Drude weight anisotropy in the doped iron pnictides: the primary role of orbital weight redistribution along the reconstructed Fermi surfaces
Extracted source text · arXiv 1810.06842
%
% y scale in anis plots ..
% ref sequence
%
%
%
\documentclass[prb,twocolumn,showpacs]{revtex4-1}
\usepackage{graphicx}
\usepackage{times}
\usepackage{color}
\pagestyle{empty}
\input pdfcolor.tex
\usepackage{graphicx}
\usepackage{epsfig,color}
\usepackage{amsfonts,amssymb}
\usepackage{amsmath}
\usepackage{color}
\usepackage{epstopdf}
\definecolor{refcolor}{rgb}{1.0,0.0,0.0}
%\usepackage{siunitx}
%\newcommand{\angstrom}{\mbox{\normalfont\AA}}
%\newcommand{\angstrom}{\textup{\AA}}
%\newcommand{\angstrom}{\text{\normalfont\AA}}
\newcommand{\be}{\begin{equation}}
\newcommand{\ee}{\end{equation}}
\newcommand{\bea}{\begin{eqnarray}}
\newcommand{\eea}{\end{eqnarray}}
\newcommand{\ba}{\begin{array}}
\newcommand{\ea}{\end{array}}
\newcommand{\phr}[1]{Phys.~Rev. {\bf #1}}
\newcommand{\phrl}[1]{Phys.~Rev.~Lett., {\bf #1}}
\newcommand{\phrb}[1]{Phys.~Rev.~B, {\bf #1}}
\newcommand{\cmat}[1]{arXiv:#1}
\newcommand{\jpsj}[1]{J.~Phys.~Soc.~Jpn.{\bf #1}}
\newcommand{\bib}{\bibitem}
\newcommand{\lb}{\left[}
\newcommand{\rb}{\right]}
\newcommand{\lp}{\left(}
\newcommand{\rp}{\right)}
\newcommand{\lf}{\left\{}
\newcommand{\rf}{\right\}}
\newcommand{\la}{\left <}
\newcommand{\ra}{\right >}
\newcommand{\G}{{\cal G}}
\renewcommand{\r}{{\bf r}}
\renewcommand{\j}{{\bf j}}
\newcommand{\p}{{\bf p}}
\newcommand{\q}{{\bf q}}
\renewcommand{\k}{{\bf k}}
\newcommand{\Q}{{\bf Q}}
\newcommand{\R}{{\bf R}}
\newcommand{\J}{{\bf J}}
\newcommand{\E}{\bf E}
\newcommand{\ua}{\uparrow}
\newcommand{\da}{\downarrow}
\newcommand{\F}{{\cal F}}
\begin{document}
\title{Drude weight anisotropy in the doped iron pnictides:
the primary role of\\
orbital weight redistribution along
the reconstructed Fermi surfaces}
\date{\today}
\author{Dheeraj Kumar Singh$^{1,2,3}$ and Pinaki Majumdar$^{3}$}
\affiliation{$^1$Department of Physics, POSTECH, Pohang, Gyeongbuk 790-784, Korea}
\affiliation{$^2$Asia Pacific Center for Theoretical Physics, Pohang, Gyeongbuk 790-784, Korea}
\affiliation{$^3$Harish-Chandra Research Institute, HBNI,
Chhatnag Road, Jhunsi, Allahabad 211019, India}
\begin{abstract}
We explore the anisotropy in low frequency conductivity as a function of
hole doping in multiorbital models of pnictides by analyzing the Drude
weight in the $x$ and $y$ directions of a ($\pi$, 0) spin-density wave
state. A reduction in the conductivity anisotropy with increased hole
doping, and subsequent sign reversal, both observed in experiments, is
found to be a common trend in these models. This behavior arises due to
the interplay of low energy orbitally-resolved density of states and the
geometrical features of the Fermi surface. An understanding of anisotropy
in the electron doped regime, however, would require additional ingredients.
\end{abstract}
\pacs{74.70.Xa,74.25.F-}
\maketitle
\section{Introduction}
The anisotropic electronic properties of the iron pnictides,
in both the collinear spin-density wave (SDW) state and
the nematic state, have attracted considerable attention.
Evidence of anisotropy is visible
in angle-resolved photoemission spectroscopy~\cite{yi,shimojima1} (ARPES), nuclear magnetic resonance
\cite{fu} (NMR), scanning tunneling microscopy~\cite{chuang},
and transport measurements~\cite{chu,tanatar,blomberg}. According to the
ARPES measurements, a significant energy splitting between $d_{xz}$ and
$d_{yz}$ orbitals is observed below the tetragonal-to-orthorhombic
transition, which may either precede~\cite{nandi}
the SDW transition
or occur simultaneously~\cite{rotter}, and states at the Fermi level
have dominantly $d_{xz}$ character~\cite{shimojima2}.
In-plane anisotropy is observed also in the optical
spectra~\cite{nakajima} up to a photonic energy of 2eV.
The easiest to measure is transport anisotropy.
Detwinned crystals show
larger conductivity, surprisingly, in the antiferromagnetic
direction compared to the ferromagnetic
direction~\cite{chu, tanatar, ying}. The ratio of resistivities
can be as large as $1.5$ in
the undoped parent compound. Similar conductivity
anisotropy has also been observed in the nematic
state~\cite{ying,blomberg}.
Theories have tried to address the anisotropy.
In the collinear SDW state of the
undoped compound
multiorbital models have been studied
employing (i)~a
combination~\cite{yin}
of the local density approximation (LDA)
and dynamical mean field theory (DMFT), as well as
(ii)~mean-field methods~\cite{sugimoto1,zhang, valenzuela}.
According to these studies, a significant difference in the
density of states (DOS) of $d_{xz}$ and $d_{yz}$
orbitals at the Fermi level may be responsible for
the counter-intuitive anisotropy in the conductivity. Suppression
of the DOS of $d_{yz}$ orbital \cite{daghofer1}
at the Fermi level, compared to that of $d_{xz}$,
in the ($\pi, 0$)-SDW state results in a larger
conductivity along the $x$-direction since a major
contribution to the Drude weight comes from the interorbital
$d_{xz}$ to $d_{xy}$ hopping~\cite{zhang}.
Conductivity in the $y$-direction,
facilitated primarily by hopping between
$d_{yz}$ and $d_{xy}$, is smaller. Another
study~\cite{valenzuela} emphasizes the role of Fermi
surface `reconstruction'.
What about doped systems?
Experiments reveal that the conductivity anisotropy
decreases and changes sign as hole doping increases~\cite{blomberg} while it grows on electron doping~\cite{chu, ying}.
In order to describe the doping dependence the
role of doping induced disorder has been invoked.
It was shown that disorder with isotropic impurity potential
can lead to elongated magnetic droplets in the nematic
state, with anisotropy increasing as one approaches
the SDW transition~\cite{gastiasoro,timm,wang,dheeraj}. Deep inside the SDW state,
the disorder has been suggested to be
important particularly close to zero doping,
where the Dirac
point is situated near the Fermi level ~\cite{sugimoto}.
The role of both magnetic and non-magnetic
disorder has indeed been explored experimentally.
In the tetragonal state
subjected to external strain, electron-doped
samples having different
residual resistivity
show similar resistivity anisotropies~\cite{kuo}. Studies
on electron- and hole-doped samples, and the presence of
magnetic impurities, drew a similar conclusion.
This suggests that while disorder controls the
magnitude of the resistivity, the anisotropy is
more a consequence of the bandstructure
in the doped system~\cite{kobyashi}.
To conclusively settle this issue, we
examined the Drude weight both
in (a)~the
($\pi, 0$)-SDW state of the three-orbital model proposed by
Daghofer \textit{et al.}~\cite{daghofer} and (b)~the more
sophisticated five-orbital models of Graser \textit{et al.}~
\cite{graser} and Ikeda \textit{et al.}~\cite{ikeda}.
We find the following:
(i)~Despite considerable differences
% in Fermi-surface structure and the orbital resolved DOS
all these models exhibit the robust common
feature that conductivity anisotropy decreases with
hole doping, and subsequently reverses it's sign
- all consistent with experiments.
(ii)~We explain this in terms of the
orbitally resolved weight along the Fermi surface, the
hopping anisotropy,
and shape of the Fermi surface.
(iii)~In the electron-doped region,
the ratio of Drude weights
along ferromagnetic and antiferromagnetic direction approaches
$\sim 1$ asymptotically. Thus,
experimental trend observed on electron doping
is not captured within this approach, suggesting additional
effects at play.
\section{Theory}
We consider a multiorbital model Hamiltonian with the kinetic
part defined as
\begin{equation}
\mathcal{H}_0 = -\sum_{{\bf i},{\bf j}}^{\mu,\nu, \sigma}
t_{{\bf i}{\bf j}}^{\mu\nu} d_{{\bf i}\mu\sigma}^\dagger d_{{\bf j}\nu\sigma}
\end{equation}
where the operator $d_{{\bf i} \mu \sigma}^\dagger$ ($d_{{\bf i}
\mu \sigma}$) creates (destroys) an electron in the $\mu$-th orbital
of site ${\bf i}$ with spin $\sigma$, and $t_{{\bf i}{\bf j}}^{\mu\nu}$
are the hopping elements from orbital $\mu$ and $\nu$ of sites ${\bf i}$
and ${\bf j}$, respectively. The orbitals $\mu$ and $\nu$ belong to the set
of five $d$-orbitals $d_{xz}$, $d_{yz}$, $d_{xy}$, $d_{x^2-y^2}$, and
$d_{3z^2-r^2}$ depending on the model.
The interaction part of the Hamiltonian is given by
\begin{eqnarray}
\mathcal{H}_{int} &=& U \sum_{{\bf i},\mu} n_{{\bf i}\mu
\uparrow} n_{{\bf i}\mu \downarrow} + (U' - \frac{J}{2})
\sum_{{\bf i}, \mu<\nu, \sigma \sigma^{\prime}} n_{{\bf i}
\mu \sigma} n_{{\bf i} \nu \sigma^{\prime}}
\cr
&-& 2 J \sum_{\bf i}^{\mu<\nu} {\bf{S_{{\bf i} \mu}}}
\cdot {\bf{S_{{\bf i} \nu}}}
+ J \sum_{\bf i,\sigma}^{\mu<\nu}
d_{{\bf i} \mu \sigma}^{\dagger}d_{{\bf i} \mu
\bar{\sigma}}^{\dagger}d_{{\bf i} \nu \bar{\sigma}}
d_{{\bf i} \nu \sigma} \nonumber
\label{int}
\end{eqnarray}
which includes the intraorbital (interorbital) Coulomb
interaction term as the first (second) term.
The third term describes the Hund’s coupling, and the
fourth term represents the pair hopping energy. Rotation-invariant
interaction is ensured provided that $U^{\prime}$ = $U$ - $2J$.
The Hamiltonian after mean-field decoupling of the interaction
term in the $(\pi, 0)$-SDW state is given in the two sublattice
basis by
\begin{equation}
\mathcal{H}_{mf} = \sum_{\bf k \sigma} \Psi^{\dagger}_{{\bf k}
\sigma} (T_{{\bf k} \sigma} + M_{{\bf k} \sigma})
\Psi_{{\bf k} \sigma},
\end{equation}
where matrix elements $T_{{\bf k} \sigma}^{ll^{\prime}}$ and
$M^{ll^{\prime}}_{{\bf k} \sigma} = -s \sigma \Delta_{l^{\prime}}
\delta^{ll^{\prime}} + \frac{5J - U}{2} n_{l^{\prime}} \delta^{ll^{\prime}}$
are due to the kinetic and interaction parts, respectively.
$l$, $l^{\prime}$ $\in$ $s \otimes \mu $ with $s$ and
$\mu$ denoting a sublattice and an orbital, respectively.
$s$ and $\sigma$ in front of $\Delta_{ll^{\prime}} \delta^{ll^{\prime}}$
take value 1 (-1) for A (B) sublattice and $\uparrow$-spin ($\downarrow$-spin),
respectively. The electron field operator and the exchange fields are
defined as $\Psi^{\dagger}_{k} = (d^{\dagger}_{A{\bf k}1
\uparrow},d^{\dagger}_{A{\bf k}2 \uparrow},. .. ,d^{\dagger}_
{B{\bf k}1 \uparrow},d^{\dagger}_{B{\bf k}2 \uparrow},...)$ and $2\Delta_{l} =
Um_{l} + J \sum_{l \ne l^{\prime}}m_{l^{\prime}}$, respectively.
Charge density $n_l$ and magnetization
$m_l$ in each of the orbitals are determined
self-consistently by diagonalizing the Hamiltonian.
% -------------------------------------------------------------------
\begin{figure}[b]
\begin{center}
\psfig{figure=fig1.pdf,width=6.2cm,height=8.0cm,angle=-90}
\end{center}
\vspace{-4mm}
\caption{Doping dependence of the (a) magnetic moment,
(b) electronic density, and (c) DOS at the
Fermi level for each of the orbitals in the ($\pi, 0$)-SDW
state of the three-orbital model of Daghofer \textit{et al}..
(d) Drude weights $D_x$ and $D_y$ in the $x$-direction and $y$-direction,
respectively. For convenience, the dimensionless
quantity $R_{an} = D_x$/$D_y -1$, which provides a
measure of the anisotropy, is also plotted.}
\label{daghofer1}
\end{figure}
% -------------------------------------------------------------------
\begin{figure}[t]
\begin{center}
\psfig{figure=fig2.pdf,width=80mm,height=5.6cm,angle=0}
\end{center}
\vspace{-4mm}
\caption{Fermi surfaces in the ($\pi, 0$)-SDW state of
the model of Daghofer \textit{et. al.} for hole dopings
(a) $x_h = 0.24$, (b) $x_h = 0.0$, and electron doping
(c) $x_e = 0.1$ with the predominant orbital densities,
where $U = 1.22$eV.}
\label{daghofer2}
\end{figure}
% -------------------------------------------------------------------
The optical conductivity along $\alpha = x,y$ given by
\cite{dagotto,valenzuela1}
\begin{eqnarray} \sigma_{\mathcal{\alpha}} &=& D_{\alpha} \delta{(\omega)}+
\frac {1}{N} \sum_{{\bf k}, n \ne n^{\prime}}
\frac{ |j^{\alpha}_{nn^{\prime}}({\bf k})|^2}{E_{n^\prime {\bf k}} -
E_{n {\bf k}}} \nonumber\\
&\times&\theta (-E_{n^\prime {\bf k}})\theta (E_{n {\bf k}})
\delta(\omega-E_{n^\prime {\bf k}}+E_{n {\bf k}})
\end{eqnarray}
with Drude weight
\begin{eqnarray} \frac {D_{\alpha}}{2 \pi} &=& \frac{1}{2N}
\sum_{\bf k} T^{\alpha}_{nn} ({\bf k}) \theta (-E_{n {\bf k}}) -
\frac {1}{N} \sum_{{\bf k}, n \ne n^{\prime}}
\frac{ |j^{\alpha}_{nn^{\prime}}({\bf k})|^2}{E_{n^\prime {\bf k}} -
E_{n {\bf k}}} \nonumber\\
&\times&\theta (-E_{n^\prime {\bf k}})\theta (E_{n {\bf k}}).
\end{eqnarray}
$\theta$ is the step function, $E_{n{\bf k}}$ is the single
particle energy in the ordered state with $n$ as a band index, and
\begin{eqnarray} T^{\alpha}_{nn} = \sum_{\mu \nu} T^{\alpha;\mu \nu}_{nn}
= \sum_{\mu \nu} \frac {\partial^2
t_{\mu \nu} ({\bf k})}{\partial k^2_x}
a^*_{{\bf k} \mu n} a_{{\bf k}\nu n}
\nonumber\\
j^{\mathcal{\alpha}}_{nn^{\prime}} = -\sum_{\mu \nu} j^{\alpha;\mu \nu}_{nn^{\prime}}
= -\sum_{\mu \nu} \frac {\partial
t_{\mu \nu} ({\bf k})}{\partial k_x}
a^*_{{\bf k} \mu n} a_{{\bf k}\nu n},
\end{eqnarray}
$a_{{\bf k} \mu n}$ is the unitary matrix element
between the orbital $\mu$ and band $n$ in ($\pi, 0$)-SDW state. To
investigate the role of electronic reconstruction in the conductivity anisotropy,
electron scattering by impurities or spin fluctuations can be assumed to be isotropic
for simplicity, which may not be the case in actuality~\cite{kemper}.
However, absence of anisotropy
due to impurity scattering is also supported by experiments. Then,
$\delta$ function is approximated by Lorentzian with a small but non vanishing
broadening parameter, which is same in both directions.
In this work we focus on the anisotropy in $\omega \rightarrow 0$ limit. To gain
a better understanding of the origin of
anisotropy, it is useful to define the components of
Drude weight as~\cite{zhang}
\begin{eqnarray}
\frac {D^{\alpha\beta}_{l}}{2 \pi} &=& \frac{1}{2N}
\sum_{\bf k} T^{l;\alpha \beta}_{nn} ({\bf k}) \theta
(-E_{n {\bf k}}) \nonumber\\
&-& \frac {1}{N} \sum_{{\bf k}, n \ne n^{\prime}} Re
\frac{ j^{x(y);\alpha {\beta}^* }_{nn^{\prime}}({\bf k})
j^{x(y)}_{nn^{\prime}}({\bf k})}{E_{n^\prime {\bf k}}
- E_{n {\bf k}}}\theta (-E_{n^\prime {\bf k}})\theta (E_{n {\bf k}}). \nonumber\\
\end{eqnarray}
In the following, Hund's coupling $J$ is set to be
$U$/4 unless stated otherwise~\cite{schafgans},
whereas intraorbital Coulomb interaction is chosen so
that the total magnetic moment per site is nearly unity
for the undoped case.
\section{Results}
\subsection{Three band model}
Fig.~\ref{daghofer1}(a) and (b) show the mean-field magnetic
order parameter and the charge density of each of the orbitals in the
($\pi$, 0)-SDW state for $U = 1.22$eV. Total magnetic moment
increases continuously on moving from the electron- to hole-overdoped
region. The difference in magnetization $m_{xz}-m_{yz}$ of two
orbitals $d_{xz}$ and $d_{yz}$ reverses it's sign slightly
below band-filling $n = 4.0$ which
corresponds to the undoped case. A similar but opposite trend is
observed for the ferro-orbital order $n_{xz}-n_{yz}$.
As can also be seen from Fig.~\ref{daghofer1}(d), sign reversal
of the orbital order is not related to the reversal of anisotropy
$R_{an} = D_x$/$D_y -1$, where $D_x$ and $D_y$ are Drude weights along
$x$- and $y$-directions, respectively. Latter occurs near hole
doping $x_h \approx 0.22$ in agreement with the experiments. Note that SDW state
for $x_h \approx 0.22$ may be stabilized only at higher temperature \cite{avci}.
$R_{an}$ exhibits a maximum near zero doping and it's
behavior in the electron doped region where it decreases is in
contrast with the experiments.
Note that $d_{xz}$ DOS remains dominant at
the Fermi level for the entire hole-doping regime as shown
in Fig.~\ref{daghofer1}(c). Thus, suppression of $d_{yz}$ DOS
at Fermi level alone is not sufficient to explain the sign
reversal. Apart from $d_{xz}$, $d_{xy}$ orbital also plays a
very important role in the anisotropy. As seen from Table I,
interorbital $D^{13}$ and intraorbital $D^{33}$ contributions
to the Drude weight are significant since
the intraorbital hopping parameters $t_{13}$ and $t_{33}$ are
considerably larger than $t_{11}$ or $t_{22}$, where superscript
1, 2, and 3 have been used for the orbitals $d_{xz}$, $d_{yz}$,
and $d_{xy}$, respectively. $D^{13}_{x}$ remains larger than
$D^{13}_{y}$ despite a reduction in it's value for the entire
hole doping regime considered. At the same time, $D^{33}_{y}$
becomes larger than $D^{33}_{x}$ and contributes, apart from
small contributions from $D^{11}_{y}$ and $D^{22}_{y}$,
substantially to the sign-reversal.
\begin{table}[t]
\caption{Components of Drude weight in the three orbital
model for $U$ = 1.22eV.
$\alpha$ and $\beta$ are orbital
indices having values 1, 2, and 3 corresponding to the
$d_{xz}$, $d_{yz}$, and $d_{xy}$ orbitals, respectively.
Subscripts 1, 2 and 3 of the components correspond
to $x_h$ = 0.0, $x_h = 0.24$ and $x_e = 0.1$
respectively.} % title of Table
\centering % used for centering table
\setlength{\tabcolsep}{3pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ & $11$ & $12$ & $13$ & $22$ & $23$ & $33$ \\ [0.5ex] % inserts table
%heading
%\hline % inserts single horizontal line
\hline
$D^{\alpha \beta}_{x;1}$ &$0.010$ & $0.000$ & $0.098$ & $-0.001$ & $0.006$ & $0.067$ \\
\hline
$D^{\alpha \beta}_{y;1}$ &$0.016$ & $0.000$ & $-0.010$ & $0.001$ & $0.007$ & $0.070$ \\
\hline
$D^{\alpha \beta}_{x;2}$ &$0.002$ & $0.000$ & $0.045$ & $0.000$ & $0.017$ & $0.017$ \\
\hline
$D^{\alpha \beta}_{y;2}$ &$0.011$ & $0.000$ & $0.019$ & $0.012$ & $0.012$ & $0.041$ \\
\hline
$D^{\alpha \beta}_{x;3}$ &$0.002$ &$ 0.001$ &$ 0.117$ &$ -0.001$ &$ 0.006$ &$ 0.074$ \\
\hline
$D^{\alpha \beta}_{y;3}$ &$0.009$ &$ 0.000$&$ -0.014$ &$ 0.002$ &$ 0.012$ &$ 0.104$ \\
[1ex] % [1ex] adds vertical space
\hline \hline %inserts single line
\end{tabular}%\setlength{\tabcolsep}{6em}
%\label{table:nonlin} % is used to refer this table in the text
\end{table}
% ---------------------------------------------------------------
\begin{figure}[b]
\begin{center}
\psfig{figure=fig3.pdf,width=6.2cm,height=8cm,angle=-90}
\end{center}
\vspace{-4mm}
\caption{(a) Orbital magnetizations (b) orbital densities
(c) orbital resolved DOS at the Fermi level as a function of doping
in the five-orbital model of Graser \textit{et. al.}. (d)
Drude weight in the $x$-direction, $y$-direction, and the anisotropy. }
\label{graser1}
\end{figure}
% ---------------------------------------------------------------
\begin{figure}[t]
\begin{center}
\psfig{figure=fig4.pdf,width=80mm,height=5.6cm,angle=0}
\end{center}
\vspace{-4mm}
\caption{Fermi surfaces for hole dopings (a) $x_h = 0.1$, (b) $x_h = 0.0$,
and electron doping (c) $x_e = 0.06$ with dominant orbital character in
the model of Graser \textit{et. al}., where $U = 1.44$eV. }
\label{graser2}
\end{figure}
% ---------------------------------------------------------------
To understand the doping dependence
it is essential to take into account the orbital-resolved DOS
across the Fermi surface as well as the geometrical
structure of the Fermi surfaces. Reconstruction of the hole pocket~\cite{daghofer,yi1,lu,liu}
around $\Gamma$ is not seen (Fig.~\ref{daghofer2}(b)) which
is in contrast with the experiments. This happens
primarily because of larger
interaction parameters chosen to yield magnetic
moment $m_{tot} \approx 1$. However,
reconstruction of the hole pockets for
same magnetization can be clearly seen
in the five-orbital model as presented below,
which can disappear again on increasing the interaction~\cite{kovacic}.
The contribution to
$D^{13}_{x(y)}$ comes from those regions of the Fermi surfaces,
which have significant proportions of both the orbitals
$d_{xz}$ and $d_{yz}$. In the undoped case, such regions are near
($\pm 0.24\pi,\pm \pi$) of the electron pockets at
($0, \pm \pi$) and also near the boundary
separating $d_{yz}$ and $d_{xy}$ dominated regions of the hole pockets
at ($0, \pm0.5\pi$) (Fig.\ref{daghofer2}(b)). Moreover,
components of the Fermi velocity along those regions follow the
relation $v^{f}_x$ $>$ $v^{f}_y$. For this reason,
$D^{13}_{x}$ is significantly larger than $D^{13}_{y}$, and thus
primarily responsible for the Drude weight
anisotropy. Other important contributor $D^{33}_{x(y)}$
do not differ substantially in the $x$- and $y$-directions for zero doping.
This is because nearly equal regions of the electron
pocket at ($0, \pm \pi$) and the hole pockets
at ($\pm 0.5\pi, 0$) dominated by the $d_{xy}$
orbital have Fermi velocity components mainly along
$y$- and $x$-direction, respectively. The
hole pocket around ($0, 0$) is nearly circular and
predominantly of $d_{xz}$ character, thus does
not contribute to the anisotropy.
On electron doping, $d_{xy}$ dominated sections of
the electron pockets and
hole pockets get shrunk and enlarged, respectively,
(Fig.~\ref{daghofer2}(c)). This leads to $D^{33}_{y}
> D^{33}_{x}$ causing a drop in $R_{an}$. Whereas
contribution due to $D^{13}_{(x)y}$ becomes smaller
(Fig.~\ref{daghofer2}(a)) on hole doping. Finally, the regions
of square-like Fermi surface with dominant $d_{xz}$ and $d_{xy}$ orbitals
have $v^f_y > v^f_x$ in large region leading to $D^{33}_{y} > D^{33}_{x}$,
which results in the reversal of Drude-weight anisotropy.
\subsection{Five band model of Graser et al.}
In the five-orbital model of Graser \textit{et. al.},
orbital magnetization shows a quick rise on hole doping
so that $\sum_{\mu} m_{\mu} \ge 1.6$ as soon as doping
reaches 0.1, while it decreases rapidly on electron
doping and vanishes near $n \approx 6.12$
(Fig.~\ref{graser1}(a)). $n \approx 6.0$
corresponds to the undoped case $x = 0$.
There is a significant difference in the
magnetizations as well as in the charge densities of
$d_{xz}$ and $d_{yz}$ orbitals. The ferro-orbital
order $n_{xz} - n_{yz}$ has a sign opposite to that of
$m_{xz} - m_{yz}$ in the entire region of
electron density considered here (Fig.~\ref{graser1}(b)).
Unlike the three-orbital model, there is no sign reversal of these
parameters.
Before discussing the Drude weight anisotropy
in this model, it is interesting to look first
at the orbitally-resolved relative
DOS at the Fermi level. As seen from
Fig.~\ref{graser1}(c), the DOS of $d_{xy}$ orbital is
the largest for the undoped case in contrast
with the three-orbital model, which decreases upon doping either
holes or electrons. On the other hand, DOSs
of $d_{xz}$ and $d_{yz}$ orbitals, though small and nearly
equal for zero doping, increase on doping either electrons or holes.
\begin{table}[t]
\caption{Elements of Drude weight in the
model of Graser \textit{et. al.} for $U = 1.44$eV. Here,
orbital indices having values 1, 2, 3, 4,
and 5, which correspond to the $d_{xz}$, $d_{yz}$,
$d_{x^2-y^2}$, $d_{xy}$, and $d_{3z^2-r^2}$ orbitals, respectively.
Subscripts 1, 2 and 3
correspond to $x_h$ = 0.0, $x_h = 0.1$ and $x_e = 0.06$, respectively} % title of Table
\centering % used for centering table
\setlength{\tabcolsep}{2.5pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ & $11$ & $14$ & $23$ & $24$ & $35$ & $44$ & $45$ \\ [0.5ex] % inserts table
%heading
%\hline % inserts single horizontal line
\hline
$D^{\alpha \beta}_{x;1}$ & $0.017$ & $0.037$ & $0.002$ & $0.000$ & $-0.001$ & $0.019$ & $0.000$ \\
\hline
$D^{\alpha \beta}_{y;1}$ & $0.010$ & $0.000$ & $0.001$ & $0.004$ & $0.000$ & $0.017$ & $0.000$ \\
\hline
$D^{\alpha \beta}_{x;2}$ & $0.016$ & $0.035$ & $0.000$ & $0.000$ & $0.000$ & $0.024$ & $0.000$ \\
\hline
$D^{\alpha \beta}_{y;2}$ & $0.010$ & $-0.001$ & $0.021$ & $0.043$ & $0.008$ & $0.003$ & $0.019$ \\
\hline
$D^{\alpha \beta}_{x;3}$ & $ 0.030$ & $ 0.085$ & $ -0.002$ & $ 0.000$ & $ 0.000$ & $ 0.010$ & $ 0.000$\\
\hline
$D^{\alpha \beta}_{y;3}$ & $0.007$ & $ 0.003$ & $ 0.000$ & $ 0.021$ & $ 0.000$ & $ 0.069$ & $ 0.000$\\
[1ex] % [1ex] adds vertical space
%not to be included 11,13,22,25,33,34,55
\hline \hline%inserts single line
\end{tabular}
\end{table}
% -------------------------------------------------------------
\begin{figure}[b]
\begin{center}
\psfig{figure=fig5.pdf,width=6.2cm,height=8cm,angle=-90}
\end{center}
\vspace{-4mm}
\caption{(a) Magnetization, (b) charge density (c)
DOS at the Fermi level for each of the orbitals.
(d) Drude weights in $x$- and $y$-directions, and
their ratios in the model of Ikeda \textit{et. al}..}
\label{ikeda1}
\end{figure}
% -------------------------------------------------------------
% -------------------------------------------------------------
\begin{figure}[t]
\begin{center}
\psfig{figure=fig6.pdf,width=80mm,height=5.6cm,angle=0}
\end{center}
\vspace{-4mm}
\caption{Fermi surfaces for hole dopings (a)
$x_h = 0.1$, (b) $x_h = 0.0$, and
(c) $x_e = 0.06$ with color showing leading
orbital character in the five-orbital model of
Ikeda \textit{et. al.}, where $U = 1.12$eV. }
\label{ikeda2}
\end{figure}
% -------------------------------------------------------------
$R_{an}$ is maximum near $x_h \approx 0$, drops rapidly,
and changes sign near $x_h \approx 0.07$ (Fig.~\ref{graser1}(d)).
This drop is significantly faster in comparison
with the three-orbital model. Another important difference
from the previous model is that the Drude
weight is larger in the $x$-direction
despite DOSs of $d_{xz}$ and $d_{yz}$ orbitals being
nearly equal at the Fermi level in the undoped case.
Although the same component $D^{14}$ is responsible
for this anisotropy (superscript 4 here refers to
$d_{xy}$ orbital, Table II), which implies again
a crucial role of the orbitally-resolved DOS distribution along
the Fermi surfaces. The sections of the Fermi
surfaces having significant proportions of both
$d_{xz}$ and $d_{xy}$ orbitals are near ($\pm 0.12\pi, \pm \pi$)
of the electron pockets at ($0, \pm \pi$) and are also near
the $d_{xy}$ dominated regions of the small hole pockets
at ($\pm 0.27 \pi, 0$) (Fig.~\ref{graser2}(b)).
Moreover, $v^f_x > v^f_y$ in these regions,
which is mainly responsible for the anisotropy.
Like the three-orbital model, the hole pocket at (0, 0) is predominantly
of $d_{xz}$ orbital character ($\approx$ 88\%),
thus does not play an important role in the anisotropy.
% -------------------------------------
\begin{figure*}
\begin{center}
\psfig{figure=fig7.pdf,width=160mm,height=4.8cm,angle=0}
\end{center}
\vspace{-4mm}
\caption{(a) ARPES Fermi surfaces obtained for
3.5$\%$ electron-doped CaFe$_2$As$_2$ in
the orthorhombic state with collinear
magnetic order. $k_x$ and $
k_y$ are the reciprocal lattice vectors defined
corresponding to the unit cell consisting of two Fe atoms (Ref.
[34]). High
symmetry directions in the SDW state are
rotated by 45$^{\circ}$. (b) Resistivity anisotropy
$\rho = \rho_{b}/\rho_{a} - 1$ in BaFe$_2$As$_2$
at 0.95$T_{\rm N}$ reproduced from Ref. [7], where $b$ and $a$ are the
ferromagnetic and the antiferromagnetic axes,
respectively. (c) Integrated absolute difference (IAD) from X-ray emission
spectroscopy as a function of hole doping
for BaFe$_2$As$_2$, which provides a qualitative estimate of the behavior of
local moments as a function of hole doping (Ref. [41]).}
\label{exp}
\end{figure*}
% -------------------------------------
The electron pocket grows with a significant modification
in the orbitally-resolved DOS distribution on electron doping
in such a way that $D^{14}_x$ increases further while $D^{14}_y$
remains vanishingly small (Fig.~\ref{graser2}(c)). This is
because the regions of the pocket with nearly equal
proportions of both $d_{xz}$ and $d_{xy}$ orbitals
have $v^f_x > v^f_y$ . However, $D^{44}_y$
becomes significantly larger than $D^{44}_x$
(table II) as $d_{xy}$ orbital dominates now only
those sections which have $v^f_x < v^f_y$.
For this reason, there is a net decline in $R_{an}$.
On the other hand, the electron pocket becomes
smaller and disappear on hole doping $x_h \approx 0.1$. Meanwhile,
the hole pockets at ($\pm 0.27 \pi, 0$) grows
larger in size until they get connected to the
hole pocket at (0, 0) and also extended in the
opposite direction (Fig.~\ref{graser2}(a)).
As seen from table II, anisotropy in $D^{14}$ remains largely
unaffected because of the regions of hole pockets
having $v^f_x > v^f_y$ significant proportions of
$d_{xz}$ and $d_{xy}$ orbitals. Whereas the flat
regions of the hole pocket with $v^f_y > v^f_x$
(vanishingly small $v^f_x$) have significant
proportions of $d_{yz}$ as well as of either $d_{xy}$ or
$d_{x^2-y^2}$ orbitals, which leads to
$D^{23}_y > D^{23}_x$ and $D^{24}_y > D^{24}_x$
(Table II), hence leading to the
sign reversal of $R_{an}$.
\subsection{Five band model of Ikeda et al.}
% ---------------------------------------------
\begin{table}[b]
\caption{Elements of Drude weight in the model
of Ikeda \textit{et. al.} for $U = 1.12$ eV. Here,
orbital indices having values 1, 2, 3, 4, and 5,
which correspond to the $d_{xz}$, $d_{yz}$, $d_{x^2-y^2}$,
$d_{xy}$, and $d_{3z^2-r^2}$ orbitals, respectively.
Subscripts 1, 2 and 3
correspond to $x_h$ = 0.0, $x_h = 0.1$ and $x_e = 0.06$, respectively}
% title of Table
\centering % used for centering table
\setlength{\tabcolsep}{2.5pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ &$11$ & $14$ & $23$ & $24$ & $35$ &$44$ & $45$ \\ [0.5ex] % inserts table
%heading
%\hline % inserts single horizontal line
\hline
$D^{\alpha \beta}_{x;1}$ & $0.026$& $0.038$ & $-0.002$ & $0.000$ & $-0.001$ &$0.010$ & $0.000$ \\
\hline
$D^{\alpha \beta}_{y;1}$ &$0.012$ & $-0.001$ & $0.000$ & $0.014$ & $0.005$ & $0.000$ & $0.000$ \\
\hline
$D^{\alpha \beta}_{x;2}$ & $0.027$& $0.021$ & $0.002$ & $0.000$ & $-0.004$ & $0.006$ & $-0.002$ \\
\hline
$D^{\alpha \beta}_{y;2}$ & $0.013$& $-0.001$ & $0.003$ & $0.024$ & $0.011$ & $0.001$& $0.006$ \\
\hline
$D^{\alpha \beta}_{x;3}$ & $0.025$& $ 0.057$ & $ 0.003$ & $0.001$ & $ 0.002$ &$0.034$ & $ -0.001$\\
\hline
$D^{\alpha \beta}_{y;3}$ & $0.013$ & $0.001$ & $ 0.000$ & $0.032$ & $ 0.005$ &$0.024$ & $ 0.000$\\
[1ex] % [1ex] adds vertical space
\hline \hline%inserts single line
\end{tabular}
\end{table}
% ---------------------------------------------
% ----------------------------------------------------
\begin{table}[b]
\caption{Components of Drude weight in the model
of Daghofer \textit{et. al.} for $\delta = 80$meV.
$\alpha$ and $\beta$ are orbital indices having
values 1, 2, and 3 corresponding
to the $d_{xz}$, $d_{yz}$, and $d_{xy}$ orbitals,
respectively.} % title of Table
\centering % used for centering table
\setlength{\tabcolsep}{3pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ & $11$ & $12$ & $13$ & $22$ & $23$ & $33$ \\ [0.5ex] % inserts table
\hline
$D^{\alpha \beta}_{x}$ &$0.008$ & $0.000$ & $0.136$ & $0.029$ & $-0.003$ & $0.099$ \\
\hline
$D^{\alpha \beta}_{y}$ &$0.013$ & $-0.001$ & $-0.010$ & $0.034$ & $0.135$ & $0.121$ \\
[1ex] % [1ex] adds vertical space
\hline\hline %inserts single line
\end{tabular}%\setlength{\tabcolsep}{6em}
\end{table}
% ----------------------------------------------------
\begin{table}[t]
\caption{Elements of Drude weight in the model of
Graser \textit{et. al.} for $\delta = 80$meV. Here,
orbital indices having values 1, 2, 3, 4, and 5,
which correspond to the
$d_{xz}$, $d_{yz}$, $d_{x^2-y^2}$, $d_{xy}$,
and $d_{3z^2-r^2}$ orbitals, respectively.} % title of Table
\centering % used for centering table
\setlength{\tabcolsep}{2.5pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ & $11$ & $22$ & $33$ & $44$ & $55$ & $13$ & $14$ \\ [0.5ex] % inserts table
%heading
%\hline % inserts single horizontal line
\hline
$D^{\alpha \beta}_{x}$ & $0.043$ & $0.024$ & $0.001$ & $0.063$ & $0.000$ & $0.007$ & $0.100$ \\
\hline
$D^{\alpha \beta}_{y}$ & $0.023$ & $0.045$ & $0.002$ & $0.080$ & $0.000$ & $0.026$ & $0.004$ \\
[1ex] % [1ex] adds vertical space
\hline\hline %inserts single line
\end{tabular}
\end{table}
% ----------------------------------------------------
\begin{table}[t]
\caption{Elements of Drude weight in the model of
Ikeda \textit{et. al.} with parameters and conventions as
in Table V.} % title of Table
\centering % used for centering table
\setlength{\tabcolsep}{2.5pt} % General space between cols (6pt standard)
\renewcommand{\arraystretch}{1} % General space between rows (1 standard)
\begin{tabular}{c|c c c c c c c c c c c c c c c c} % centered columns (4 columns)
\hline\hline %inserts double horizontal lines
$\alpha \beta$ & $11$ & $22$ & $33$ & $44$ & $55$ & $13$ & $14$ \\ [0.5ex] % inserts table
%heading
%\hline % inserts single horizontal line
\hline
$D^{\alpha \beta}_{x}$ & $0.050$ & $0.020$ & $0.004$ & $0.071$ & $0.000$ & $0.007$ & $0.096$ \\
\hline
$D^{\alpha \beta}_{y}$ & $0.020$ & $0.050$ & $0.005$ & $0.092$ & $0.000$ & $0.033$ & $0.004$ \\
[1ex] % [1ex] adds vertical space
\hline\hline %inserts single line
\end{tabular}
\end{table}
% ----------------------------------------------------
The doping dependence of magnetic moment and charge
density of each of the orbitals is similar to that
of the model of Graser \textit{et. al.} (Fig.~\ref{ikeda1}(a)
and (b)). However,
the DOS of $d_{yz}$ orbital is largest at
the Fermi level in the undoped case in
contrast with the previously discussed two
models (Fig.~\ref{ikeda1}(c)).
Behavior of $R_{an}$ as a function of
hole doping is more or less similar to
the model of Graser \textit{et. al.} except that
the peak and zero of the anisotropy are shifted
towards a point near $x_h \approx 0.04$ and $x_h \approx 0.09$,
respectively (Fig.~\ref{ikeda1}(d)).
Another significant difference from the
previous model is the absence of the electron pocket
around ($0, \pm\pi$) for zero doping.
The leading source of anisotropy for
the undoped case is $D^{14}_x > D^{14}_y$, which originates
mainly from the small electron pockets near
($\pm 0.25\pi, 0$), which are extended
along $y$-direction. The reason is similar
to that described in the previous model in terms of the orbitally-resolved
DOS distributions (Fig.~\ref{ikeda2}(b)). The
anisotropy vanishes close to $x_h \approx 0.1$
because of the hole pockets being
nearly circular (Fig.~\ref{ikeda2}(b)). Major components
contributing to the Drude weight within the range
$0 \lesssim x_h \lesssim 0.1$ remains similar to that within the model of
Graser {\it et. al.}. Moreover, their behavior as a function of
doping is also similar. But there are important differences in the electron-doped
region. Here, the suppression of anisotropy does not
result from a large anisotropy in the $D^{44}$ component ($D_y^{44} >> D_x^{44}$). In fact,
$D_x^{44}$ remains greater than $D_y^{44}$ largely because of the difference
in orbital-weight distribution
along the electron pocket near ($\pm\pi, 0$). Thus, unlike $D^{44}$,
$D^{24}$ is mainly responsible for the reduction in anisotropy on electron doping.
\subsection{Anisotropy in the nematic phase}
In order to understand the conductivity
anisotropy in the nematic phase, which is found to
be significant $\sigma_a/\sigma_b \sim 1.2$
in the transport measurements, we consider
an explicit orbital-splitting
term as observed by the ARPES measurements
in addition to the kinetic part of the Hamiltonian (Eq. (1))
\begin{equation}
\mathcal{H}_{OO} = \frac{\delta}{2}\sum_{i}
(d^{\dagger}_{{\bf i} yz \sigma} d_{{\bf i} yz \sigma} -
d^{\dagger}_{{\bf i} xz \sigma} d_{{\bf i} xz \sigma}).
\end{equation}
Consideration of the term has limited motivation
that is to focus on it's impact on the anisotropy instead
of it's origin. Recent works trace the origin to many-body
effects beyond the random-phase approximation level \cite{onari}. Furthermore, an
alternative scenario of conductivity anisotropy, where electrons gets scattered by
the critical-spin fluctuations, has also been proposed~\cite{fernandes}.
Table IV, V and VI show relevant
components of Drude weight obtained for $\delta = 80$meV in
the undoped case. Interestingly, we find anisotropy
in all the three models to be arising due to the
anisotropy in the interorbital component of Drude
weight corresponding to the $d_{xy}$ orbital
indicating that the orbital splitting affects
orbital distribution of $d_{xy}$ the most along
the Fermi surfaces. Anisotropy is
significant $\sim$ 8$\%$ ($D_y > D_x$ ) in each
model but smaller against experimentally
observed $\sim$ 20$\%$. Thus, orbital splitting
induced redistribution of $d_{xy}$ orbital may contribute
significantly to the conductivity anisotropy in the nematic phase.
\section{Discussion}
The ratio of the Drude weights along
the antiferromagnetic and ferromagnetic directions drops from
$D_x/D_y \sim 2$ to a lower value on moving
away from $x \sim 0$. This compares well with
with $\sigma_x/\sigma_y \sim 1.5$ for
$\omega \rightarrow 0$ and $x \sim 0$ in the optical
conductivity measurement where it is found to
be $\sim 1.5$~\cite{nakajima}. A similar value of
conductivity ratio is obtained in the transport
measurements highlighting the important role of
band structure.
Despite the similarities exhibited by these models with respect to
doping dependence of local moment and Drude weight
anisotropy, several features of their Fermi surface
are different. In the models of Daghofer \textit{ et. al.}
and Graser \textit{et. al.}, a very important role in the
anisotropy is played by the electron pockets at ($0, \pm \pi$),
absent in the model of Ikeda \textit{et. al.}. ARPES measurements
do find a weak signature of small electron pockets around X. However, most crucial
features such as the electron pockets placed slightly away from $\Gamma$
along $\Gamma$-X and $\Gamma$-Y of the model of Ikeda \textit{et. al.}
are in very good agreement with the experiments except the ellipse-like
hole pocket around $\Gamma$ which may be either absent or whose
signature may be very weak to be visible \cite{wang}. In other words,
yellow electron pockets and the electron pocket with two different
regions having different leading orbital character
(Fig.~\ref{ikeda2}(c)) show a good correspondence to the blue and
the green electron pockets of experimental Fermi surfaces
(Fig.~\ref{exp}(a)), respectively. Overall, a better agreement
with the experiments regarding the Fermi-surface characteristics
is exhibited within this model. It is also important to note
that the electron pockets are mainly responsible for the
anisotropy even in this model. Moreover, similar electron
pockets occur also in the model of Graser \textit{et. al.} though
their role is not as significant as in the model of Ikeda
\textit{et. al.} because of their smaller sizes.
Drude weight anisotropy in all the three model explored here
exhibit a common trend. Continuous decline and
sign reversal on hole doping compares very well the experiments
while the decline on electron doping is in contrast with the
measurements~\cite{blomberg}. The decrease of anisotropy on hole
doping is not so fast in the three-orbital model
owing largely to a slow modification in the electronic structure,
whereas a sharp decline is exhibited in the five-orbital models.
Magnetic moment grows on moving from the electron-doped towards
the hole-doped region irrespective of the models. This is
not surprising because in each of the models, one
approaches half filling on doping holes, which
suppresses electron movement in the lattice leading to the
departure from metallicity, and finally to the Mottness as
also pointed out by a recent density functional theory (DFT)
combined with slave-spin mean-field approach~\cite{medici}. Evidence in
support of growing magnetic moment on hole doping has been
substantiated recently by the X-ray emission spectroscopy using
integrated absolute difference (IAD)~\cite{lafuerza}.
% ------------------------------------------------
\begin{figure}[t]
\vspace{-0mm}
\begin{center}
\psfig{figure=fig8.pdf,width=85mm,angle=0}
\end{center}
\vspace{-5mm}
\caption{Electronic dispersion in the ($\pi, 0$)
-SDW state along high-symmetry directions
within the three-orbital model of Daghofer \textit{et. al.}. }
\label{disp1}
\end{figure}
% ------------------------------------------------
\begin{figure}[]
\vspace{-0mm}
\begin{center}
\psfig{figure=fig9.pdf,width=85mm,angle=0}
\end{center}
\vspace{-5mm}
\caption{Same as in Fig.\ref{disp1} but for the five-orbital model
of Graser \textit{et. al.}.}
\label{disp2}
\end{figure}
% ------------------------------------------------
\begin{figure}[b]
\vspace{-0mm}
\begin{center}
\psfig{figure=fig10.pdf,width=85mm,angle=0}
\end{center}
\vspace{-5mm}
\caption{Same as in
Fig.\ref{disp2} but for the model of Ikeda \textit{et. al.}.}
\label{disp3}
\end{figure}
% ------------------------------------------------
\section{conclusions}
We have established the role of bandstructure
on the doping dependence of conductivity anisotropy
by examining the Drude weights within the widely used models of pnictides.
In all of them, the anisotropy exhibits a
maximum near zero doping, decreases on doping holes, and then
reverses sign at a certain hole doping, all in qualitative
agreement with the experiments. However, we also find that
the anisotropy decreases on electron doping and vanishes, a feature in
contrast with the experiments, clearly implying
the role of other factors such as
doping induced disorder.
We explain the origin of anisotropy and it's
doping dependence
in terms of the topology and shape of the
Fermi surfaces as well as the distribution of
orbital-resolved DOS along them - not just the dominance of
some particular orbital at the Fermi level. This is
reflected in the similar anisotropy behavior of different models
despite the dominance of different low energy orbitals
in them. The leading contributor
to the anisotropy common for all the models is
interorbital hopping processes involving
$d_{xz}$ and $d_{xy}$ orbitals.
In the nematic phase, a
significant anisotropy is obtained due to the redistribution of
orbital weight which affects $d_{xy}$ orbital the most. Overall, the
$d_{xy}$ orbital plays the most
crucial role in the conductivity anisotropy in the iron pnictides.
\section*{Acknowledgment}
We acknowledge use of the HPC Clusters at HRI.
\section*{Appendix: Electronic dispersion in the ($\pi, 0$)-SDW state}
Electronic dispersions for the different models are presented for the parameters
discussed above. Fig.~\ref{disp1} shows dispersion along high-symmetry
directions for the dopings (a) $x_h = 0.24$, (b) $x_h = 0.0$, and (c) $x_e = 0.1$
in the three-orbital model of Daghofer \textit{et. al.} with $U = 1.22$eV.
Fig.~\ref{disp2}
shows dispersion for the dopings (a) $x_h = 0.1$, (b)
$x_h = 0.0$, and (c) $x_e = 0.06$ in the five-orbital model of Graser \textit{et. al.}
with $U = 1.44$eV.
Fig.~\ref{disp3} shows dispersion for the dopings (a) $x_h = 0.1$, (b) $x_h = 0.0$,
and (c) $x_e = 0.06$ in the five-orbital model of Ikeda \textit{et. al.} with $U = 1.12$eV.
\begin{thebibliography}{0}
\bibitem{yi}
M. Yi, D. Lu, J.-H. Chu, J. G. Analytis, A. P. Sorini, A. F.
Kemper, B. Moritz, S.-K. Mo, R. G. Moore, M. Hashimoto,
W.-S. Lee, Z. Hussain, T. P. Devereaux, I. R. Fisher, and Z.-
X. Shen, Proc. Natl. Acad. Sci. {\bf 108}, 6878 (2011).
\bibitem{shimojima1}
T. Shimojima, T. Sonobe, W. Malaeb, K. Shinada, A. Chainani,
S. Shin, T. Yoshida, S. Ideta, A. Fujimori, H. Kumigashira,
K. Ono, Y. Nakashima, H. Anzai, M. Arita, A. Ino, H. Namatame,
M. Taniguchi, M. Nakajima, S. Uchida, Y. Tomioka, T. Ito, K. Ki-
hou, C. H. Lee, A. Iyo, H. Eisaki, K. Ohgushi, S. Kasahara,
T. Terashima, H. Ikeda, T. Shibauchi, Y. Matsuda, and K. Ishizaka,
Phys. Rev. B {\bf 89}, 045101 (2014).
\bibitem{fu}
M. Fu, D. A. Torchetti, T. Imai, F. L. Ning, J.-Q. Yan, and A. S. Sefat,
\phrl {109}, 247001 (2012).
\bibitem{chuang}
T.-M. Chuang, M. P. Allan, J. Lee, Y. Xie, N. Ni, S. L. Bud’ko, G. S.
Boebinger, P. C. Canfield, and J. C. Davis, Science {\bf 327}, 181 (2010).
\bibitem{chu}
J-H. Chu, J.-H. Chu, J. G. Analytis, K. De Greve, P. L. McMahon,
Z. Islam, Y. Yamamoto, and I. R. Fisher, Science {\bf 329}, 824 (2010).
\bibitem{tanatar}
M. A. Tanatar, E. C. Blomberg, A. Kreyssig, M. G. Kim, N. Ni,
A. Thaler, S. L. Bud'ko, P. C. Canfield, A. I. Goldman, I. I. Mazin,
and R. Prozorov, Phys. Rev. B {\bf 81}, 184508 (2010).
\bibitem{blomberg}
E. C. Blomberg, M. A. Tanatar, R. M. Fernandes, I. I. Mazin, B. Shen,
H.-H. Wen, M. D. Johannes, J. Schmalian, and R. Prozorov, Nat.
Commun. {\bf 4}, 1914 (2013).
\bibitem{nandi} S. Nandi, M. G. Kim, A. Kreyssig, R. M. Fernandes, D. K. Pratt, A. Thaler, N. Ni, S. L. Bud’ko, P. C. Canfield, J. Schmalian, R. J. McQueeney, and A. I. Goldman, \phrl {104}, 057006 (2010).
\bibitem{rotter} M. Rotter, M. Tegel, D. Johrendt, I. Schellenberg, W. Hermes, and R. P\"{o}ttgen, \phrb {78}, 020503R (2008).
\bibitem{shimojima2}
T. Shimojima, K. Ishizaka, Y. Ishida, N. Katayama, K. Ohgushi, T. Kiss, M. Okawa, T. Togashi,
X.-Y. Wang, C.-T. Chen, S. Watanabe, R. Kadota, T. Oguchi, A. Chainani, and S. Shin,
Phys. Rev. Lett. {\bf 104}, 057002 (2010).
\bibitem{nakajima}
M. Nakajima, T. Liang, S. Ishida, Y. Tomioka, K. Kihou, C. H. Lee, A.
Iyo, H. Eisaki, T. Kakeshita, T. Ito, and S. Uchida, Proc. Natl. Acad.
Sci. U.S.A. {\bf 108}, 12238 (2011).
\bibitem{ying}
J. J. Ying, X. F. Wang, T. Wu, Z. J. Xiang, R. H. Liu, Y. J. Yan, A. F.
Wang, M. Zhang, G. J. Ye, P. Cheng, J. P. Hu, and X. H. Chen, Phys.
Rev. Lett. {\bf 107}, 067001 (2011).
\bibitem{yin}
Z. P. Yin, K. Haule, and G. Kotliar,
Nat. Phys. {\bf 7}, 294 (2011).
\bibitem{sugimoto1}
K. Sugimoto, E. Kaneshita, and T. Tohyama, J. Phys. Soc. Jpn.
{\bf 80}, 033706 (2011).
\bibitem{zhang}
X. Zhang and E. Dagotto, Phys. Rev. B {\bf 84}, 132505 (2011).
\bibitem{valenzuela}
B. Valenzuela, E. Bascones, and M. J. Calder\'{o}n, Phys. Rev. Lett. {\bf 105}, 207202 (2010).
\bibitem{daghofer1}
M. Daghofer, Q.-L. Luo, R. Yu, D. X. Yao, A. Moreo, and
E. Dagotto, Phys. Rev. B {\bf 81}, 180514(R) (2010).
\bibitem{liang}
S. Liang, G. Alvarez, C. Sen, A. Moreo, and E. Dagotto, Phys. Rev.
Lett. {\bf 109}, 047001 (2012).
\bibitem{gastiasoro}
M. N. Gastiasoro, I. Paul, Y. Wang, P. J. Hirschfeld, and B. M.
Andersen, Phys. Rev. Lett. {\bf 113}, 127001 (2014).
\bibitem{timm}
M. Breitkreiz, P. M. R. Brydon, and C. Timm, Phys. Rev. B {\bf 90},
121104(R) (2014).
\bibitem{dheeraj}
D. K. Singh and P. Majumdar,
Phys. Rev. B {\bf 96}, 235111 (2017); D. K. Singh, A. Akbari, and P. Majumdar, arXiv:1805.09582
\bibitem{wang}Y. Wang, Maria N. Gastiasoro, Brian M. Andersen, M. Tomic,
H. O. Jeschke, R. Valentí, I. Paul, and P. J. Hirschfeld, Phys. Rev. Lett. {\bf 114}, 097003 (2015).
\bibitem{sugimoto}
K. Sugimoto, P. Prelov\v{s}ek, E. Kaneshita, and T. Tohyama, Phys. Rev.
B {\bf 90}, 125157 (2014).
\bibitem{kuo}
H.-H. Kuo and I. R. Fisher, Phys. Rev. Lett. {\bf 112}, 227001 (2014).
\bibitem{kobyashi}
T. Kobayashi, K. Tanaka, S. Miyasaka, and S. Tajima,
\jpsj {84}, 094707 (2015).
\bibitem{daghofer}
M. Daghofer, A. Nicholson, A. Moreo, and E. Dagotto, Phys. Rev.
B {\bf 81}, 014511 (2010).
\bibitem{graser}
S. Graser, T. Maier, P. Hirschfeld, and D. Scalapino, New J. Phys.
{\bf 11}, 025016 (2009).
\bibitem{ikeda}
H. Ikeda, R. Arita, and J. Kunes, Phys. Rev. B {\bf 81}, 054502
(2010).
\bibitem{dagotto}
E. Dagotto, Rev. Mod. Phys. {\bf 66}, 763 (1994).
\bibitem{valenzuela1}
B. Valenzuela, M. J. Calder\'{o}n, G. Le\'{o}n, and E. Bascones,
Phys. Rev. B {\bf 87}, 075136 (2013).
\bibitem{kemper}
A. F. Kemper, M. M. Korshunov, T. P. Devereaux,
J. N. Fry, H-P. Cheng,
and P. J. Hirschfeld, Phys.
Rev. B {\bf 83}, 184516 (2011).
\bibitem{schafgans}
A. A. Schafgans, S. J. Moon, B. C. Pursley, A. D. LaForge, M. M. Qazilbash,
A. S. Sefat, D. Mandrus, K. Haule, G. Kotliar, and D. N. Basov,
Phys. Rev. Lett. {\bf 108}, 147002 (2012).
\bibitem{avci} S. Avci, O. Chmaissem, D. Y. Chung, S. Rosenkranz, E. A. Goremychkin, J. P. Castellan,
I. S. Todorov, J. A. Schlueter, H. Claus, A. Daoud-Aladine, D. D. Khalyavin,
M. G. Kanatzidis, and R. Osborn, Phys. Rev. B {\bf 85}, 184507 (2012).
\bibitem{yi1} M. Yi, D. H. Lu, J. G. Analytis, J.-H. Chu, S.-K. Mo,
R.-H. He, M. Hashimoto, R. G. Moore, I. I. Mazin,
D. J. Singh, Z. Hussain, I. R. Fisher and Z.-X. Shen,
Phys. Rev. B {\bf 80}, 174510 (2009).
\bibitem{lu} D. H. Lu, M. Yi, S.-K. Mo, J. G. Analytis, J.-H. Chu,
A. S. Erickson, D. J. Singh, Z. Hussain, T. H. Geballe,
I. R. Fisher, Z.-X. Shen, Physica C {\bf 469}, 452 2009.
\bibitem{liu} C. Liu, T. Kondo, R. M. Fernandes, A. D. Palczewski, E. D. Mun, N. Ni, A.
N. Thaler, A. Bostwick, E. Rotenberg, J. Schmalian, S. L. Bud’ko,
P. C. Canfield and A. Kaminski, Nat. Phys. {\bf 6}, 419 (2010).
\bibitem{kovacic}
M. Kovacic, M. H. Christensen, M. N. Gastiasoro and B. M. Andersen,
Phys. Rev. B {\bf 91}, 064424 (2015).
\bibitem{onari}
S. Onari and H. Kontani,
Phys. Rev. B {\bf 96}, 094527 (2017).
\bibitem{fernandes}
R. M. Fernandes, E. Abrahams and J. Schmalian, Phys. Rev. Lett {\bf 107}, 217002 (2011).
\bibitem{medici}
L. de$^{\prime}$ Medici, G. Giovannetti, and M. Capone, Phys. Rev. Lett. {\bf 112},
177001 (2014).
\bibitem{lafuerza}
S. Lafuerza, H. Gretarsson, F. Hardy, T. Wolf, C. Meingast, G. Giovannetti, M. Capone,
A. S. Sefat, Y. -J. Kim, P. Glatzel, and L. de$^{\prime}$ Medici, Phys. Rev. B {\bf 96}, 045133 (2017).
\end{thebibliography}
\end{document}