# Synopsis: Magnetic order is no match for the lattice

#### Interplay between magnetic properties and Fermi surface nesting in iron pnictides

A. N. Yaresko, G.-Q. Liu, V. N. Antonov, and O. K. Andersen

Published April 16, 2009

Understanding magnetic order in the two major families of iron-based superconductors with $\text{FeAs}$ layers—namely, electron-doped ${\text{LaFeAsO}}_{1-x}{\text{F}}_{x}$ and hole-doped ${\text{Ba}}_{1-2y}{\text{K}}_{2y}{\text{Fe}}_{2}{\text{As}}_{2}$—is important because magnetism seems to be intimately tied with superconductivity. At about $150\phantom{\rule{0.333em}{0ex}}\text{K}$, the undoped compounds ($x$ and $y$ = 0) acquire a ferromagnetic stripe ordering along the shorter axis of the square $\text{Fe}$ sublattice, while displaying antiferromagnetic ordering along the longer axis and between the $\text{Fe}$ layers. Doping with enough carriers suppresses the magnetic order and induces superconductivity in both compounds, though in ${\text{Ba}}_{1-2y}{\text{K}}_{2y}{\text{Fe}}_{2}{\text{As}}_{2}$ magnetism and superconductivity coexist for $0.10.

Although density-functional calculations overestimate the value of the $\text{Fe}$ moment, they generally reproduce the observed magnetic structure for undoped pnictides. In an article appearing in Physical Review B, Alexander Yaresko, Guo-Qiang Liu, Viktor Antonov, and Ole Krogh Andersen at the Max-Planck Institute in Stuttgart, Germany, present comprehensive calculations on experimentally observed crystal structures of ${\text{LaFeAsO}}_{1-x}{\text{F}}_{x}$ and ${\text{Ba}}_{1-2y}{\text{K}}_{2y}{\text{Fe}}_{2}{\text{As}}_{2}$ to determine the magnetic behavior as a function of doping. They find that electron doping above $x>0.1$ in ${\text{LaFeAsO}}_{1-x}{\text{F}}_{x}$ destabilizes the stripe magnetic order and leads to an incommensurate spin spiral order, whereas hole-doping in ${\text{Ba}}_{1-2y}{\text{K}}_{2y}{\text{Fe}}_{2}{\text{As}}_{2}$ leaves the magnetic stripe order intact up to $y=0.25$. The authors find that in both compounds, the classical Heisenberg model with nearest and next-nearest neighbor spin interactions is inadequate to describe the magnetic order and may require additional terms to accurately compute the energy. – Sarma Kancharla