1: \documentclass[a4paper,prb,twocolumn,showpacs]{revtex4}
2: %\documentclass[12pt]{article}
3: %\def\baselinestretch{1.5}
4: \usepackage{epsfig}
5: \usepackage{hyperref}
6: %\usepackage{feynmf}%%%{feynmp}
7: \unitlength=1mm
8: \begin{document}
9: \title{Electronic pseudogap of optimally doped
10: Nd$_{2-x}$Ce$_x$CuO$_4$}
11: % \author{D.K.~Sunko\thanks{email: dks\@phy.hr}~~and S.~Bari\v si\``c\\
12: % Department of Physics,\\Faculty of Science,\\ % University of Zagreb,\\
13: % Bijeni\v cka cesta 32,\\ % HR-10000 Zagreb, Croatia.}
14: \author{D.~K.~Sunko}
15: \email{dks@phy.hr}
16: \author{S.~Bari\v si\'c}
17: \email{sbarisic@phy.hr}
18: \affiliation{Department of Physics, Faculty of Science, University of
19: Zagreb,\\ Bijeni\v cka cesta 32, HR-10000 Zagreb, Croatia.}
20: %\date{}
21: \pacs{74.25.Jb,74.72-h,79.60-i}
22: \newcommand{\ReS}{\mathrm{Re}\,\Sigma}
23: \newcommand{\ImS}{\mathrm{Im}\,\Sigma}
24: \newcommand{\eV}{\;\mathrm{eV}}
25: \begin{abstract}
26:
27: We study the effect of antiferromagnetic correlations in the three-band Emery
28: model, in comparison with the experimental angle-resolved photoemission
29: (ARPES) spectra in optimally doped NCCO. The same calculation, formerly used
30: to describe BSCCO, is applied here, but in contrast to BSCCO, where quantum
31: paramagnon fluctuations are important, the characteristic energy of the
32: dispersive paramagnons in NCCO is of the order of $T_c$. The wide dispersing
33: features of the single-electron spectrum in NCCO are analogous to the BSCCO
34: hump. The Fermi surface is pseudogapped in both the nodal and antinodal
35: directions, although the detailed features differ, being dominated by loss of
36: intensity in the nodal direction, and loss of coherence in the antinodal one.
37: Direct oxygen-oxygen hopping is important in NCCO as well as in BSCCO, in
38: order to obtain overall agreement with the measured ARPES spectra.
39:
40: \end{abstract}
41:
42: \maketitle
43:
44: %\section{Introduction}
45:
46: High-temperature superconductivity (SC) is one of the premier problems of
47: physics today. It occurs across a broad range of copper oxide perovskites,
48: whose electronic responses are quite dissimilar. The $n$-doped
49: high-temperature superconductors show a number of outstanding differences with
50: the $p$-doped ones. First, $T_c$ is much lower, by a factor of 2--5. Also,
51: static antiferromagnetism (AF) extends to much higher doping, so that muon
52: resonance~\cite{Uefuji01} measurements clearly show a N\'eel temperature $T_N$
53: \emph{above} the superconducting (SC) $T_c$ in underdoped NCCO,
54: Nd$_{2-x}$Ce$_x$CuO$_4$ for $x<0.15$. At optimal doping, $x=0.15$, on which we
55: concentrate in the following, the static AF response abruptly disappears, but
56: another signal, indicating AF correlations without long-range order, persists
57: for $T<T_c$.\cite{Uefuji01,Yamada99} In contrast, the $p$-doped compounds have
58: a wide ``pseudogap'' region in the phase diagram, between the AF and SC
59: phases.\cite{Ding96,Loeser96}
60:
61: The ARPES ``effective band'' electron response in the two classes of compounds
62: also appears to be quite different. In BSCCO and YBCO, the SC ARPES profile at
63: the van Hove (vH, antinodal) point is dominated by the so-called peak-dip-hump
64: structure.\cite{Campuzano99,Fedorov99,Lu01} There is a clear gap of the order
65: of several $T_c$, the ``leading-edge scale,'' which tapers to zero at the
66: nodal point in optimally doped (OP) BSCCO, but persists in the slightly
67: underdoped case, albeit without the narrow peak. In NCCO, a wide structure
68: disperses quickly across the Fermi level at both nodal and antinodal points. A
69: visible pseudogap persists only at the ``hot spot'' for AF scattering, where
70: the ARPES intensity drops sharply as the Fermi surface crosses the
71: skew-diagonal of the
72: zone.\cite{Armitage01,Onose01,Armitage01-1,Sato01,Armitage02,Blumberg02} This
73: is a precursor of the AF gap, which opens on the skew-diagonal for $T<T_N$.
74: The Fermi surface surmised from these measurements is hole-like in both BSCCO
75: and NCCO, but with a different shape, a rounded square around the $(\pi,\pi)$
76: point for BSCCO, and nearly a circle for NCCO.
77:
78: These results are discussed here within the three band Hubbard model (Emery
79: model),\cite{Emery87,Varma87} which treats the $p$- and $n$-doped planar
80: CuO$_2$ cuprates on equal footing from the outset. It introduces the
81: difference $\Delta_{pd}$ between the O and Cu site energies, and the Cu--O and
82: O--O hoppings, $t_{pd}$ and $t_{pp}$ respectively, together with the large
83: interaction $U_d$ between the two holes on the Cu site. These parameters
84: reflect the chemistry of the CuO$_2$ planes and are therefore expected to be
85: similar for a given material upon both $p$ ($\delta>0$) and $n$ ($\delta<0$)
86: dopings. The physical regime which appears to be appropriate for high-T$_c$ SC
87: corresponds~\cite{Mrkonjic03,Mrkonjic03-2} to $\Delta_{pd}>|t_{pd}|>|t_{pp}|$,
88: with $|t_{pp}|$ and $t_{pd}^2/\Delta_{pd}$ of the same order of magnitude. The
89: AF order is suppressed asymetrically, typically for $\delta>0.01$ on the $p$
90: side and $-\delta>0.1$ on the $n$ side.
91:
92: An alternative approach often applied to NCCO is the one-band Hubbard model,
93: viewed ``from the insulator.''\cite{Kuroki99,Kusko02,Kyung04} It attempts to
94: face experiment by taking into account the oxygen degree of freedom at least
95: through a large $t'\cos k_x\cos k_y$ term, a contribution of the same symmetry
96: as direct O--O hopping.\cite{Mrkonjic03} Models with more than three bands
97: are also sometimes considered,\cite{Bansil05} while on the $p$-doped side
98: three-band approaches have been used to discuss (stripe)
99: inhomogeneities.\cite{Lorenzana02,Abbamonte05}
100:
101: In the Emery model, the asymmetry of AF and other physical properties between
102: the $p$- and $n$-dopings comes from the different role of $U_d$ in
103: the two cases. It is conveniently observed in the mean-field (MF) slave-boson
104: (SB) approximation for $U_d\to\infty$, which replaces the ``bare'' chemical
105: parameters $\Delta_{pd}$ and $t_{pd}$ with their renormalized values
106: $\Delta_{pf}$ and $t$, respectively, with $t_{pp}$ unchanged, in an effective
107: free three-band model, with a definite dependence of the renormalized
108: parameters on both the chemical ones and on $\delta$. Although crude, the MF,
109: paramagnetic, translationally invariant version of this theory, combined with
110: harmonic fluctuations of the SB field around the MF saddle point,
111: gives~\cite{Sunko05-1,Mrkonjic03} physically reasonable band structures in the
112: metallic regime.\cite{Lorenzana02}
113:
114: The possible three-band regimes in the Emery model in such a situation have
115: been extensively classified.\cite{Mrkonjic03} In the presence of the
116: unrenormalized direct oxygen-oxygen hopping $t_{pp}$, the signature of strong
117: renormalization is a regime change,\cite{Mrkonjic04-1} from
118: $0<-t_{pp}<t_{pd},\Delta_{pd}/4$ to $-t_{pp}>t,\Delta_{pf}/4>0$. In
119: particular, $\Delta_{pf}< 4|t_{pp}|$ means that the intrinsic band-width
120: of the oxygen band exceeds the effective copper-oxygen splitting, so there
121: occurs an ``anticrossing,'' in which the lowest (bonding) hole band acquires a
122: significant oxygen character.
123:
124: Although the main effects of $U_d$ on the overlap $t$ are similar on the $p$-
125: and $n$-sides, the renormalization mechanism is different. In the former,
126: paramagnetic charge interactions push the doped holes onto the oxygen sites.
127: This leads to a paramagnetic ``resonant band'' regime, $\Delta_{pf}\approx
128: 4|t_{pp}|$, useful in BSCCO.~\cite{Sunko05-1} In NCCO, doping puts carriers on
129: the copper sites. Long-range AF survives to large $n$-dopings, but at some
130: point there are enough carriers for a significant gain in kinetic energy, if
131: they could spend more time on the oxygens. This is expressed by a band
132: renormalization into the anticrossing regime, $-4t_{pp}\gg\Delta_{pf}>0$,
133: which thus corresponds to a paramagnetic ``lower Hubbard band.''
134:
135: The correlations omitted by the MF approach can also be described in terms of
136: the same bare parameters. In this scheme, far enough from the AF transition,
137: the magnetic correlations can be treated as dispersive paramagnons, and
138: included in a one-loop calculation:\cite{Sunko05-1}
139: %\begin{widetext}
140: $$
141: \Sigma_R(\mathbf{k},\omega)\propto-F
142: \int d^2q\int_{-\infty}^\infty d\omega'
143: $$
144: $$
145: \left[
146: \chi_R(\mathbf{Q+q},\omega-\omega')(1-f(\omega'))
147: \mathrm{Im}\,G^{(0)}_R(\mathbf{k-q-Q},\omega')
148: \right.
149: $$
150: \begin{equation}
151: \left. +G^{(0)}_R(\mathbf{k-q-Q},\omega-\omega')
152: n(\omega')\mathrm{Im}\,\chi_R(\mathbf{Q+q},\omega')\right],
153: \label{eqsigma}
154: \end{equation}
155: %\end{widetext}
156: where $F$ is an effective interaction constant and $\mathbf{Q}=(\pi,\pi)$. The
157: first term is the magnon propagator convoluted with the electron response, the
158: second, vice versa. Both are equally important in BSCCO, but the one with the
159: boson occupation number dominates in NCCO.
160:
161: %\section{Results}\label{res}
162:
163: \begin{figure}
164: \begin{tabular}{cc}
165: \hskip 3mm\raisebox{7mm}{\includegraphics[height=30mm]{ncco-cuts}}&
166: \includegraphics[height=40mm]{figlowbh-col}
167: \end{tabular}
168: \vskip -5mm
169: \caption{Left: experimental~\cite{Armitage01-1} NCCO ARPES intensity at the
170: Fermi level for $T<20$~K, $T_c=24$~K. Right: present calculation.}
171: \label{figfer}
172: \end{figure}
173: In Figure~\ref{figfer} we compare our results with experiment. The Fermi
174: surface is first fitted to obtain the renormalized parameters of the
175: three-band model. We can easily reproduce the shape of the Fermi surface in
176: NCCO in a regime very similar to the one in BSCCO,\cite{Sunko05-1} namely with
177: $\Delta_{pf}>-t_{pp}\gg t>0$, the main difference being that $\Delta_{pf}$ is
178: now smaller than in BSCCO: $\Delta_{pf}=1.6$~eV, $t_{pp}=-1.2$~eV and
179: $t=0.3$~eV, deep in the anticrossing regime. It is noteworthy that the
180: zeroth-order band prediction indicates that the effect of the oxygen hopping
181: $t_{pp}$ is as large in the $n$-doped cuprates as in the $p$-doped ones. The
182: Fermi surface in Fig.~\ref{figfer} is actually found beyond the anticrossing
183: point of the bonding dispersion, where the wave-functions have significant
184: oxygen character, making the argument in favor of anti-crossing
185: self-consistent. This regime cannot be reached consistently from the
186: single-band Hubbard model.
187:
188: \begin{figure}
189: \begin{tabular}{cc}
190: \hskip -10mm\includegraphics[height=6.5cm]{ncco-exp}&
191: \hskip 0mm\includegraphics[height=6.5cm]{ncco-th}
192: \end{tabular}
193: \caption{Left: experimental~\cite{Armitage01-1} ARPES profiles for the cuts
194: (a)--(d) in Fig.~\ref{figfer}. The positions in the zone are indicated as
195: percentages along the ranges given by arrows. Right: calculated ARPES profiles
196: for the same positions, without the Fermi factor. The nodal $(\pi/2,\pi/2)$,
197: van Hove $(\pi,0)$, and hot-spot $(0.65\pi,0.36\pi)$ profiles are indicated
198: with heavy lines.}
199: \label{figedc}
200: \end{figure}
201:
202: The paramagnon fit parameters are: band-edge
203: $\widetilde{\omega}=0.001\eV<kT=0.002\eV<\gamma=0.004\eV$ (damping), cutoff
204: $\omega_0=0.15\eV$, coupling constant $F=0.77\eV$, giving a self-energy range
205: shown in Fig.~\ref{sigma} below. The main input scales are all taken from
206: experiment.\cite{Uefuji01,Armitage01-1} The calculation is not sensitive to
207: the magnon damping and cutoff, kept roughly similar as in BSCCO. The only free
208: parameter left is the coupling constant $F$. With this one adjustment, all
209: ARPES scales, dispersions and widths in Fig.~2 were obtained rather well. Even
210: smaller experimental features have their counterparts in theory, such as the
211: flattening of the peak on the $(\pi,0)$--$(\pi,\pi)$ line, discussed below,
212: and the difference in dispersion shape (trend of the peaks) between panels (c)
213: and (d) in Fig.~\ref{figedc}.
214:
215: The chemical potential (relative to the vH point) in this fit, $\mu=0.75$~eV,
216: is more than twice as large than experiment seems to suggest. The experimental
217: panel (c) in Fig.~\ref{figedc} shows that the ARPES profile at the vH point
218: has a maximum at about 0.35~eV binding energy. The lowest curve in the
219: theoretical panel (c) in the same figure explains the discrepancy: the
220: ``bump'' there is in fact the upper wing of a spectral intensity split by AF
221: interactions. In this way the apparent Fermi level mismatch is resolved,
222: without spoiling the zeroth-order Fermi surface fit, as evident from
223: Fig.~\ref{figfer}. The narrow lower wing at nearly 1~eV binding corresponds to
224: the experimental ARPES intensity rising again at higher energy. In the
225: approach from the paramagnetic metal, high-energy features are subject to
226: further corrections, so let us first focus on the wide dispersive upper wings.
227:
228: Our calculation in Fig.~\ref{figedc} gives the impression that all the wide
229: features observed in OP NCCO are incoherent.
230: Experimentally,~\cite{Armitage01-1} in addition to the hot spot, there is also
231: a drop in intensity in the nodal direction \emph{before} the Fermi level is
232: approached. This is exactly the opposite behavior than one would expect from a
233: quasiparticle. To emphasize the point, theoretical curves are shown without
234: the Fermi factor, and it is clear that the same phenomenon is occuring in the
235: calculation as well. We extend the experimental
236: observation,\cite{Armitage01-1} and claim that the Fermi surface in OP NCCO is
237: pseudogapped everywhere.
238:
239: \begin{figure}
240: \includegraphics[height=50mm]{ncco-sigma}
241: \caption{Spectral intensities and the corresponding self-energies at selected
242: points in the Brillouin zone. Thin dashed line: vH point. Thin full line: 25\%
243: of the zone away from it in the direction of the AF point. Thick full line:
244: hot spot. Thick dashed line: 50\% of the zone in the nodal direction
245: $k_x=k_y$. The transition to the pseudogapped regime around $E_F$ is
246: emphasized by drop-down thin lines at $\omega-\mu=-0.14\eV$.}
247: \label{sigma}
248: \end{figure}
249: A detailed analysis is given in Figure~\ref{sigma}. At the vH point, the
250: chemical potential puts a quasiparticle at 0.75~eV binding energy, but the
251: self-energy shows that this is in the middle of a forbidden region: $\ReS$ is
252: rising, $\ImS$ is large around there. The quasiparticle is split, and the wide
253: structure at 0.45~eV appears as the upper part of a split quasiparticle, not
254: necessarily incoherent. As $k_y$ increases along the $(\pi,0)$--$(\pi,\pi)$
255: line, this signal enters another forbidden region, at about $\pm 0.1$~eV
256: around $E_F$. We have purposefully put a rather low boson damping in the
257: calculation, to observe the moment of transition as a fine structure in the
258: profile at $k_y=0.25\pi/a$: the small sharp spike is at the edge of the
259: quasiparticle region, while the peak at right is already incoherent. (The
260: experimental profiles at 25\% and 29\% of the zone in panel (c) of
261: Fig.~\ref{figedc} show a similar flattening at the top.) Loss of coherence as
262: the Fermi energy is approached indicates a pseudogap regime, validating the
263: initial impression.
264:
265: A similar phenomenon occurs in the nodal direction. The dominant feature is
266: loss of intensity rather than loss of coherence, because the Fermi surface is
267: nearly nested. The quasiparticle loses intensity before entering the pseudogap
268: region around $E_F$, and reappears at the other side, without really crossing
269: $E_F$. Succintly, the antinodal region is pseudogapped because an incoherent
270: signal crosses the Fermi level, while the nodal one is pseudogapped because a
271: coherent signal does not cross it. Technically, $\ReS$ increases everywhere at
272: the Fermi level, precluding a quasiparticle, while $\ImS$ is large at the
273: nodal Fermi crossing, because of nesting, but fairly small at the antinodal
274: crossing, which is not nested.
275:
276: At the hot spot, both effects combine: an incoherent structure is suppressed
277: in intensity before it can cross the Fermi level. Finally, as one moves from
278: the vH point to the $\Gamma$-point, the increase in binding energy is
279: accompanied with a loss of both coherence and intensity, as the upper wing of
280: the split quasiparticle approaches the forbidden region below $\sim$0.5~eV
281: binding, where the slope $\ReS$ is positive, and $\ImS$ is large (see panel
282: (b) of Fig.~\ref{figedc}).
283:
284: %\section{Discussion}
285:
286: The present work establishes a number of parallels and distinctions between
287: BSCCO and NCCO. The chemical potential, relative to the vH point, and the
288: paramagnon coupling constant are much larger in NCCO, the first by a factor of
289: $\sim$30, the second $\sim$5. This makes the main ARPES scales in NCCO
290: significantly larger, $\sim$$0.5$ \emph{vs.} $\sim$$0.1$~eV in BSCCO. The
291: larger coupling constant is a plausible consequence of doping on the coppers,
292: where the carriers enter an otherwise strongly AF-polarized environment, as
293: evident from the wider extension of AF on the $n$-doped side. On the other
294: hand, when enough carriers are doped on the oxygens, the superexchange is only
295: indirectly involved in their magnetic correlations.
296:
297: The paramagnon physical regime turns out to be significantly different in the
298: two pseudogaps. In SC BSCCO, paramagnons are quantum fluctuations, whose
299: energy scale (41~meV) is much higher than $T_c$. To put NCCO in the same
300: regime requires a paramagnon band-edge $\widetilde{\omega}\sim 10$~meV. We
301: find this at odds with the distinct hot spot in experiment. If the band-edge
302: is comparable with the temperature, and especially if it is lower as here, the
303: hot spot becomes clearly discerned.\cite{Sunko05-2} We conclude the
304: paramagnons in SC NCCO are semiclassical, or transitional (the results are
305: similar if $\widetilde{\omega}\sim T_c$). Paramagnons in the vicinity of the
306: AF state indeed have low energies.~\cite{Uefuji01}
307:
308: Conversely, if the paramagnon band-edge is lowered in BSCCO, the narrow
309: ``antiadiabatic''~\cite{Sunko05-1} peak near the vH point disappears. This is
310: observed in underdoped BSCCO, as expected if the AF transition is approached.
311: We have argued~\cite{Sunko05-1} that the low-energy peak only obscures the
312: pseudogap in OP BSCCO, because it is part of the pseudogap profile, not a true
313: quasiparticle. Without it, there remains the usual splitting of the spectral
314: intensity in two peaks, due to AF scattering. In BSCCO, the splitting
315: straddles the Fermi level, and the lower (occupied) wing is sufficiently close
316: to $E_F$ at the $(\pi,0)$--$(\pi,\pi)$ line to be incoherent. In NCCO, the
317: chemical potential places both wings below the Fermi energy at the vH point,
318: and the upper one becomes incoherent as it disperses towards $E_F$ along the
319: same line. Hence the wide dispersive wing in NCCO is analogous to the hump in
320: BSCCO.
321:
322: Paramagnon damping was a critical parameter in BSCCO, where a switch from
323: over- to underdamped produced the change in ARPES profile observed at
324: T$_c$.\cite{Sunko05-1} By contrast, the parametrization used here for NCCO is
325: only weakly sensitive to the damping regime, which neatly parallels the
326: experimental situation: the changes in ARPES profiles as NCCO becomes
327: superconducting are very slight.\cite{Armitage01,Armitage01-1} This
328: strenghtens our assertion~\cite{Sunko05-1} that SC has also an indirect effect
329: on the ARPES spectra, by gapping out the magnon damping.
330:
331: In both the BSCCO and NCCO calculations, the region of about $\pm 150$~meV
332: around $E_F$ does not support a quasiparticle, a scale set by the AF
333: interaction. Other sources of decoherence are absent in the numerics, coupling
334: to Mott charge fluctuations and to stripes.~\cite{Kivelson03} Both
335: are a natural next step in the present framework.\cite{Sunko05-1}
336: Thus Mott fluctuations are expected to account for the rise in response
337: below 0.5~eV binding, found in experiment, by broadening the narrow deeply
338: bound peaks visible in the calculation of Fig.~\ref{figedc}. A similar role of
339: Mott fluctuations at the $\Delta_{pf}$ scale has been found in the extended
340: Raman background, in the non-crossing approximation.\cite{Niksic95}
341:
342: We expect our predictions to remain robust at less than 0.5~eV binding.
343: The lowest-energy feature is the coherent-incoherent crossover at
344: $\sim$0.15~eV, still high above the SC scale. Neither high-energy (Mott)
345: perturbations, operating at 1--2~eV, nor low-energy (SC) effects should
346: significantly change the picture given here in the energy window from T$_c$ to
347: $\sim$0.5~eV. The precise position of the crossover at 0.15~eV is
348: parameter-dependent, but its existence is not.
349:
350: The intermediate-energy ($\sim$0.1 eV) features of ARPES spectra in both
351: BSCCO~\cite{Sunko05-1} and NCCO, down to the leading-edge scale at the
352: antinodal point in BSCCO, are reproduced by accounting for AF fluctuations
353: alone. If superconductivity, of whatever origin, is simply making use of the
354: available density of states, as calculated here, it obviously follows that
355: $T_c$ should be much higher in BSCCO than in NCCO. This indicates a somewhat
356: more complex structure of the SC gap than a single $d$-wave, not simply
357: related to the order parameter symmetry.
358:
359: To conclude, we have accounted for ARPES spectra in NCCO focussing exclusively
360: on dispersive paramagnons, without low-energy magnetic responses, in the same
361: simple approach~\cite{Sunko05-1} as previously for BSCCO. We explain the
362: observed differences by a change in the physical regime of the paramagnons,
363: semiclassical in NCCO as opposed to quantum in BSCCO, and a much larger value
364: of the effective coupling constant. Both are consistent with the vicinity of
365: the AF transition in NCCO. Optimally doped NCCO is pseudogapped on the whole
366: Fermi surface, and the wide dispersive features are analogous to the hump in
367: BSCCO. Direct oxygen-oxygen hopping is as important in NCCO as in BSCCO. The
368: phenomenological value of the Cu--O hopping in NCCO indicates, as expected,
369: that it is renormalized by the strong on-site repulsion.
370:
371: %\acknowledgments
372:
373: This work was supported by the Croatian Government under Project~$0119256$.
374:
375: \begin{thebibliography}{28}
376: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
377: \expandafter\ifx\csname bibnamefont\endcsname\relax
378: \def\bibnamefont#1{#1}\fi
379: \expandafter\ifx\csname bibfnamefont\endcsname\relax
380: \def\bibfnamefont#1{#1}\fi
381: \expandafter\ifx\csname citenamefont\endcsname\relax
382: \def\citenamefont#1{#1}\fi
383: \expandafter\ifx\csname url\endcsname\relax
384: \def\url#1{\texttt{#1}}\fi
385: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
386: \providecommand{\bibinfo}[2]{#2}
387: \providecommand{\eprint}[2][]{\url{#2}}
388:
389: \bibitem[{\citenamefont{Uefuji et~al.}(2001)\citenamefont{Uefuji, Kubo, Yamada,
390: Fujita, Kurahashi, Watanabe, and Nagamine}}]{Uefuji01}
391: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Uefuji}}
392: \bibnamefont{et~al.},
393: \bibinfo{journal}{Physica C} \textbf{\bibinfo{volume}{357-60}},
394: \bibinfo{pages}{208} (\bibinfo{year}{2001}).
395:
396: \bibitem[{\citenamefont{Yamada et~al.}(1999)\citenamefont{Yamada, Kurahashi,
397: Endoh, Birgeneau, and Shirane}}]{Yamada99}
398: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Yamada}}
399: \bibnamefont{et~al.}, \bibinfo{journal}{J. Phys. Chem. Solids}
400: \textbf{\bibinfo{volume}{60}}, \bibinfo{pages}{1025} (\bibinfo{year}{1999}).
401:
402: \bibitem[{\citenamefont{Ding et~al.}(1996)\citenamefont{Ding, Yokoya,
403: Campuzano, Takahashi, Randeria, Norman, Mochiku, Kadowaki, and
404: Giapintzakis}}]{Ding96}
405: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Ding}}
406: \bibnamefont{et~al.}, \bibinfo{journal}{Nature}
407: \textbf{\bibinfo{volume}{382}}, \bibinfo{pages}{51} (\bibinfo{year}{1996}).
408:
409: \bibitem[{\citenamefont{Loeser et~al.}(1996)\citenamefont{Loeser, Shen, Dessau,
410: Marshall, Park, Fournier, and Kapitulnik}}]{Loeser96}
411: \bibinfo{author}{\bibfnamefont{A.~G.} \bibnamefont{Loeser}}
412: \bibnamefont{et~al.}, \bibinfo{journal}{Science}
413: \textbf{\bibinfo{volume}{273}}, \bibinfo{pages}{325} (\bibinfo{year}{1996}).
414:
415: \bibitem[{\citenamefont{Campuzano et~al.}(1999)\citenamefont{Campuzano, Ding,
416: Norman, Fretwell, Randeria, Kaminski, Mesot, Takeuchi, Sato, Yokoya
417: et~al.}}]{Campuzano99}
418: \bibinfo{author}{\bibfnamefont{J.~C.} \bibnamefont{Campuzano}}
419: \bibnamefont{et~al.},
420: \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{83}},
421: \bibinfo{pages}{3709} (\bibinfo{year}{1999}).
422:
423: \bibitem[{\citenamefont{Fedorov et~al.}(1999)\citenamefont{Fedorov, Valla,
424: Johnson, Li, Gu, and Koshizuka}}]{Fedorov99}
425: \bibinfo{author}{\bibfnamefont{A.~V.} \bibnamefont{Fedorov}}
426: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
427: \textbf{\bibinfo{volume}{82}}, \bibinfo{pages}{2179} (\bibinfo{year}{1999}).
428:
429: \bibitem[{\citenamefont{Lu et~al.}(2001)\citenamefont{Lu, Feng, Armitage, Shen,
430: Damascelli, Kim, Ronning, Shen, Bonn, Liang et~al.}}]{Lu01}
431: \bibinfo{author}{\bibfnamefont{D.~H.} \bibnamefont{Lu}}
432: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
433: \textbf{\bibinfo{volume}{86}}, \bibinfo{pages}{4370} (\bibinfo{year}{2001}).
434:
435: \bibitem[{\citenamefont{Armitage
436: et~al.}(2001{\natexlab{a}})\citenamefont{Armitage, Lu, Feng, Kim, Damascelli,
437: Shen, Ronning, {Z.-X.~Shen}, Onose, Taguchi et~al.}}]{Armitage01}
438: \bibinfo{author}{\bibfnamefont{N.~P.} \bibnamefont{Armitage}}
439: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
440: \textbf{\bibinfo{volume}{86}}, \bibinfo{pages}{1126}
441: (\bibinfo{year}{2001}{\natexlab{a}}).
442:
443: \bibitem[{\citenamefont{Onose et~al.}(2001)\citenamefont{Onose, Taguchi,
444: Ishizaka, and Tokura}}]{Onose01}
445: \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Onose}}
446: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
447: \textbf{\bibinfo{volume}{87}}, \bibinfo{pages}{217001}
448: (\bibinfo{year}{2001}).
449:
450: \bibitem[{\citenamefont{Armitage
451: et~al.}(2001{\natexlab{b}})\citenamefont{Armitage, Lu, Kim, Damascelli, Shen,
452: Ronning, Feng, Bogdanov, {Z.-X.~Shen}, Onose et~al.}}]{Armitage01-1}
453: \bibinfo{author}{\bibfnamefont{N.~P.} \bibnamefont{Armitage}}
454: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
455: \textbf{\bibinfo{volume}{87}}, \bibinfo{pages}{147003}
456: (\bibinfo{year}{2001}{\natexlab{b}}).
457:
458: \bibitem[{\citenamefont{Sato et~al.}(2001)\citenamefont{Sato, Kamiyama,
459: Takahashi, Kurahashi, and Yamada}}]{Sato01}
460: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Sato}}
461: \bibnamefont{et~al.}, \bibinfo{journal}{Science}
462: \textbf{\bibinfo{volume}{291}}, \bibinfo{pages}{1517} (\bibinfo{year}{2001}).
463:
464: \bibitem[{\citenamefont{Armitage et~al.}(2002)\citenamefont{Armitage, Ronning,
465: Lu, Kim, Damascelli, Shen, Feng, Eisaki, {Z.-X.~Shen}, Mang
466: et~al.}}]{Armitage02}
467: \bibinfo{author}{\bibfnamefont{N.~P.} \bibnamefont{Armitage}}
468: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
469: \textbf{\bibinfo{volume}{88}}, \bibinfo{pages}{257001}
470: (\bibinfo{year}{2002}).
471:
472: \bibitem[{\citenamefont{Blumberg et~al.}(2002)\citenamefont{Blumberg, Koitzsch,
473: Gozar, Dennis, Kendziora, Fournier, and Greene}}]{Blumberg02}
474: \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Blumberg}}
475: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
476: \textbf{\bibinfo{volume}{88}}, \bibinfo{pages}{107002}
477: (\bibinfo{year}{2002}).
478:
479: \bibitem[{\citenamefont{Emery}(1987)}]{Emery87}
480: \bibinfo{author}{\bibfnamefont{V.~J.} \bibnamefont{Emery}},
481: \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{58}},
482: \bibinfo{pages}{2794} (\bibinfo{year}{1987}).
483:
484: \bibitem[{\citenamefont{Varma et~al.}(1987)\citenamefont{Varma, Schmitt-Rink,
485: and Abrahams}}]{Varma87}
486: \bibinfo{author}{\bibfnamefont{C.~M.} \bibnamefont{Varma}}
487: \bibnamefont{et~al.}, \bibinfo{journal}{Solid State Commun.}
488: \textbf{\bibinfo{volume}{62}}, \bibinfo{pages}{681} (\bibinfo{year}{1987}).
489:
490: \bibitem[{\citenamefont{Mrkonji\'{c} and
491: Bari\v{s}i\'{c}}(2003{\natexlab{a}})}]{Mrkonjic03}
492: \bibinfo{author}{\bibfnamefont{I.}~\bibnamefont{Mrkonji\'{c}}}
493: \bibnamefont{and}
494: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Bari\v{s}i\'{c}}},
495: \bibinfo{journal}{Eur. Phys. J. B} \textbf{\bibinfo{volume}{34}},
496: \bibinfo{pages}{69} (\bibinfo{year}{2003}{\natexlab{a}}).
497:
498: \bibitem[{\citenamefont{Mrkonji\'{c} and
499: Bari\v{s}i\'{c}}(2003{\natexlab{b}})}]{Mrkonjic03-2}
500: \bibinfo{author}{\bibfnamefont{I.}~\bibnamefont{Mrkonji\'{c}}}
501: \bibnamefont{and}
502: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Bari\v{s}i\'{c}}},
503: \bibinfo{journal}{Eur. Phys. J. B} \textbf{\bibinfo{volume}{34}},
504: \bibinfo{pages}{441} (\bibinfo{year}{2003}{\natexlab{b}}).
505:
506: \bibitem[{\citenamefont{Kuroki et~al.}(1999)\citenamefont{Kuroki, Arita, and
507: Aoki}}]{Kuroki99}
508: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Kuroki}}
509: \bibnamefont{et~al.}, \bibinfo{journal}{J. Low Temp. Phys.}
510: \textbf{\bibinfo{volume}{117}}, \bibinfo{pages}{247} (\bibinfo{year}{1999}).
511:
512: \bibitem[{\citenamefont{Kusko et~al.}(2002)\citenamefont{Kusko, Markiewicz,
513: Lindroos, and Bansil}}]{Kusko02}
514: \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Kusko}}
515: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. B}
516: \textbf{\bibinfo{volume}{66}}, \bibinfo{pages}{140513(R)}
517: (\bibinfo{year}{2002}).
518:
519: \bibitem[{\citenamefont{Kyung et~al.}(2004)\citenamefont{Kyung, Hankevych,
520: Daré, and Tremblay}}]{Kyung04}
521: \bibinfo{author}{\bibfnamefont{B.}~\bibnamefont{Kyung}}
522: \bibnamefont{et~al.}, \bibinfo{journal}{Phys. Rev. Lett.}
523: \textbf{\bibinfo{volume}{93}}, \bibinfo{pages}{147004}
524: (\bibinfo{year}{2004}).
525:
526: \bibitem[{\citenamefont{Bansil et~al.}(2005)\citenamefont{Bansil, Lindroos,
527: Sahrakorpi, and Markiewicz}}]{Bansil05}
528: \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Bansil}}
529: \bibnamefont{et~al.}, \bibinfo{journal}{New J. Phys.}
530: \textbf{\bibinfo{volume}{7}}, \bibinfo{pages}{140} (\bibinfo{year}{2005}).
531:
532: \bibitem[{\citenamefont{Lorenzana and Seibold}(2002)}]{Lorenzana02}
533: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Lorenzana}} \bibnamefont{and}
534: \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Seibold}},
535: \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{89}},
536: \bibinfo{pages}{136401} (\bibinfo{year}{2002}).
537:
538: \bibitem[{\citenamefont{Abbamonte et~al.}(2005)\citenamefont{Abbamonte, Rusydi,
539: Smadici, Gu, Sawatzky, and Feng}}]{Abbamonte05}
540: \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Abbamonte}}
541: \bibnamefont{et~al.}, \bibinfo{journal}{Nature Physics}
542: \textbf{\bibinfo{volume}{1}}, \bibinfo{pages}{155} (\bibinfo{year}{2005}).
543:
544: \bibitem[{\citenamefont{Sunko and
545: Bari\v{s}i\'{c}}(2005{\natexlab{a}})}]{Sunko05-1}
546: \bibinfo{author}{\bibfnamefont{D.~K.} \bibnamefont{Sunko}} \bibnamefont{and}
547: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Bari\v{s}i\'{c}}},
548: \bibinfo{journal}{Eur. Phys. J. B} \textbf{\bibinfo{volume}{46}},
549: \bibinfo{pages}{269} (\bibinfo{year}{2005}{\natexlab{a}}).
550:
551: \bibitem[{\citenamefont{Mrkonji\'{c} and Bari\v{s}i\'{c}}(2004)}]{Mrkonjic04-1}
552: \bibinfo{author}{\bibfnamefont{I.}~\bibnamefont{Mrkonji\'{c}}}
553: \bibnamefont{and}
554: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Bari\v{s}i\'{c}}},
555: \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{92}},
556: \bibinfo{pages}{129701} (\bibinfo{year}{2004}).
557:
558: \bibitem[{\citenamefont{Sunko and
559: Bari\v{s}i\'{c}}(2005{\natexlab{b}})}]{Sunko05-2}
560: \bibinfo{author}{\bibfnamefont{D.~K.} \bibnamefont{Sunko}} \bibnamefont{and}
561: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Bari\v{s}i\'{c}}}, in
562: \emph{\bibinfo{booktitle}{Proc. SPIE, Strongly Correlated Electron Materials:
563: Physics and Nanoengineering}}, edited by
564: \bibinfo{editor}{\bibfnamefont{I.}~\bibnamefont{Bozovic}} \bibnamefont{and}
565: \bibinfo{editor}{\bibfnamefont{D.}~\bibnamefont{Pavuna}}
566: (\bibinfo{publisher}{SPIE}, \bibinfo{year}{2005}{\natexlab{b}}), vol.
567: \bibinfo{volume}{5932}, pp. \bibinfo{pages}{71--83}.
568:
569: \bibitem[{\citenamefont{Kivelson et~al.}(2003)\citenamefont{Kivelson, Bindloss,
570: Fradkin, Oganesyan, Tranquada, Kapitulnik, and Howald}}]{Kivelson03}
571: \bibinfo{author}{\bibfnamefont{S.~A.} \bibnamefont{Kivelson}}
572: \bibnamefont{et~al.}, \bibinfo{journal}{Rev. Mod. Phys.}
573: \textbf{\bibinfo{volume}{75}}, \bibinfo{pages}{1201} (\bibinfo{year}{2003}).
574:
575: \bibitem[{\citenamefont{Nik\v{s}i\'{c}
576: et~al.}(1995)\citenamefont{Nik\v{s}i\'{c}, Tuti\v{s}, and
577: Bari\v{s}i\'{c}}}]{Niksic95}
578: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Nik\v{s}i\'{c}}}
579: \bibnamefont{et~al.}, \bibinfo{journal}{Physica C}
580: \textbf{\bibinfo{volume}{241}}, \bibinfo{pages}{247} (\bibinfo{year}{1995}).
581:
582: \end{thebibliography}
583: \end{document}
584: