1: \documentclass[prl,aps,floatfix,twocolumn]{revtex4}%
2: \usepackage{epsfig}
3: \usepackage{amsmath}
4: \usepackage{amsfonts}
5: \usepackage{amssymb}
6: \usepackage{graphicx}%
7: \setcounter{MaxMatrixCols}{30}
8: \providecommand{\U}[1]{\protect\rule{.1in}{.1in}}
9: \begin{document}
10: \title{Phase-sensitive tests of the pairing state symmetry in Sr$_{2}$RuO$_{4}$}
11: \author{Igor \v{Z}uti\'{c} and Igor Mazin }
12: \affiliation{Center for Computational Materials Science,
13: Naval Research Laboratory,
14: Washington, D.C. 20375}
15:
16: \begin{abstract}
17: Exotic superconducting properties of Sr$_{2}$RuO$_{4}$ have
18: provided strong support for an unconventional pairing symmetry. However, the
19: extensive efforts over the past decade have not yet unambiguously resolved the
20: controversy about the pairing symmetry in this material. While recent
21: phase-sensitive experiments using flux modulation in Josephson junctions
22: consisting of Sr$_{2}$RuO$_{4}$ and a conventional superconductor have been
23: interpreted as conclusive evidence for a chiral spin-triplet pairing, we
24: propose here an alternative interpretation. We show that an overlooked chiral
25: spin-singlet pairing is also compatible with the observed phase shifts in
26: Josephson junctions
27: and propose further experiments which would distinguish it from its
28: spin-triplet counterpart.
29:
30: \end{abstract}
31:
32: \pacs{74.50.+r,74.70.Pq,74.25.Fy}
33: \maketitle
34:
35:
36: \vspace{-0.6cm}
37: %]
38: %\newpage
39: Unambiguous determination of the pairing state symmetry is one of the key
40: steps towards understanding the pairing mechanism in a continously growing
41: class of unconventional superconductors~\cite{unconv}. Phase-sensitive
42: experiments, capable of identifying angular dependence of the superconducting
43: order parameter, have provided a crucial evidence for a dominant $d$-wave
44: orbital symmetry in cuprate
45: superconductors~\cite{Wollman1993:PRL,Wei1998:PRL,Klemm}. However, much less
46: is known for other unconventional superconductors such as heavy fermions,
47: charge transfer salts,
48: and cobaltates. In particular, there is compelling evidence for an
49: unconventional pairing in Sr$_{2}$RuO$_{4}$%
50: ~\cite{Maeno1994:N,Mackenzie2003:RMP}, with the strong possibility of
51: spin-triplet superconductivity which would be a solid-state analog of
52: superfluid He$^{3}$~\cite{Agterberg1997:PRL}.
53:
54: In superconductors with inversion symmetry an order parameter (gap matrix) can
55: be expressed as $\hat{\Delta}(\mathbf{k})=\Delta_{0}(\mathbf{k}) i \hat
56: {\sigma}_{y}$ for spin singlet and $\hat{\Delta}(\mathbf{k})=\mathbf{\hat
57: {\sigma}} \cdot\mathbf{d(k)} i \hat{\sigma}_{y}$ for spin-triplet pairing.
58: Here $\mathbf{\hat\sigma}$ are the Pauli spin matrices and scalar (vector)
59: $\Delta_{0}$ $(\mathbf{d})$ has even (odd) parity in the wavevector
60: \textbf{k}. Often the symmetry of both the orbital and the spin part of
61: $\hat{\Delta}(\mathbf{k})$
62: remains to be identified and the lack of related understanding
63: comes from the difficulty in performing phase-sensitive experiments.
64:
65: While numerous previous experiments probed the pairing symmetry of
66: Sr$_{2}$RuO$_{4}$~\cite{Mackenzie2003:RMP}, in this context, recent
67: phase-sensitive experiments~\cite{Nelson2004:S} that provide
68: angle-resolved
69: information are particularly important. The corresponding measurements were
70: performed in a superconducting quantum interference device (SQUID) geometry,
71: consisting of a pair of Au$_{0.5}$In$_{0.5}$/Sr$_{2}$RuO$_{4}$ Josephson
72: junctions. Since Au$_{0.5}$In$_{0.5}$ is a conventional $s$-wave
73: superconductor, the observed modulation of critical current in an applied
74: magnetic field was interpreted as conclusive support for the phase shifts
75: from an odd-parity spin-triplet pairing in Sr$_{2}$RuO$_{4}$%
76: ~\cite{Nelson2004:S,Rice2004:S}.
77:
78: A similar SQUID geometry was initially proposed~\cite{Geshkenbein1987:PRB} to
79: study possible $p$-wave pairing in heavy fermions and later also used for
80: identifying $d$-wave pairing in cuprates~\cite{Wollman1993:PRL}. The critical
81: current $I_{c}$ is modulated in the applied magnetic field as~\cite{typo}
82: \begin{equation}
83: I_{c}\propto\cos(\Phi/\Phi_{0}+\delta_{12}/2), \label{eq:squid}%
84: \end{equation}
85: where $\Phi$ is the flux threading the SQUID, $\Phi_{0}$ is the flux quantum,
86: and $\delta_{12}$ is the intrinsic phase shift of the order parameter between
87: the two tunneling directions. For a conventional $s$-wave SQUID $\delta
88: _{12}=0$ and $I_{c}$ has a maximum at $\Phi=0$. In contrast, a phase shift
89: $\delta_{12}=\pi$, characteristic of unconventional pairing~\cite{pi},
90: yields a minimum $I_{c}$ at $\Phi=0$. The modulation of external flux together
91: with the fabrication of junctions with varying tunneling directions in SQUID
92: geometry therefore provides an angle-resolved phase-sensitive information about
93: the superconducting pairing symmetry.
94:
95: The suggested chiral $p$-wave (C\textit{p}W) state with the triplet
96: order parameter~\cite{Ishida1998:N},
97: \begin{equation}
98: \mathbf{d(k)}\propto(k_{x}+ik_{y})\mathbf{\hat{z}}, \label{eq:CpW}%
99: \end{equation}
100: in which the spins of the Cooper pairs lie in the RuO$_2$ plane
101: ($\perp {\bf d}$),
102: is indeed compatible with the experiment~\cite{Nelson2004:S}. However,
103: we show here
104: that it is not the only candidate. There exists another pairing state, allowed
105: by the tetragonal symmetry of Sr$_{2}$RuO$_{4}$, the singlet chiral $d$-wave
106: (C\textit{d}W) state $^{1}E_{g}(c)$ with $\Delta_{0}(\mathbf{k})\propto
107: (k_{x}+ik_{y})k_{z}$, or, more accurately~\cite{Annet1990:AP}
108: \begin{equation}
109: \Delta_{0}(\mathbf{k})\propto(k_{x}+ik_{y})\sin k_{z}c, \label{eq:CdW}%
110: \end{equation}
111: which is equally consistent~\cite{T-note} with the phase shifts observed in
112: Ref.~\cite{Nelson2004:S}. We use our findings to propose an experimental test
113: which would discriminate between C\textit{p}W and C\textit{d}W pairing symmetries.
114:
115: Could experimental and theoretical reasons be used to rule out C\textit{d}W
116: and favor only the C\textit{p}W state? The two main arguments in favor of the
117: C\textit{p}W come from
118: muon spin resonance
119: and Knight shift
120: experiments~\cite{Mackenzie2003:RMP,Luke1998:N}. The former indicate a
121: time-reversal symmetry breaking below the transition temperature $T_{c}$,
122: incompatible with the
123: $d_{x^2-y^2}$-wave state %encountered
124: in cuprates,
125: but fully compatible with either
126: C\textit{p}W or C\textit{d}W symmetry. The Knight shift ($K$) was initially
127: interpreted as firm evidence for a triplet state with in-plane spins
128: (like C\textit{p}W), since no change of the in-plane spin susceptibility below
129: $T_{c}$ was found.
130:
131: Even in singlet superconductors
132: ($e.g.$, vanadium),
133: $K$ could remain invariant below $T_{c}$.
134: Such behavior is usually attributed to (a)
135: spin-orbit induced spin-flip scattering which
136: suppresses the effect of the superconductivity on $K$ and (b)
137: an accidental cancellation of the spin, dipole, and orbital
138: contributions of the Fermi-level electrons to $K$ which leaves only
139: superconductivity-insensitive contributions such as the Van Vleck
140: susceptibility. However, a quantitative analysis~\cite{Pavarini2005:P} shows
141: that the spin-orbit coupling in Sr$_{2}$RuO$_{4}$ is too weak for
142: scenario (a) while the accidential cancellation, required for scenario (b),
143: does not occur~\cite{Knight}.
144: Thus, neither of the two explanations of a constant $K$ arising from
145: singlet pairing is applicable.
146:
147: This would have made the Knight shift argument for C{\it p}W very
148: convincing, if not for the recent
149: experiment showing the same result in a field perpendicular to the
150: plane~\cite{Murukawa2004:PRL}. It was proposed~\cite{Murukawa2004:PRL}
151: that the testing field of
152: 0.02 T may be enough to induce a phase transition from the C\textit{p}W
153: in Eq.~(\ref{eq:CpW})
154: to a state with
155: \textbf{d}$\Vert\mathbf{\hat{x}}$. However, this is highly unlikely: (i)
156: the \textbf{d}$\Vert\mathbf{\hat{x}}$ state would have an additional
157: horizontal line node, as compared to the
158: \textbf{d}$\propto(k_{x}+ik_{y})\mathbf{\hat{z}}$ state and therefore lose
159: a large part of the pairing energy ( $\sim$ $\Delta$ per electron,
160: $\Delta\gtrsim1.4$ K $\gg\mu_{B}\times$0.02 T);
161: (ii) although in the \textbf{d}$\Vert\mathbf{\hat{x}}$ state
162: the spins of the pairs lie in the $yz$ plane, there is no $y-z$ symmetry (as
163: opposed to the $xy$ plane) and it is not $a$ $priori$ clear whether the
164: magnetic susceptibility of the Cooper pair will be the same as for the normal
165: electrons. Since \textbf{d}$\Vert\mathbf{\hat{x}}$ is
166: not allowed for a tetragonal symmetry,
167: it may only appear as a result of a second phase transition below
168: $T_{c}$, which has never been observed in Sr$_{2}$RuO$_{4}$; (iii) the
169: spin-orbit part of the pairing interaction, which keeps the spins in the $xy$
170: plane, despite $z$ being the easy magnetization axis~\cite{Maeno1997:JPSJ},
171: would have to be weaker than $\mu_{B}\times$0.02 T=1.1 $\mu$eV=0.013 K, an
172: energy scale much too small for the spin-orbit coupling in Sr$_{2}$RuO$_{4}$.
173: So, neither the old theories for the lack of a Knight shift reduction below
174: $T_{c}$, nor the new explanation in terms of the magnetic-field rotated order
175: parameter withstand quantitative scrutiny;
176: the Knight shift in Sr$_{2}$RuO$_{4}$
177: remains a challenge for theorists. Until this puzzle is resolved, we cannot
178: use the Knight shift argument for the pairing symmetry determination.
179:
180: We now turn to the experiments of Ref.~\cite{Nelson2004:S} and compare
181: Josephson tunneling between an $s$-wave superconductor and either (a) an even
182: parity (spin singlet) superconductor or (b) an odd parity (triplet)
183: superconductor. In the first case, the Josephson current between a
184: conventional $s$-wave superconductor and an unconventional spin-singlet
185: superconductor, represented by the order parameters $\Delta_{s-\mathrm{wave}}$
186: and $\Delta_{0}(\mathbf{k})$, respectively, can be expressed
187: as~\cite{Geshkenbein1986:PZETF}
188: \begin{equation}
189: J \propto\left\langle T_{\mathbf{k}}\mathrm{Im}[\Delta
190: _{s-\mathrm{wave}}^{\ast}\Delta_{0}(\mathbf{k})]\right\rangle _{FS},
191: \label{eq:jofk}%
192: \end{equation}
193: which depends on the relative phase between the superconducting
194: order parameters.
195: The averaging is over all states at the Fermi surface (FS) where the Fermi
196: velocity, $\mathbf{v}_{F}$, has a positive projection on the tunneling
197: direction represented by the unit normal \textbf{n} (perpendicular to the
198: interface plane, see Fig.~\ref{fig:1}) and $T_{\mathbf{k}}$ is the tunneling
199: probability. For a thick rectangular barrier of width $w$ and height
200: $U$~\cite{Mazin2001:EPL} we can obtain
201: \begin{equation}
202: T_{\mathbf{k}}=\frac{16m^{2}\kappa^{2}v_{L}v_{R}\exp[-2\kappa w-k_{\Vert}%
203: ^{2}w/\kappa]}{(\kappa^{2}+m^{2}v_{L}^{2})(\kappa^{2}+m^{2}v_{R}^{2})},
204: \label{eq:tkthick}%
205: \end{equation}
206: where $\kappa=\sqrt{2m(U-\mu)}$
207: such that $w\kappa\gg1$ (in the thick barrier limit), $m$ is the free-electron
208: mass, $\mu$ is the Fermi energy, and we set $\hbar=1$. We use $v_{L,R}$ to
209: denote normal components of the Fermi velocities in the two superconductors
210: and $k_{\Vert}$ is the component parallel to the interface. From
211: Eq.~(\ref{eq:tkthick}) we see that $T_{\mathbf{k}}$ is sharply peaked when
212: $\mathbf{v}_{F}\parallel\mathbf{n}$.
213:
214: In the second case, the Josephson current between a singlet and a triplet
215: superconductor becomes~\cite{Geshkenbein1986:PZETF,Millis1988:PRB}
216: \begin{equation}
217: J\propto\left\langle \tilde{T}_{\mathbf{k}}\mathrm{Im}[\Delta
218: _{0}^{\ast}(\mathbf{k})\mathbf{d(k)}\cdot(\mathbf{n}\times\mathbf{k}%
219: )]\right\rangle _{FS}, \label{eq:MRS}%
220: \end{equation}
221: where we use $\tilde{T}_{\mathbf{k}}$ to denote that, unlike
222: $T_{\mathbf{k}}$, it contains matrix elements
223: corresponding to spin-flip tunneling,
224: for example, due to magnetic interfaces or spin-orbit coupling. For
225: nonmagnetic barriers and in the absence of spin-orbit,
226: there is no spin-flip scattering, therefore
227: $\tilde{T}_{\mathbf{k}%
228: }=0$ and the Josephson current vanishes
229: identically~\cite{Geshkenbein1986:PZETF,Millis1988:PRB,Pals1977:PRB}.
230:
231: \begin{figure}[ptb]
232: \caption{ Schematic sample and the Fermi surface geometry for phase-sensitive
233: SQUID measurements from Ref.~\cite{Nelson2004:S}. (a) Au$_{0.5}$In$_{0.5}%
234: $/Sr$_{2}$RuO$_{4}$ junction geometry with an interface normal $\mathbf{n}$.
235: (b) Possible deviation of $\mathbf{n}$ from the $ab$ crystallographic plane.
236: (c) Warping of the Sr$_{2}$RuO$_{4}$ Fermi surface. The magnitude of the Fermi
237: wavevector $\mathbf{k}_{F}$ is generally different from the one corresponding
238: to the cylindrical Fermi surface ($k_{F0})$. }%
239: \label{fig:1}
240: \centerline{\psfig{file=sro0.eps,width=1\linewidth,angle=0}}\end{figure}From
241: Eqs.~(\ref{eq:jofk}) and (\ref{eq:MRS}) we can directly infer that for c-axis
242: tunneling, $\mathbf{n}\parallel\mathbf{c}$, $J=0$ for both
243: C\textit{d}W and C\textit{p}W states [$\int dk_{x}dk_{y}(k_{x}+ik_{y})=0$].
244: For tunneling \textit{precisely} in the $ab$ plane, the current is also zero
245: for C\textit{d}W while for C\textit{p}W it only vanishes at $\mathbf{n}%
246: \parallel\mathbf{k}$.
247:
248:
249: We consider a model of a quasi-two-dimensional (2D) layered superconductor
250: which has
251: a nearly cylindrical FS with a small $c$-axis dispersion originating from the
252: weak inter-layer hopping. In Fig.~\ref{fig:1}(a) we show the sample geometry
253: used Ref.~\cite{Nelson2004:S} and in Fig.~\ref{fig:1}(b)
254: represent a warping of the Fermi surface. In the first approximation for
255: Sr$_{2}$RuO$_{4}$ such a warping can be expressed as~\cite{Bergemann2003:AP}
256: \begin{equation}
257: \mu=\frac{k_{F}^{2}(z)}{2m}[1+\varepsilon\cos k_{z} c], \label{eq:mu}%
258: \end{equation}
259: where $|\varepsilon| \ll1$ is the warping parameter ($\varepsilon\approx-7
260: \times10^{-4}$ \cite{Bergemann2003:AP}), $c$ is the lattice constant along the
261: crystallographic $z$-direction, and $\mathbf{k}_{F}(z)$ is the $z$-dependent
262: projection of the Fermi wavevector in the $xy$ plane [see Fig.~\ref{fig:1}%
263: (c)]. It is convenient to resolve the Fermi wavevector in cylindrical
264: coordinates ($k_{F}(z)$, $\varphi$, $k_{z}$) with
265: \begin{equation}
266: k_{F}(z)=k_{F0}/[1+\varepsilon\cos k_{z} c ]^{1/2}, \label{eq:kF}%
267: \end{equation}
268: where $k_{F0}=(2m \mu)^{1/2}$ and for $\varphi=0$ [see Fig.~\ref{fig:1}(c)]
269: $k_{F}(z)\rightarrow k_{Fx}$.
270:
271: Schematic representation of the SQUID geometry in Fig.~\ref{fig:1}(a) (adapted
272: from Ref.~\cite{Nelson2004:S}) is an oversimplification. While efforts were
273: made to fabricate edges either precisely parallel or perpendicular to the
274: $c$-axis, in the actual samples the direction of interface planes or their
275: corresponding normals changes gradually from $a$ to $c$-direction. In several
276: samples~\cite{Nelson2004:S} an interface nearly parallel to the $ab$ plane, at
277: Au$_{0.5}$In$_{0.5}$/Sr$_{2}$RuO$_{4}$ junction, was covered by an insulating
278: oxide [see Fig.~\ref{fig:1}(a)]. It is then plausible to expect that the
279: normal to such interface could deviate from the $ab$ plane. In
280: Fig.~\ref{fig:1}(b) we depict a generalized situation in which an interface
281: normal, $\mathbf{n}=(n_{\rho},n_{\varphi},n_{z})$ with $|n_{z}|\ll1$, need not
282: lie exactly in the crystallographic $ab$ plane of Sr$_{2}$RuO$_{4}$. We show
283: below that the analysis of phase-sensitive measurements in terms of the two
284: small but \emph{finite} parameters, $\varepsilon$ and $n_{z}$, can provide a
285: qualitatively different interpretation from those which a priori assume
286: $\varepsilon=n_{z}\equiv0$.
287:
288: For a conventional superconductor with the FS larger than the one of Sr$_{2}%
289: $RuO$_{4}$, the Josephson tunneling across a thick rectangular barrier
290: can be
291: obtained from Eqs.~(\ref{eq:jofk}) and (\ref{eq:tkthick}) as
292: \begin{align}
293: J & \propto\int_{\mathbf{v}_{F}\cdot{\mathbf{n}}>0}%
294: d\mathbf{k}\delta(\epsilon_{\mathbf{k}}-\mu)\mathbf{v}_{F}\mathbf{\cdot n}%
295: \exp(-k_{\parallel}^{2}w/2\kappa)\nonumber\\
296: & \times\mathrm{Im}(k_{x}+ik_{y})\sin k_{z}c, \label{eq:jk}%
297: \end{align}
298: where $k_{\parallel}^{2}=\mathbf{k}_{F}^{2}-(\mathbf{k}_{F}\cdot
299: \mathbf{n})^{2}$, $\kappa^2 \gg m v_{L,R}^2$,
300: and the projection of the Fermi velocity in Sr$_{2}$RuO$_{4}$
301: along $\mathbf{n}$ is
302: \begin{equation}
303: \mathbf{v}_{F}\mathbf{\cdot n}=\frac{k_{F0}[1+\varepsilon\cos k_{z}c]^{1/2}%
304: }{m}n_{\rho}-\frac{k_{F0}^{2}\varepsilon c\sin k_{z}c}{2m[1+\varepsilon\cos
305: k_{z}c]}n_{z}. \label{eq:vf}%
306: \end{equation}
307: For a thick barrier, the integration can be simplified by noting that the
308: dominant contribution comes from $k_{\parallel}=0$. The right hand side of
309: Eq.~(\ref{eq:jk}), in the leading order in $\varepsilon$ and $n_{z}$,
310: can be then reduced to
311: $\sqrt{\pi\kappa/w}ck_{F0}^{2}n_{z}(1-\varepsilon)$, such that
312: \begin{equation}
313: J_{\: \sqcap}=A \: n_{z}(1-\varepsilon), \label{eq:thick}%
314: \end{equation}
315: where $A$ characterizes the normal state barrier transparency. Thus, with a
316: tilted interface ($n_{z}\neq0$) there is a finite Josephson current even in
317: the absence of any Fermi surface warping ($\varepsilon=0$).
318:
319: To verify that our findings of finite Josephson current in C\textit{d}W state
320: are not limited to the specific assumption of a thick rectangular barrier, we
321: also consider the rather different case of a strong $\delta$-function
322: barrier. The corresponding transmission probability
323: is~\cite{Zutic1999:PRB,Mazin2001:EPL}
324: \begin{equation}
325: T_{\mathbf{k}}=\frac{4v_{L}v_{R}}{(v_{L}+v_{R})^{2}+4U^{2}},\label{eq:tkdelta}%
326: \end{equation}
327: where $v_{L,R}$ are the normal components of the Fermi velocities in the two
328: superconductors and $U$ ($\gg v_{L,R}^{2}$) is the scattering strength.
329: From Eqs.~(\ref{eq:jofk}),~(\ref{eq:tkdelta}) we obtain
330: \begin{equation}
331: J\propto\int_{\mathbf{v}_{F}\mathbf{\cdot n}>0}d\mathbf{k}%
332: \delta(\epsilon_{\mathbf{k}}-\mu)\mathbf{v}_{F}\mathbf{\cdot n} \: \mathrm{Im}%
333: (k_{x}+ik_{y})\sin k_{z}c,\label{eq:jofk3}%
334: \end{equation}
335: where, unlike in the case of a thick barrier, we perform
336: $\varphi$ and $k_{z}$ integration.
337: In the leading order, the right hand side of Eq.~(\ref{eq:jofk3}) is $-\pi
338: k_{F0}^{2}\varepsilon n_{z},$ and yields
339: \begin{equation}
340: J_\delta =-A \: \varepsilon n_{z},\label{eq:delta}%
341: \end{equation}
342: where again $A$ characterizes the normal state transparency. In contrast to
343: the thick-barrier result, the current now vanishes in the absence of FS
344: warping. From Eqs.~(\ref{eq:thick}),~(\ref{eq:delta}) one can conjecture
345: that for a
346: general case $J\approx A(s-\varepsilon)$, where $0\lesssim s\lesssim1$.
347:
348: The presence of small parameters $\varepsilon$ and $n_{z}$ in
349: Eqs.~(\ref{eq:thick}),~(\ref{eq:delta}) shows that the Josephson current in the
350: C\textit{d}W state would be reduced as compared to the conventional SQUID with
351: $s$-wave electrodes. However, the alternative picture, based on the
352: C\textit{p}W state, also contains small parameters which should be kept in mind
353: when interpreting the experiment of Ref.~\cite{Nelson2004:S}. In addition to
354: the small relative strength of the spin-orbit coupling (quantified by the
355: admixture of $S_{\downarrow}$ into a nonrelativistic $S_{\uparrow}$ state,
356: or the spin-orbit induced band shift relative to the band width
357: \cite{small}), there can also be another small factor --- a ratio of the
358: lattice constant and the superconducting coherence length~\cite{Fenton1985:SSC},
359: approximately $6\times10^{-3}$ \cite{Mackenzie2003:RMP}.
360:
361: Results from Eq.~(\ref{eq:thick}),~(\ref{eq:delta}) confirm that the
362: C\textit{d}W
363: state could be compatible with the phase shifts observed in
364: Ref.~\cite{Nelson2004:S}. Furthermore, the azimuthal dependence of an order
365: parameter coincides for both C\textit{d}W and C\textit{p}W states.
366: While the proposed symmetry
367: arguments~\cite{Nelson2004:S,Rice2004:S} exclude most of superconducting
368: states allowed in the tetragonal symmetry~\cite{Asano2005:P},
369: these arguments alone are not
370: sufficient to unambiguously identify the odd-pairing of C\textit{p}W state.
371: Instead, to confirm that a C\textit{p}W state has indeed been observed, one
372: would need to accurately calculate the expected magnitudes of the Josephson
373: current for both chiral states. In particular we propose a modification of
374: the experimental configuration~\cite{Nelson2004:S} such that the
375: interface plane would be slanted at $\approx45^{o}$ with the $c$-axis. If the
376: corresponding ratio of the Josephson current to the normal state conductivity
377: becomes substantially larger ($n_{z}$ is no longer small) than in
378: Ref.~\cite{Nelson2004:S},
379: it would be strong support for the chiral singlet state in
380: Eq.~(\ref{eq:CdW}).
381:
382:
383: Another important distinction between the two
384: chiral states is the presence of nodes in the superconducting gap.
385: In contrast to the C{\it p}W state, C{\it d}W requires by symmetry
386: a horizontal line node
387: [see Eqs.~(\ref{eq:CpW}),~(\ref{eq:CdW})].
388: The idea of a horizontal line node~\cite{Mackenzie2003:RMP} has
389: been entertained by experimentalists~\cite{Tanatar2001:PRL} and
390: theorists~\cite{Zhitomirsky2001:PRL}
391: for a while, although recently it has fallen
392: out of favor. Still, some researchers insist on the existence of a horizontal
393: line node~\cite{Contreras2004:PRB}. Moreover, in the Josephson experiments of
394: Ref.~\cite{Nelson2004:S} evidence was found for a substantial $k_{z}$
395: dependence, albeit not necessarily for horizontal nodes,
396: of the order parameter in
397: Sr$_{2}$RuO$_{4}$~\cite{Liu2005:PC}. How could such a material with nearly 2D
398: electronic structure develop a highly 3D superconducting state? To answer
399: this question we point out the following facts: (a) practically no
400: ferromagnetic spin fluctuations, favorable for a $p$-wave pairing, have been
401: experimentally found in Sr$_{2}$RuO$_{4}$; (b) antiferromagnetic spin
402: fluctuations at \textbf{q}$=(2/3,2/3,q_{z})$ have negligible $z$
403: dispersion; (c) the crystal structure of Sr$_{2}$RuO$_{4}$, as opposed to its
404: electronic structure, is fairly 3D, so one can expect the electron-phonon
405: coupling to be quite 3D as well (d) there is a sizeable O isotope effect in
406: Sr$_{2}$RuO$_{4}$, which strongly changes with introduction of pair-breaking
407: symmetries~\cite{Mao2001:PRB}. While electron-phonon coupling $per$ $se$
408: can only induce an $s$-wave pairing, such a pairing would be prevented by
409: the strong antiferromagnetic spin fluctuations. However, for the
410: proposed C\textit{d}W
411: state, any 2D interaction cancels out, including the magnetic
412: interaction. Should the electron-phonon coupling have a maximum say, at
413: $(1/2,1/2,1/2)$, as opposed to $(1/2,1/2,0)$, the C\textit{d}W state would
414: have been immediately stabilized providing a plausible scenario for
415: spin-singlet superconductivity in Sr$_{2}$RuO$_{4}$.
416:
417: In conclusion, we have revealed that a completely overlooked chiral $d$-wave
418: pairing state in Sr$_{2}$RuO$_{4}$
419: is equally compatible with the existing body of experimental data as
420: the generally accepted chiral $p$-wave state.
421: We have proposed phase-sensitive experiments in a SQUID geometry
422: with a variable tilting angle capable of unambiguously distinguishing
423: between the two chiral states.
424:
425: We thank R. A. Klemm, Y. Liu, and D. J. Van Harlingen for useful discussions.
426: This work was supported by the US ONR. I. \v{Z}. acknowledges financial
427: support from the National Research Council.
428:
429:
430: \begin{references}
431: \bibitem{unconv}
432: In an unconventional superconductor, often synonymous to
433: a non $s$-wave orbital symmetry, there is at least one more
434: symmetry broken in addition to the gauge symmetry.
435:
436: \bibitem{Wollman1993:PRL}
437: D. A. Wollman {\it et al.},
438: Phys. Rev. Lett. {\bf 71}, 2134 (1993);
439: D. J. {Van Harlingen}, Rev. Mod. Phys. {\bf 67}, 515 (1995).
440:
441: \bibitem{Wei1998:PRL}
442: J. Y. T. Wei {\it et al.}, Phys. Rev. Lett. {\bf 81}, 2542 (1998);
443: C. C. Tsuei and J. R. Kirtley, Rev. Mod. Phys. {\bf 72}, 969 (2000);
444: S. Kashiwaya and Y. Tanaka, Rep. Prog. Phys. {\bf 63}, 1671 (2000)i;
445: G. Deutsher, Rev. Mod. Phys. {\bf 77}, 109 (2005).
446: \bibitem{Klemm} There still remain views which suggest other
447: symmetries [R. A. Klemm, Phil. Mag. {\bf 85}, 801 (2005)].
448: %\bibitem{Thalmeier2004:P}
449: %P. Thalmeier
450: %{\it et al.},
451: %cond-mat/0409363.
452: %\bibitem{Ishiguro:1998}
453: %T. Ishiguro, K. Yamaji, and G. Saito, {\it Organic Superconductors}
454: %(Springer, Berlin, 1998).
455: %\bibitem{Sengupta2001:PRB}
456: %K. Sengupta {\it et al.},
457: %Phys. Rev. B {\bf 63}, 144531 (2001).
458: \bibitem{Maeno1994:N}
459: Y. Maeno {\it et al.},
460: Nature {\bf 372}, 532 (1994).
461: \bibitem{Mackenzie2003:RMP}
462: A. P. Mackenzie and Y. Maeno, Rev. Mod. Phys. {\bf 75}, 657 (2003).
463: \bibitem{Agterberg1997:PRL}
464: T. M. Rice and M. Sigrist, J. Phys.: Condens. Matter {\bf 7}, L643 (1995);
465: D. F. Agterberg, T. M. Rice, and M. Sigrist, Phys. Rev. Lett. {\bf 78}, 3374 (1997).
466: \bibitem{Nelson2004:S}
467: K. D. Nelson {\it et al.}, % Z. Q. Mao, Y. Maeno, and Y. Liu,
468: Science {\bf 306}, 1151 (2004).
469: \bibitem{Rice2004:S}
470: T. M. Rice, Science {\bf 306}, 1142 (2004).
471: \bibitem{Geshkenbein1987:PRB}
472: V. B. Geshkenbein, A. I. Larkin, and A. Barone,
473: Phys. Rev. B {\bf 36}, 235 (1987).
474: \bibitem{typo}
475: We correct a typo in the prefactor of $\delta_{12}$
476: from Ref.~\protect{\cite{Wollman1993:PRL}}.
477: \bibitem{pi} There are also alternative explanations
478: for the origin of $\delta_{12}=\pi$ or so called $\pi$-junctions in
479: magnetic [A. I. Buzdin, cond-mat/0505583]
480: and nomagnetic geometries
481: [J. J. A. Baselmans, Phys. Rev. Lett. {\bf 89}. 207002 (2002)].
482: \bibitem{Ishida1998:N}
483: K. {Ishida \it et al.}, Nature {\bf 396}, 658 (1998).
484: \bibitem{Annet1990:AP}
485: J. F. Annet, Adv. Phys. {\bf 39}, 83 (1990).
486: \bibitem{T-note}Note that neither scenario can fully explain the data of Ref.
487: \protect{\cite{Nelson2004:S}}; for instance, the fact that phase shift
488: depends on temperature and is $\pi$ only at $T\simeq T_c$ presents
489: a challenge to any theory.
490: \bibitem{Luke1998:N}
491: G. M. {Luke \it et al.}, Nature {\bf 394}, 558 (1998).
492: \bibitem{Pavarini2005:P}
493: E. Pavarini and I. I. Mazin, to be published (2005).
494: \bibitem{Knight}
495: In fact, the Knight shift appears to be dominated by
496: the core-polarization part of the spin contribution
497: \protect{\cite{Pavarini2005:P}}.
498: \bibitem{Murukawa2004:PRL}
499: H. Murukawa {\it et al.},
500: Phys. Rev. Lett. {\bf 93}, 167004 (2004).
501: \bibitem{Maeno1997:JPSJ}
502: Y. Maeno {\it et al.},
503: J. Phys. Soc. Jap. {\bf 66}, 1405 (1997).
504: \bibitem{Geshkenbein1986:PZETF}
505: V. B. Geshkenbein and A. I. Larkin, Pis'ma Zh. Eksp. Teor. Fiz.
506: {\bf 43}, 306 (1986).
507: \bibitem{Mazin2001:EPL}
508: I. I. Mazin, Europhys. Lett. {\bf 55}, 404 (2001).
509: \bibitem{Millis1988:PRB}
510: A. Millis, D. Rainer, and J. A. Sauls,
511: Phys. Rev. B {\bf 38}, 4504 (1988).
512: \bibitem{Pals1977:PRB}
513: J. A. Pals, W. {van Haeringen}, and M. H. {van Maaren},
514: Phys. Rev. B {\bf 15}, 2592 (1977).
515: \bibitem{Bergemann2003:AP}
516: C. Bergemann {\it et al.},
517: Adv. Phys. {\bf 52}, 639 (2003).
518: The Fermi surface of Sr$_2$RuO$_4$ has three ($\alpha$, $\beta$, $\gamma$)
519: sheets. Equation~(7) holds for the $\gamma$-sheet; for the
520: $\beta$-sheet the leading warping term is $\propto \cos (k_z c/2)$ and
521: its magnitude is about 7 times larger.
522: \bibitem{Zutic1999:PRB}
523: I. \v{Z}uti\'c and S. {Das Sarma},
524: Phys. Rev. B {\bf 60}, R16 322 (1999).
525: \bibitem{small}
526: Which can be estimated from band structure calculations in
527: Ref.~\protect{\cite{Pavarini2005:P}} to be of the order of $10^{-3}$.
528: \bibitem{Fenton1985:SSC} E. W. Fenton, Solid State Commun. {\bf 54}, 709 (1985).
529: \bibitem{Asano2005:P}
530: For example, from the phase shifts alone it is easy to rule
531: out the chiral $d_{x^2-y^2}+id_{xy}$
532: state [Y. Asano {\it et al.}, preprint, cond-mat/0502082].
533: \bibitem{Tanatar2001:PRL}
534: M. A. Tanatar {\it et al.},
535: Phys. Rev. Lett. {\bf 86}, 2649 (2001);
536: K. Izawa {\it et al.},
537: Phys. Rev. Lett. {\bf 86}, 2653 (2001).
538: \bibitem{Zhitomirsky2001:PRL}
539: M. E. Zhitomirsky and T. M. Rice,
540: Phys. Rev. Lett. {\bf 87}, 057001 (2001).
541: \bibitem{Contreras2004:PRB}
542: P. Contreras, M. Walker, and K. Samokhin,
543: Phys. Rev. B {\bf 70}, 184528 (2004).
544: \bibitem{Liu2005:PC}
545: Y. Liu, private communication.
546: \bibitem{Mao2001:PRB} Z. Q. {Mao \it et al.},
547: Phys. Rev. B {\bf 63}, 144514 (2001).
548: \end{references}
549:
550: \end{document}
551: