1: % Template article for preprint document class `elsart'

2: % SP 2001/01/05

3: 

4: \documentclass{elsart3p}

5: 

6: % Use the option doublespacing or reviewcopy to obtain double line spacing

7: %\documentclass[doublespacing]{elsart}

8: 

9: 

10: % if you use PostScript figures in your article

11: % use the graphics package for simple commands

12: \usepackage{graphics}

13: % or use the graphicx package for more complicated commands

14: \usepackage{graphicx}

15: % or use the epsfig package if you prefer to use the old commands

16: %\usepackage{epsfig}

17: 

18: % The amssymb package provides various useful mathematical symbols

19: \usepackage{amssymb}

20: 

21: 

22: \newcommand{\PR}[1]{{\it Phys. Rev. B} {\bf {#1}},}

23: \newcommand{\PRB}[1]{{\it Phys. Rev. B} {\bf {#1}},} 

24: \newcommand{\PRL}[1]{{\it Phys.\ Rev.\ Lett.} {\bf {#1}},} 

25: \newcommand{\JPSJ}[1]{{\it J.\ Phys.\ Soc.\ Jpn.} {\bf #1},} 

26: \newcommand{\sgt}{$\raisebox{-0.6ex}{$\stackrel{>}{\sim}$}$}

27: \newcommand{\slt}{$\raisebox{-0.6ex}{$\stackrel{<}{\sim}$}$}

28: \newcommand{\bvec}[1]{\mbox{\boldmath $#1$}}

29: \newcommand{\D}{\delta }

30: \newcommand{\lam}{\lambda }

31: 

32: \def\vq{{\bf q}}

33: \def\vQ{{\bf Q}}

34: \def\vr{{\bf r}}

35: \newcommand{\eq}[1]{Eq.~(\ref{#1})}

36: \newcommand{\fig}[1]{Fig.~\ref{#1}}

37: \newcommand{\be}{\begin{equation}}

38: \newcommand{\ee}{\end{equation}}

39: 

40: 

41: \begin{document}

42: 

43: \begin{frontmatter}

44: 

45: %titles, authors and addresses

46: 

47: % use the thanksref command within \title, \author or \address for footnotes;

48: % use the corauthref command within \author for corresponding author footnotes;

49: % use the ead command for the email address,

50: % and the form \ead[url] for the home page:

51: % \title{Title\thanksref{label1}}

52: % \thanks[label1]{}

53: % \author{Name\corauthref{cor1}\thanksref{label2}}

54: % \ead{email address}

55: % \ead[url]{home page}

56: % \thanks[label2]{}

57: % \corauth[cor1]{}

58: % \address{Address\thanksref{label3}}

59: % \thanks[label3]{}

60: 

61: \title{Theory of the in-plane anisotropy of magnetic excitations in 

62: YBa$_{2}$Cu$_{3}$O$_{6+y}$}

63: 

64: % use optional labels to link authors explicitly to addresses:

65: % \author[label1,label2]{}

66: % \address[label1]{}

67: % \address[label2]{}

68: 

69: \author{Hiroyuki Yamase and Walter Metzner}

70: 

71: \address{Max-Planck-Institute for Solid State Research, 

72: Heisenbergstrasse 1, D-70569 Stuttgart, Germany}

73: 

74: 

75: 

76: \begin{abstract}

77: A pronounced $xy$-anisotropy was observed in recent neutron scattering

78: experiments for magnetic excitations in untwinned 

79: YBa$_{2}$Cu$_{3}$O$_{6+y}$ (YBCO). 

80: The small anisotropy of the bare band structure due to

81: the orthorhombic crystal symmetry seems to be enhanced by correlation

82: effects. A natural possibility is that the system is close to a

83: Pomeranchuk instability associated with a $d$-wave Fermi surface 

84: deformation ($d$FSD).

85: We investigate this possibility in the bilayer $t$-$J$ model within

86: a self-consistent slave-boson mean-field theory.

87: We show that the $d$FSD correlations drive a pronounced $xy$-anisotropy

88: of magnetic excitations at low doping and at relatively high

89: temperatures, providing a scenario for the observed $xy$-anisotropy 

90: in optimally doped as well as underdoped YBCO, including the pseudogap 

91: phase. 

92: \end{abstract}

93: 

94: \begin{keyword}

95: % keywords here, in the form: keyword; keyword

96: %Firstkeyword\sep Secondkeyword\sep Thirdkeyword

97: magnetic excitation\sep Y-based cuprates\sep Fermi surface

98: 

99: % PACS codes here, in the form: \PACS code \sep code

100: \PACS 74.25.Ha\sep 74.72.Bk\sep 74.20.Mn\sep 71.10.Fd

101: \end{keyword}

102: \end{frontmatter}

103: 

104: % main text

105: %\section{Title of the first section}

106: %\label{labelOfFirstSection}

107: The interpretation of neutron scattering data for high-$T_{c}$ 

108: cuprates within a spin-charge stripe scenario has attracted

109: much interest\cite{tranquada95}. 

110: It predicts a satellite signal at either $\vq=(\pi,\pi\pm 2\pi\eta)$ or 

111: $(\pi\pm 2\pi\eta,\pi)$, depending on the direction of the stripes. 

112: Recently Hinkov {\it et al.}\cite{hinkov0406} performed 

113: neutron scattering experiments for untwinned YBa$_{2}$Cu$_{3}$O$_{6+y}$ 

114: to check this scenario. 

115: The untwinned crystals  

116: are purely orthorhombic and thus it is expected that 

117: the stripe direction is determined uniquely by the orthorhombicity, 

118: leading to only one satellite signal. 

119: However, they observed two satellite 

120: signals at both $\vq=(\pi,\pi\pm 2\pi\eta_{y})$ and 

121: $(\pi\pm 2\pi\eta_{x},\pi)$, with $\eta_{x}\ne \eta_{y}$ and 

122: different intensity, 

123: and concluded that the stripe scenario did not work. 

124: Then how can we understand this $xy$-anisotropy of 

125: magnetic excitations? 

126: Our scenario is that it comes from 

127: electron correlation effects associated with a 

128: $d$-wave Fermi surface deformation ($d$FSD). 

129: 

130: The $d$FSD was first discussed in the $t$-$J$\cite{yamase00} 

131: and the Hubbard\cite{halboth00} model: 

132: the Fermi surface expands 

133: along the $k_{x}$ direction and shrinks along the $k_{y}$ direction, 

134: or vice versa.  

135: The $d$FSD is driven by forward scattering interactions of 

136: quasi-particles (Pomeranchuk instability).  

137: The $d$FSD competes with the $d$-wave superconductivity ($d$SC). 

138: In the slave-boson mean-field analysis of the $t$-$J$ 

139: model\cite{yamase00}, the dominant instability is the $d$SC and the 

140: spontaneous $d$FSD does not occur. 

141: However, the system still has appreciable $d$FSD correlations, 

142: which yield an enhanced anisotropy of a 

143: renormalized band structure in the presence of an external 

144: anisotropy such as orthorhombicity of a lattice\cite{yamase00}. 

145: 

146: We explore this $d$FSD correlation effect and 

147: compute magnetic excitations in the bilayer 

148: $t$-$J$ model in the slave-boson mean-field scheme with a small 

149: $xy$-anisotropy in $t$ and $J$. 

150: The magnetic excitations contain the even and the odd channel. 

151: Comprehensive results for the odd channel as well as our detailed 

152: formalism are presented in 

153: Ref.\cite{yamase06}. We focus on the even channel in this paper 

154: and take the same model parameters as in Ref.\cite{yamase06}. 

155: 

156: $\vq$ maps of magnetic excitation spectra  

157: are shown in \fig{qmapde} 

158: for a sequence of hole-doping rates $\delta$ at low $T$. 

159: The left-hand panels are self-consistent results while 

160: the right-hand panels are corresponding non-self-consistent results 

161: where $d$FSD correlations are switched off, keeping just the

162: bare input anisotropy. 

163: The strong spectral weight forms a diamond shaped distribution around 

164: $\vq=(\pi,\pi)$ with prominent weight at 

165: $\vq=(\pi,\pi\pm 2\pi\eta_{y})$ and $(\pi\pm 2\pi\eta_{x},\pi)$. 

166: A robust measure of the anisotropy of magnetic excitations 

167: is given by $\Delta\eta=\eta_{x}-\eta_{y}$\cite{yamase06}.  

168: Although the input anisotropy is fixed, we can read off 

169: appreciable $\delta$ dependence of $\Delta\eta$ from \fig{qmapde}. 

170: In particular, the bare anisotropy effect (right-hand panels) 

171: becomes less effective for lower $\D$, where 

172: an almost isotropic distribution is seen. 

173: The $d$FSD correlations, on the other hand, enhance the bare anisotropy 

174: effect and yield a sizable anisotropy even for low $\D$. 

175: 

176: \begin{figure}[t]

177: %\begin{figure}[!ht]

178: \begin{center}

179: \includegraphics[width=0.37\textwidth]{qmapde-v.eps}

180: \end{center}

181: \caption{(Color online) 

182: $\vq$ maps of magnetic excitation spectra for a sequence of 

183: $\D$ for $T=0.01J$ and $\omega=0.30J$ 

184: in the even channel; $5\% (10\%)$ anisotropy is introduced in $t(J)$; 

185: the right-hand panels show a bare anisotropy effect, namely 

186: without $d$FSD correlations; 

187: $\vq$ is  scanned in $0.6\pi \leq q_{x}, q_{y} \leq 1.4\pi$ 

188: except for the panels for $\D=0.20$ where 

189: $0.5\pi\leq q_{x}, q_{y} \leq 1.5\pi$.}

190: \label{qmapde}

191: \end{figure}

192: 

193: The importance of the $d$FSD correlations appears also 

194: in the $T$ dependence. 

195: The diamond shaped distribution at low $T$ (\fig{qmapde}) is rather 

196: robust against $T$ although 

197: the spectral weight around $\vq=(\pi,\pi)$ increases with $T$. 

198: Such spectral weight finally 

199: becomes dominant [\fig{qmapTe}(a)] and a 

200: pronounced $xy$-anisotropy appears. 

201: Figure~\ref{qmapTe} shows a comparison between results including 

202: $d$FSD correlations (left-hand panels) and results based on the 

203: bare anisotropy only (right-hand panels), for a sequence of $T$. 

204: We see that $d$FSD correlations drive a pronounced anisotropy 

205: for temperatures around $T_{\rm RVB}$ (corresponding to the

206: pseudogap temperature) while they become less effective for 

207: higher $T$. 

208: 

209: \begin{figure}[t]

210: %\begin{figure}[!ht]

211: \begin{center}

212: \includegraphics[width=0.37\textwidth]{qmapTe-v.eps}

213: \end{center}

214: \caption{(Color online) 

215: $\vq$ maps of magnetic excitation spectra for a sequence of $T$ 

216: in $0.6\pi\leq q_{x}, q_{y} \leq 1.4\pi$ for   

217: $\D=0.10$ and $\omega=0.30J$ in the even channel; 

218: $5\% (10\%)$ anisotropy is introduced in $t(J)$; 

219: right-hand panels show a bare anisotropy effect; 

220: the $d$-wave singlet is realized below 

221: $T_{\rm RVB}=0.133J (0.139J)$ 

222: %$T<0.133J (0.139J)$ 

223: in the left-hand (right-hand) cases.} 

224: \label{qmapTe}

225: \end{figure}

226: 

227: We have seen that the $d$FSD correlations lead to a pronounced 

228: anisotropy of magnetic excitations at low $\D$ and at relatively 

229: high $T$, which are common features in both the odd\cite{yamase06} 

230: and the even channel. 

231: Anisotropic magnetic excitations are observed 

232: in YBa$_{2}$Cu$_{3}$O$_{6.6}$\cite{hinkov0406} 

233: and YBa$_{2}$Cu$_{3}$O$_{6.5}$\cite{stock04},  

234: and in the pseudogap phase\cite{hinkov0406}. 

235: These data can not be understood by the bare anisotropy effect only 

236: (see right-hand panels in Figs.~\ref{qmapde} and \ref{qmapTe}), 

237: implying the importance of $d$FSD correlations. 

238: 

239: \vspace {-3mm}

240: 

241: \begin{thebibliography}{00}

242: \bibitem{tranquada95}

243: J . M. Tranquada {\it et al.}, Nature 375 (1995) 561. 

244: \bibitem{hinkov0406}

245: V. Hinkov {\it et al.}, Nature 430 (2004) 650; V. Hinkov {\it et al.}, 

246: cond-mat/0601048.  

247: \bibitem{yamase00}

248: H. Yamase and H. Kohno, J. Phys. Soc. Jpn. 69 (2000) 332; 

249: 69 (2000) 2151.

250: \bibitem{halboth00}

251: C. J. Halboth and W. Metzner, Phys. Rev. Lett. 85 (2000) 5162.

252: \bibitem{yamase06} 

253: H. Yamase and W. Metzner, Phys. Rev. B 73 (2006) 214517. 

254: \bibitem{stock04}

255: C. Stock {\it et al.} Phys. Rev. B 69 (2004) 14502.  

256: 

257: 

258: % notes:

259: % \bibitem{label} \note

260: 

261: % subbibitems:

262: % \begin{subbibitems}{label}

263: % \bibitem{label1}

264: % \bibitem{label2}

265: % If there is a note, it should come last:

266: % \bibitem{label3} \note

267: % \end{subbibitems}

268: \end{thebibliography}

269: 

270: \end{document}

271: 

272: