin for superconducting pairing. Therefore, resolution of the dichotomy between (pi, 0) magnetic order in the parent compound FeTe and superconductivity with (pi, pi) magnetic resonance in Se-substituted samples is a key challenge to our emerging understanding of iron-based superconductivity.;In this dissertation we aim to elucidate the puzzling relationship between magnetism and superconductivity by studying the evolution of superconductivity and magnetism in Fe1.02(Te1-xSe x). We first performed preliminary studies of the evolution of superconductivity, magnetism, and structural transition in Fe1.22 (Te1-xSex) using polycrystalline samples. Our results and analyses suggest that superconductivity in this system is associated with magnetic fluctuations and therefore may be unconventional in nature.;In follow-up studies, we established the complete phase diagram of electronic and magnetic properties for Fe1.02(Te1- xSex) using high-quality single crystal samples. We find that for low Se content long-range AFM order is formed with a magnetic wave vector (pi, 0). Dynamic magnetic correlations with a (pi, pi) wave vector, however, do co-exist in a wide range of the phase diagram. Increasing Se doping tunes the relative strength of these distinct correlations. Bulk superconductivity occurs only in a composition range where (pi, 0) magnetic correlations are sufficiently suppressed and (pi, pi) spin fluctuations associated with the nearly nesting Fermi surface dominate. This indicates that iron chalcogenide and iron pnictide superconductors, despite a competing magnetic instability in the former, have a similar mechanism for superconductivity.;In addition, we address another important issue in Fe(Te, Se) system: the effect of excess iron at interstitial sites of the (Te, Se) layers on electronic properties. Our results show that the excess Fe not only suppresses superconductivity, but also leads to weak charge carrier localization. Our results suggest that such weak charge carrier localization is related to the magnetic coupling between the excess Fe and the adjacent Fe sheets, which is responsible for the superconductivity suppression caused by the excess Fe. |
