The response of the worldwide scientific community to the discovery in 2008 of superconductivity at T c = 26 K in the Fe-based compound LaFeAsO1−x F x has been very enthusiastic. In short order, ot...

The puzzle of high temperature superconductivity in layered iron pnictides and chalcogenides

Over the past decade, discussion of the interaction between superconductivity and magnetism has been overshadowed by the omnipresence of the oxide‐based, high‐temperature superconductors. But interest in the interaction between these two generally competing effects has a history that predates high‐Tc materials by several decades. Starting with seminal work by Bernd Matthias and his coworkers, it was found that magnetic impurities strongly suppress superconductivity in pure elements and binary compounds. This rapid suppression of the superconducting transition temperature Tc was due to the local magnetic moment of the impurity preventing the formation of the spin‐up/spin‐down conduction‐electron pairs that are responsible for superconductivity. The early measurements were made on dilute alloys, and for many solid‐state physicists of the time, the quest was to find compounds in which superconductivity coexists with an ordered lattice of local magnetic moments.

New Magnetic Superconductors: A Toy Box for Solid‐State Physicists

The interaction of rare-earth magnetism and superconductivity has been a topic of interest for many years. In classical magnetic superconductors (Chevrel phases, ternary rhodium borides, etc) as well as in the high-Tc cuprates the superconducting state usually coexists with antiferromagnetic order on the rare-earth sublattice. In these compounds the magnetic ordering temperature TN is much below the superconducting transition temperature Tc. The discovery of superconducting borocarbides RT2B2C with R = Sc, Y, La, Th, Dy, Ho, Er, Tm or Lu and T = Ni, Ru, Pd or Pt (where not all of these combinations of R and T result in superconductivity) has reanimated the research on the coexistence of superconductivity and magnetic order. Most of these borocarbides crystallize in the tetragonal LuNi2B2C type structure which is an interstitial modification of the ThCr2Si2 type. Contrary to the behaviour of Cu in the cuprates Ni does not carry a magnetic moment in the borocarbides. Various types of antiferromagnetic structures on the rare-earth sublattice have been found to coexist with superconductivity in RNi2B2C for R = Tm, Er, Ho and Dy. Particularly of interest is the case of HoNi2B2C for which three different types of antiferromagnetic structures have been observed: (i) a commensurate one with Ho moments aligned ferromagnetically within layers perpendicular to the tetragonal c axis where consecutive layers are aligned in opposite directions, (ii) an incommensurate spiral along the c axis and (iii) an incommensurate a-axis-modulated structure with a modulation vector τ≈(0.55,0,0). This wave vector emerges in various RNi2B2C compounds with magnetic as well as nonmagnetic R elements and is connected with Fermi surface nesting. Both incommensurate magnetization structures have been shown to be related to the near-reentrant behaviour observed in HoNi2B2C whereas the commensurate structure coexists well with the superconducting state in this compound. The variation of TN and Tc with the de Gennes factor can roughly be drawn on straight lines from Lu to Gd and from Lu to Tb, respectively, with the exception of Yb. Consequently, Tc>TN holds for Tm, Er, Ho and Tc<TN for Dy. However, the study of various pseudoquaternary (R,R')Ni2B2C compounds has shown that this so-called de Gennes scaling is not universal for the borocarbides and it breaks down in some cases, which is attributed to effects of details of the electron structure, crystalline electric fields, the difference in the R and R' ionic radii or to the effect of nonmagnetic impurities in an antiferromagnetic superconductor. In an external magnetic field some of the RNi2B2C compounds show metamagnetic transitions combined with a large negative magnetoresistance. A small net magnetization found at low temperatures and zero magnetic field in TbNi2B2C has been interpreted as weak ferromagnetism of Dzyaloshinsky-Moriya type. A similar phenomenon observed for superconducting ErNi2B2C is still under discussion.

https://iopscience.iop.org/article/10.1088/0034-4885/64/8/202/pdf

Interaction of superconductivity and magnetism in borocarbide superconductors

Since 1986, most of the interest in superconductivity became focused on high-T c cuprates. The discovery of the superconducting quaternary borocarbide system Y–Ni–B–C with T c as high as ∼12 K inspired research into intermetallic superconductors (IMS) once again. Several reasons can be attributed to this revival of interest in IMS: (i) In the tetragonal quaternary magnetic superconductors RNi2B2C, superconductivity and magnetism occur with T c and T N ∼ 10 K, thereby allowing studies of exotic phenomena associated with, and arising from, the interplay of superconductivity and magnetism. (ii) High T N's and a variety of commensurate and incommensurate magnetic structures in RNi2B2C (Fermi surface nesting playing a central role) strongly suggest that R-spins are coupled via the RKKY-exchange interaction. Hence, unlike in most other magnetic superconductors known so far, conduction electrons take part in superconductivity and magnetism. (iii) Quaternary borocarbides open up new pathways to try and synthesize...

Superconductivity and magnetism and their interplay in quaternary borocarbides RNi2B2C

On the complexity of spinels: Magnetic, electronic, and polar ground states

${}^{7}\mathrm{Li}$ and ${}^{51}\mathrm{V}$ NMR Knight shift, linewidth, spin-lattice, and spin-spin relaxation rate data are reported as a function of temperature $T$ (1.5--800 K and 74--575 K for Li and V nuclei, respectively) in the heavy fermion compound ${\mathrm{LiV}}_{2}{\mathrm{O}}_{4}$. The ${}^{7}\mathrm{Li}$ Knight shift $K$ exhibits a broad maximum at about 25 K and is nearly constant below 4.2 K, as is the linewidth. The ${}^{51}\mathrm{V}$ Knight shift is negative at 575 K and it decreases strongly as the temperature is lowered. Above \ensuremath{\sim}80 K, both ${}^{7}\mathrm{Li}$ and ${}^{51}\mathrm{V}$ Knight shifts are proportional to the susceptibility and from the slope the hyperfine coupling constant can be obtained for both nuclei. The ${}^{7}\mathrm{Li}$ spin-lattice relaxation rate increases as the temperature is lowered below room temperature reaching a maximum at about 50 K. Below 5 K the relaxation rate decreases linearly with $T$ as for normal metals but with a very high value of the Korringa product ${(T}_{1}{T)}^{\ensuremath{-}1}$. Even so, the ${}^{7}\mathrm{Li}$ Korringa ratio ${K}^{2}{T}_{1}{T/S}_{\mathrm{Li}}$ below $\ensuremath{\sim}$ 10 K is on the order of unity as in conventional metals, thus indicating heavy Fermi liquid behavior. From an analysis of the ${}^{7}\mathrm{Li}$ and ${}^{51}\mathrm{V}$ relaxation rate data we infer an approximate square root temperature dependence of the V local moment spin-relaxation rate at high temperatures $(T\ensuremath{\gtrsim}50$ K). A simplified analysis of the data yields a value of 17 meV for the magnitude of the exchange coupling constant between the local moments and the conduction electrons and 2.3 meV for that between neighboring local moments. Quantitative discrepancies in the ${}^{7}\mathrm{Li}$ and ${}^{51}\mathrm{V}$ relaxation rates at low $T$ from what is expected for a uniform heavy Fermi liquid indicate that effects such as Kondo effect screening of the V local moment and/or dynamical vanadium ${t}_{2g}$ orbital occupation correlation/fluctuation effects may be present in our system.

7 Li and 51 V NMR study of the heavy-fermion compound LiV 2 O 4

Data of ${}^{11}\mathrm{B}$ NMR in a single crystal of ${\mathrm{YbNi}}_{2}{\mathrm{B}}_{2}\mathrm{C}$ are reported in the temperature range 1.7--300 K and for two orientations of the external magnetic field with respect to the tetragonal $c$ axis of the crystal. For $Tg50\mathrm{K}$ both the Knight shift $K$ and the nuclear spin-lattice relaxation rate ${T}_{1}^{\ensuremath{-}1}$ can be accounted for by the presence of localized $4f$ moments at the ${\mathrm{Yb}}^{3+}$ site which polarize the conduction electrons via the Ruderman-Kittel-Kasuya-Yosida mechanism. On the other hand, at low temperatures, $Tl5\mathrm{K},$ the relaxation rate ${T}_{1}^{\ensuremath{-}1}$ obeys a Korringa-like law with a constant value of ${T}_{1}T$ typical of a normal metal with high density of states at the Fermi level and no localized moments.

E. Lee

Papers

7 Li and 51 V NMR study of the heavy-fermion compound LiV 2 O 4

{sup 11}B NMR in YbNi{sub 2}B{sub 2}C single crystals: Crossover from localized moments to Fermi-liquid behavior