1: \documentclass{edp-jp4}
2: \usepackage{graphicx}
3: \begin{document}
4: \newcommand {\bb}{\bibitem}
5: \newcommand {\be}{\begin{equation}}
6: \newcommand {\ee}{\end{equation}}
7: \newcommand {\bea}{\begin{eqnarray}}
8: \newcommand {\eea}{\end{eqnarray}}
9: \newcommand {\nn}{\nonumber}
10:
11: \title{Scaling Relations in the Vortex State of Nodal Superconductors}
12:
13:
14: \author{Kazumi Maki}
15: \address{Max-Planck Institute for the Physics of Complex Systems,
16: N\"{o}thnitzer Str. 38, D-01187, Dresden, Germany}
17: \secondaddress{Department of Physics and Astronomy, University of Southern
18: California, Los Angeles, CA 90089-0484 USA}
19: \author{David Parker}
20: \sameaddress{1,2}
21: \author{Hyekyung Won}
22: \sameaddress{1}
23: \secondaddress{Department of Physics, Hallym University, Chuncheon 200-702, South Korea}
24:
25:
26: \date{\today}
27: \maketitle
28: \begin{abstract}
29:
30: In contrast to multigap superconductors (e.g. MgB$_{2}$), the low-temperature
31: properties of nodal superconductors are dominated by nodal excitations.
32: Here we extend for a variety of nodal superocnductors the earlier work by
33: Simon and Lee and K\"ubert and Hirschfeld. The scaling relations seen in
34: the thermodynamics and the thermal conductivity will provide an unequivocal
35: test of nodal superconductivity.
36:
37:
38:
39:
40:
41:
42: \end{abstract}
43: \section{Introduction}
44:
45: Although nodal superconductors have been with us since 1979 \cite{1}, the
46: systematic study of the gap symmetry of these new superconductors began
47: only around 1994 with the establishment of
48: d-wave symmetry of high-T$_{c}$ cuprate superconductors
49: through the angle resolved photoemission spectrum (ARPES) \cite{2}
50: and Josephson interferometry \cite{3,4}. Unfortunately, however, these
51: powerful techniques have not been applied to other nodal superconductors
52: like Sr$_{2}$RuO$_{4}$, heavy-fermion superconductors and organic
53: superconductors.
54:
55: Since 2001 Izawa et al have succeeded in determining the gap functions
56: $|\Delta({\bf k})|$'s in Sr$_{2}$RuO$_{4}$ \cite{5}, CeCoIn$_{5}$ \cite{6},
57: $\kappa$-(ET)$_{2}$Cu(NCS)$_{2}$ \cite{7}, YNi$_{2}$B$_{2}$C \cite{8},
58: PrOs$_{4}$Sb$_{12}$, \cite{9,10}, and UPd$_{2}$Al$_{3}$ \cite{11,12}
59: through measurements
60: of the angle-dependent thermal conductivity in the vortex state. These
61: experiments are only possible now since a) high-quality single crystals
62: of these compounds are now available, b) low-temperature facilities which
63: allow one to reach 1 - 0.1 K are available, and c)the necessary theoretical
64: development following the seminal paper by Volovik \cite{13}.
65:
66: Indeed Volovik's approach has been extended in a variety of directions, as
67: reviewed in \cite{14}. Also the angle dependent magentothermal conductivity
68: and the scaling relations \cite{15} in the vortex state will provide a
69: crucial test of nodal superconductivity. For example, the multigap
70: superconductors do not exhibit the scaling relations we are going to discuss
71: in general. Therefore, if any given superconductor exhibits a scaling
72: relation discussed here, it is very likely that the material is a nodal
73: superconductor. For example, the specific heat data of Sr$_{2}$RuO$_{4}$
74: by Deguchi et al \cite{15} obeys the scaling relation given in \cite{16}.
75: Therefore the simplest choice of gap function in Sr$_{2}$RuO$_{4}$ is
76: the chiral f-wave superconductor as pointed out in \cite{5}.
77:
78: The scaling relations in the vortex state in d-wave superconductors were
79: first proposed by Simon and Lee \cite{17}. Then within the semiclassical
80: approximation, \`a la Volovik \cite{13} K\"ubert and Hirschfeld (KH)
81: \cite{18} have succeeded in deriving the scaling function for the
82: quasiparticle density of states. KH then calculated the thermal conductivity
83: in the scaling region \cite{19}. An error in \cite{19}
84: was pointed out and corrected in \cite{20}. However, in \cite{20} only the
85: asymptotic behavior of the thermal conductivity
86: ($T \ll <|{\bf v} \cdot {\bf q}|>$, where ${\bf v} \cdot {\bf q}$ is the
87: Doppler shift) has been worked out.
88:
89: In the following we shall derive the scaling relations for a class
90: of quasi-2D superconductors, where $|\Delta({\bf k})| = \Delta |f|$ and
91: f = $\cos(2\phi), \sin(2\phi)$ (d-wave superconductor), $f = e^{\pm i\phi}\cos
92: \chi$ (chiral f-wave superconductor as in Sr$_{2}$RuO$_{4}$), $f = \cos(2\chi)$
93: (g-wave superconductor as in UPd$_{2}$Al$_{3}$ \cite{12}.)
94: These superconductors have
95: the same quasiparticle density of states as in d-wave superconductors
96: \cite{21}
97: \bea
98: N(E)/N_{0} &=& G(x)
99: \eea
100: where
101: \bea
102: G(x)&=& \frac{2x}{\pi}K(x)\,\,\,\mathrm{for\,\, x \leq 1} \\
103: &=& \frac{2}{\pi}K(x^{-1})\,\,\,\mathrm{for\,\, x > 1}.
104: \eea
105: where $x=|E|/\Delta$ and K(k) is the complete elliptic integral of the
106: second kind. In particular for $|E| < 0.3 \Delta $ we have $G(E/\Delta) =
107: \frac{|E|}{\Delta}$.
108:
109: As discussed elsewhere \cite{14}, all the nodal superconductors so far
110: discovered have G(E) $\sim |E|/\Delta$. Then one can establish a variety
111: of scaling relations in the superclean limit \cite{20}, that is, for
112: $(\Gamma \Delta)^{1/2} < T,E;$ ${\bf v} \cdot {\bf q} < \Delta$ where
113: T, E, $\Delta$ and $\Gamma$ are the temperature, the quasiparticle energy,
114: the maximal value of the energy gap and the quasiparticle scattering rate
115: respectively. Therefore the scaling relations provide another test for
116: nodal superconductivity.
117:
118: \section{Quasiparticle density of states}
119:
120: Let us limit ourselves to a class of quasi-2D systems with f listed in the
121: preceding section. As already noted we have $G(E) \simeq |E|/\Delta$ for
122: $E \ll \Delta$. In the presence of a magnetic field we find
123: \bea
124: G(E,{\bf H}) &=& <|E-{\bf v} \cdot {\bf q}|>\Delta^{-1}
125: \eea
126: where ${\bf v} \cdot {\bf q}$ is the Doppler shift and $\langle
127: \ldots \rangle$ means the average
128: over the Fermi surface and the vortex lattice. When ${\bf H} \parallel
129: {\bf c}$ in the class of quasi-2D systems, the average can be performed
130: analytically and we find \cite{18}
131: \bea
132: G(E,{\bf H}) &=& \frac{4}{\pi}\frac{\epsilon}{\Delta}g(E/\epsilon)
133: \eea
134: where
135: \bea
136: g(s)&=& \frac{\pi}{4}s(1+\frac{1}{2s^{2}}), s > 1 \\
137: &=& \frac{3}{4}\sqrt{1-s^{2}}+\frac{1}{4s}(1+2s^{2})\arcsin(s),
138: \ \ \ \ s \leq 1
139: \eea
140: where $\epsilon = \frac{1}{2}v\sqrt{eH}$ and v is the Fermi velocity
141: within the ab plane. The scaling function
142: $\frac{\Delta}{\epsilon}G(E,{\bf H})$ is shown in Fig. 1. As is readily
143: \begin{figure}[h]
144: \includegraphics[width=8cm]{fig1_car.eps}
145: \caption{The scaling function G(E, {\bf H})}
146: \end{figure}
147: seen $G(E,{\bf H})$ for ${\bf H} \parallel {\bf c}$ cannot discriminate
148: between different $|\Delta({\bf k})|$'s in the above class of nodal
149: superconductors. Then the scaling function for the specific heat is given
150: by \cite{16}
151: \bea
152: C_{s}(T,{\bf H})/C_{s}(T,0) &=& F(T/\epsilon)
153: \eea
154: where
155: \bea
156: F(T/\epsilon) &=& \frac{2}{9\pi \zeta(3)}(\frac{\epsilon}{T})^{2}
157: \int_{0}^{\infty}
158: ds \, s^{2}g(s) \mathrm{sech}^{2}\left(\frac{\epsilon s}{2T}\right) \\
159: & \simeq & 1 + \frac{\ln 2}{9 \zeta(3)}(\frac{\epsilon}{T})^{2},
160: \mathrm{for \,\, \epsilon/T \leq 1} \\
161: & \simeq & \frac{4\epsilon}{9\pi \zeta(3) T}[1 +
162: \frac{1}{18}(\frac{\pi T}{\epsilon})^{2}+ \frac{7}{1800}
163: (\frac{\pi T}{\epsilon})^{4} + \ldots], \mathrm{for \,\, \epsilon/T > 1}
164: \eea
165: The scaling function and the experimental data for Sr$_{2}$RuO$_{4}$
166: \cite{15} are shown in Fig. 2. As is seen readily the scaling function gives
167: an excellent description of the experimental data. As noted in \cite{15},
168: \begin{figure}[h]
169: \includegraphics[width=8cm]{C_scale.eps}
170: \caption{The scaling function $F(T/\epsilon) = F(x)$ and Sr$_{2}$RuO$_{4}$
171: specific heat data from Ref. \cite{15}.}
172: \end{figure}
173: this clearly shows the superconductivity in Sr$_{2}$RuO$_{4}$ is consistent
174: with the chiral f-wave superconductor as discussed in \cite{23}. On the
175: other hand, as noted in \cite{24}, this is incompatible with
176: p-wave superconductivity.
177:
178: Now when {\bf H} is rotated within the a-b plane with an angle $\phi$ from the
179: a axis, we can discriminate between $f=\cos(2\phi)$ and $f=\sin(2\phi)$ (the
180: case of vertical nodes). We obtain for $f=\cos(2\phi)$
181: \bea
182: G(E,{\bf H}) & = &\frac{2}{\pi} \sum_{\pm}\left\langle
183: \frac{\epsilon_{\pm}(\phi,\chi)}
184: {\Delta}G(\frac{E}{\epsilon_{\pm}(\phi,\chi)}\right\rangle \\
185: & = & \frac{E}{\Delta}\left(1+ \frac{\epsilon^{2}}{2E^{2}}\right),
186: \mathrm{for\,\,
187: \frac{\epsilon}{E} < 1}
188: \eea
189: \bea
190: = (\frac{2}{\pi})^{2}\frac{\epsilon}{\Delta}\sum_{\pm}(\sqrt{\frac{3}{2}
191: \pm \frac{1}{2} \sin(2\phi)} E\left(\frac{1}{\sqrt{\frac{3}{2} \pm
192: \frac{1}{2} \sin(2\phi)}}\right) + \nonumber \\
193: \frac{1}{6}(\frac{E}{\epsilon})^{2}
194: \left(\frac{1}{\sqrt{\frac{3}{2} \pm \frac{1}{2} \sin(2\phi)}}\right)K
195: \left(\frac{1}{\sqrt{\frac{3}{2} \pm \frac{1}{2}
196: \sin(2\phi)}}\right) + \ldots \\
197: \simeq \frac{4\epsilon}{\pi \Delta}(0.963+0.0205\cos(4\phi)+ \frac{1}{6}
198: (1.132 - 0.081 \cos(4\phi)) \times (\frac{E}{\epsilon})^{2} + \ldots)
199: \mathrm{for \,\,\frac{\epsilon}{E} > 1}
200: \eea
201: where $\epsilon=\frac{1}{2}\tilde{v}\sqrt{eH}$ and
202: $\tilde{v} = \sqrt{vv_{c}}$. For f=$\sin(2\phi)$ we have the same formulas
203: as in Eqs. (12) and (13), except that $\cos(4\phi)$ in Eq.(14) should be
204: changed to $-\cos(4\phi)$. Also the presence of the fourfold term in
205: the specific heat has been studied by Revaz et al \cite{25}. They found no
206: fourfold term within an accuracy of 3\%. This suggests strongly that the
207: thermal conductivity provides a more sensitive test of the gap symmetry.
208:
209: For superconductors with horizontal nodes (e.g. f = $\sin \chi, \cos(2\chi),
210: \cos \chi$) the field configuration ${\bf H} \parallel b-c$ plane, with
211: $\theta$ the angle ${\bf H}$ makes from the c-axis, is more appropriate. Then
212: we find \cite{12}
213: \bea
214: G(E,{\bf H}) &=& \frac{4}{\pi} \sum_{\pm}\left\langle \frac{\epsilon_{\pm}
215: (\theta,\phi,\chi)} {\Delta}G(\frac{E}{\epsilon_{\pm}(\theta,\phi,\chi)})\right\rangle \\
216: &=& \frac{E}{\Delta}(1+ \frac{\epsilon^{2}}{2E^{2}}), \mathrm{for\,\,
217: \frac{\epsilon}{E} < 1}
218: \eea
219: \bea
220: = (\frac{2}{\pi})^{2}\frac{\epsilon}{\Delta}\sum_{\pm}\left(\sqrt{\frac{3}{2}
221: \pm \frac{1}{2} \sin(2\phi)} E(\frac{1}{\sqrt{\frac{3}{2} \pm
222: \frac{1}{2} \sin(2\phi)}})\right) + \nonumber \\
223: \frac{1}{6}(\frac{E}{\epsilon})^{2}
224: \left(\frac{1}{\sqrt{\frac{3}{2} \pm \frac{1}{2} \sin(2\phi)}}\right)K
225: \left(\frac{1}{\sqrt{\frac{3}{2} \pm \frac{1}{2} \sin(2\phi)}}\right) \\
226: \eea
227: \bea
228: & \simeq & \frac{4\epsilon}{\pi \Delta}(0.963+0.0205\cos(4\phi)+ \frac{1}{6}
229: (1.132 - 0.081 \cos(4\phi))) \times (\frac{E}{\epsilon})^{2} + \ldots
230: \mathrm{for \,\,\frac{\epsilon}{E} > 1}
231: \eea
232: \bea
233: \simeq \frac{4\epsilon}{\pi \Delta}\sqrt{x/2}(1 -
234: \frac{1}{16}\sin^{2}\theta(\sin^{2}(\theta)+16\alpha^{2}\cos^{2}\theta
235: \sin^{2}\chi_{0})x^{-2} + \nn \\
236: \frac{1}{3}(\frac{E}{\epsilon})^{2}x^{-1}(1 +\frac{3}{16}\sin^{2}\theta(\sin^{2}(\theta)+16\alpha^{2}\cos^{2}\theta
237: \sin^{2}\chi_{0})x^{-2})) \mathrm{for \,\,\frac{\epsilon}{E} \geq 1}
238: \eea
239: where $\epsilon=v\sqrt{eH}$, $\alpha = v_{c}/v$ and
240: $x = 1 +\cos^{2}\theta + 2\alpha^{2}\sin^{2}\theta \sin^{2} \chi_{0}$ and
241: $\chi_{0} = 0, \frac{\pi}{4}$ and $\frac{\pi}{2}$ for $f = \sin \chi,
242: \cos(2\chi)$ and $\cos \chi$ respectively. Therefore in the present
243: configuration the angle dependent thermal conductivity can discriminate
244: different $\Delta({\bf k})$'s with horizontal nodes.
245:
246: So far we have completely ignored the effect of impurity scattering. As
247: already indicated the present analysis is valid in the superclean limit
248: \cite{14,20}, i.e. for $(\Gamma \Delta)^{1/2} < T,E$, ${\bf v} \cdot {\bf q} <
249: \Delta$ where $\Gamma$ is the quasiparticle scattering rate in the normal state. Then the superclean limit appears to require $\Gamma/\Delta \leq 0.01$.
250:
251: \section{Thermal conductivity}
252:
253: In the past few years the angle dependent magnetothermal conductivity (ADMTC)
254: has proven itself the most powerful technique to probe the nodal structure
255: of the gap function $\Delta({\bf k})$. Also in many cases the nodal
256: structure of $\Delta({\bf k})$ is sufficient to deduce $\Delta({\bf k})$
257: itself. We are concerned that much of the confusion and the
258: controversy in the literature
259: regarding the gap functions in Sr$_{2}$RuO$_{4}$, PrOs$_{4}$Sb$_{12}$ and
260: UPd$_{2}$Al$_{3}$ may be largely
261: due to a misunderstanding of Volovik's approach.
262: References \cite{14,20} contain a detailed description of this approach.
263: Generalizing the standard expression of the thermal conductivity given in
264: \cite{26,27}, the thermal conductivity of the class of nodal superconductors
265: in the vortex state is given by \cite{14}
266: \bea
267: \kappa_{zz} &=& \frac{n}{4mT^{2}}\int_{0}^{\infty}d \omega \, \omega^{2}
268: \left\langle\frac{h(\omega, {\bf H})}{\tilde{\Gamma}(\omega, {\bf H})}
269: \right\rangle \mathrm{sech}^{2}(\omega/2T)
270: \eea
271: where
272: \bea
273: h &=& \frac{1}{2}\left(1 + \frac{|\tilde{\omega}-{\bf v}\cdot{\bf q}|^{2}-
274: \Delta^{2}f^{2}}{|(\tilde{\omega}-{\bf v}\cdot{\bf q})^{2}-\Delta^{2}f^{2}|}
275: \right)
276: \eea
277: and
278: \bea
279: \tilde{\Gamma}&=& \mathrm{Im} \sqrt{(\tilde{\omega}-{\bf v}\cdot{\bf q})^{2}-
280: \Delta^{2}f^{2}}
281: \eea
282: Here $< ... >$ denotes the averages over the Fermi surface and vortex lattice
283: \cite{20}. In the superclean limit $\tilde{\omega}$ is given by
284: \bea
285: \tilde{\omega} &=& \omega + i\Gamma \left\langle\frac{|\tilde{\omega} -
286: {\bf v}\cdot {\bf q}|}{\sqrt{(\tilde{\omega}-{\bf v} \cdot {\bf q})^{2} -
287: \Delta^{2}f^{2}}}\right\rangle \\
288: & \simeq & \omega+i\Gamma G(\omega,{\bf H})
289: \eea
290: in the Born limit. And in the unitary limit we find
291: \bea
292: \tilde{\omega} = \omega+i\Gamma G^{-1}(\omega,{\bf H})
293: \eea
294: where G($\omega,{\bf H}$) has been defined in Eq.(5).
295:
296: First we limit ourselves to the quasi-2D systems in a magnetic field
297: {\bf H} $\parallel$ {\bf c}. The in the Born limit we obtain
298: \bea
299: \kappa & = & \frac{n}{8mT^{2}\Gamma} \int_{0}^{\infty} d\omega\,\, \omega^{2}
300: \mathrm{sech}^{2}(\frac{\omega}{2T}= \frac{\pi^{2}nT}{12m\Gamma} = \frac{1}{2}
301: \kappa_{n}
302: \eea
303: where $\kappa_{n} = \frac{\pi^{2}nT}{6m\Gamma}$ is the thermal conductivity
304: in the normal state. In particular Eq.(28) gives the scaling function
305: \bea
306: F_{B}(T/\epsilon) &=& \frac{\kappa(T,{\bf H})}{\kappa(T,0)} = 1
307: \eea
308: The last result agrees with the corresponding result given in Ref.\cite{19}
309: despite the use of a rather unphysical spatial average in this work. In the
310: unitary limit, on the other hand, we obtain
311: \bea
312: \kappa &=& \frac{n}{8m(T\Delta)^{2}\Gamma}\int_{0}^{\infty}d \omega \, \omega^{2}
313: <|\omega-{\bf v}\cdot{\bf q}|>^{2}\mathrm{sech}^{2}(\omega/2T) \\
314: &=& \frac{n}{8mT^{2}\Gamma}(\frac{\pi\epsilon}{4\Delta})^{2}\int_{0}^{\infty}d \omega \, \omega^{2} G^{2}(\omega/\epsilon)\mathrm{sech}^{2}(\omega/2T)
315: \eea
316: where $G(\omega/\epsilon)$ has already been defined in Eqs. 5 and 6. This
317: has asymptotics
318: \bea
319: \kappa &=& \frac{7n \pi^4 T^{3}}{60m\Gamma\Delta^{2}}\left(
320: 1+ \frac{5}{7}(\frac{\epsilon}{\pi T})^2 + \frac{15}{28}
321: (\frac{\epsilon}{\pi T})^{4} + \ldots \right), \,\,
322: \mathrm{for\,\, \epsilon \ll T} \\
323: &=& \frac{\pi^2 nT}{12m\Gamma}(\frac{\pi\epsilon}{4\Delta})^2\left(1+ \frac{7\pi^2}{15}(\frac{T}{\epsilon})^2 + \ldots\right), \,\, \mathrm{for\, \epsilon \gg T}
324: \eea
325: where $\epsilon=\frac{v\sqrt{eH}}{2}$. Then the scaling function is given by
326: \bea
327: F_{u}(T/\epsilon) &=& \frac{\kappa(T,{\bf H})}{\kappa(T,0)} =
328: \frac{3}{2\pi^{2} T^{3}}\int_{0}^{\infty}d\omega \omega^{2} G^{2}(\omega/\epsilon) \mathrm{sech}^{2}(\omega/2T) \\
329: &=& \left(1+ \frac{5}{7}(\frac{\epsilon}{\pi T})^2 + \frac{15}{28}
330: (\frac{\epsilon}{\pi T})^{4} + \ldots \right), \,\,
331: \mathrm{for\,\, \epsilon \ll T} \\
332: &=& \frac{5}{112}(\frac{\epsilon}{\Delta})^2\left(1+ \frac{7\pi^2}{15}(\frac{T}{\epsilon})^2 + \ldots\right), \,\, \mathrm{for\, \epsilon \gg T}
333: \eea
334: These scaling functions are shown in Fig. 3. In this figure we also include
335: \begin{figure}[h]
336: \includegraphics[width=7cm,angle=270]{fig3.ps}
337: \caption{The scaling functions F$_{I}$, F$_B$, and F$_U$}
338: \end{figure}
339: the scaling function when the inversion symmetry is broken in the impurity
340: scattering \cite{28}. $F_{I}(T/\epsilon)$ describes the thermal conductivity
341: data of the non-centrosymmetric triplet superconductor CePt$_{3}$Si by
342: Izawa et al \cite{29}. The scaling function $F_{u}(T/\epsilon)$ is very
343: different from the one given in Ref. \cite{19} but describes consistently
344: the scaling behaviors of the thermal conductivity of UPt$_{3}$ as reported
345: by Suderow et al \cite{30}.
346:
347: \section{Angle dependent thermal conductivity tensor}
348:
349: Let us consider d$_{x^{2}-y^{2}}$-wave superconducivity as in the high-T$_{c}$
350: cuprates, CeCoIn$_{5}$ \cite{6} and $\kappa$-(ET)$_{2}$(NCS)$_{2}$ \cite{7}
351: in a magnetic field within the a-b plane. Then the thermal conductivity tensors within the ab-plane are given by
352: \bea
353: \kappa_{xx} &=& \kappa_{yy} = \frac{n}{8m\Gamma T^{2}} \int_{0}^{\infty}
354: d\omega \,\, \omega^{2} \left\langle \frac{4\epsilon(\phi,\chi)}{\pi \Delta}
355: G(\omega/\epsilon(\phi,\chi))\right\rangle ^{2} \mathrm{sech}^{2}(\omega/2T)
356: \eea
357: where $\epsilon(\phi,\chi) = \frac{\tilde{v}\sqrt{eH}}{2}(1 \pm \frac{1}{2}
358: \sin(2\phi) - \frac{1}{2}\cos(2\chi))^{1/2}$ and $\langle \ldots \rangle$
359: means the average over $\pm$ and over $\chi$. Here we assumed the unitary
360: impurity scattering and the superclean limit in the present derivation.
361: This gives the following asymptotics:
362: \bea
363: \kappa_{xx} &=& \kappa_{yy} = \frac{7nT}{60m\Gamma}(\frac{\pi T}{\Delta})^{2}\left(
364: 1+ \frac{5}{7}(\frac{\epsilon}{\pi T})^2 + \frac{15}{28}
365: (\frac{\epsilon}{\pi T})^{4} + \ldots \right), \,\,
366: \mathrm{for\,\, \epsilon \ll T} \\
367: &=& \frac{4nT}{3m\Gamma}(\frac{\epsilon}{\Delta})^{2}[0.927 + 0.039 \cos(4\phi)
368: + \frac{7}{15}(\frac{\pi T}{\epsilon})^{2}[1.090 - 0.055 \cos(4\phi)]
369: \mathrm{for\,\, \epsilon \gg T}
370: \eea
371: First of all, the present result is consistent with that in Ref. \cite{20}
372: for $T < \epsilon$. On the other hand, for $T > \epsilon$ there is no fourfold
373: term. In other words the present theory in the superclean limit cannot describethe fourfold symmetry in $\kappa_{xx}$ observed in YBCO for $ T > 14 K$
374: \cite{31,32,33}. We have proposed the sign inversion of the fourfold term
375: for $T > \epsilon$ in the clean limit in \cite{34}.
376:
377: Also the Hall conductivity in the present geometry is given by \cite{20}
378: \bea
379: \kappa_{xy} &=& \frac{n}{8m(T\Delta)^{2}\Gamma}\int_{0}^{\infty}
380: d\omega \,\, \omega^{2} \left\langle \sin(2\phi^{`})|\omega - {\bf v} \cdot
381: {\bf q}| \right\rangle \left\langle |\omega - {\bf v} \cdot
382: {\bf q}| \right\rangle \mathrm{sech}^{2}(\omega/2T)
383: \eea
384: Further, we find
385: \bea
386: \left\langle \sin(2\phi^{`})|\omega - {\bf v} \cdot
387: {\bf q}| \right\rangle &=& -\sin(2\phi) \frac{\epsilon^{2}}{2|\omega|}, \mathrm{for\,\,
388: \epsilon \ll |\omega|} \\
389: & \simeq & -\sin(2\phi)\frac{4\epsilon}{\pi\Delta}(0.535 -
390: (\frac{\omega}{\epsilon})^{2}0.14192)\, \mathrm{for\,\, \epsilon \gg |\omega|}
391: \eea
392: Inserting these into Eq.(40) we find
393: \bea
394: \kappa_{xy} &=& -\sin(2\phi) \frac{\pi^{2}nT}{24m\Gamma}(\frac{\epsilon}{\Delta})^{2}
395: (1 + \frac{3}{2}(\frac{\epsilon}{\pi T})^{2},
396: \mathrm{for \,\, \epsilon \ll T} \\
397: &=& -\frac{4nT}{3m\Gamma}(\frac{\epsilon}{\Delta})^{2}\sin(2\phi)(0.5152 +
398: 0.011 \cos(4\phi))-\frac{7}{30}(\frac{\pi T}{\epsilon})^{2}(0.213 + .0607
399: \cos(4\phi))\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\
400: \,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\,\, \mathrm{for \epsilon \gg T}
401: \eea
402: Therefore in the superclean limit $\kappa_{xy} \sim -\sin(2\phi)H$ independent
403: of $\epsilon/T$. Also as $\frac{T}{\epsilon}$ increases the coefficient
404: of -$\sin(2\phi) H$ decreases almost 40\%. The present result appears to be
405: consistent with the Hall conductivity data of YBCO reported by Oca\~{n}a and
406: Esquinazi \cite{35} for $\epsilon/\Delta < 1$. Also in the superclean limit
407: the sign of the Hall conductivity is the same for all T as long as
408: $T < \Delta(T)$.
409:
410: \section{Concluding Remarks}
411:
412: We have shown a) the thermal conductivity in nodal superconductors for
413: $T < 0.3 T_{c}$ is dominated by the quasiparticles or nodal excitations,
414: b) the quasiparticles in the vortex state are accurately described in terms
415: of the semiclassical approximation. Thus the angle dependent magneto-thermal
416: conductivity provides a powerful tool to determine the nodal structure
417: of the gap function as demonstrated by a series of experiments by Izawa et al
418: \cite{5,6,7,8,9,10,11,12}. Also in most cases the nodal structure of the
419: gap function is adequate to deduce $|\Delta({\bf k})|$ itself. In addition
420: we have shown that all these model superconductors exhibit a variety of
421: scaling relations. We have proposed scaling relations for PrOs$_{4}$Sb$_{12}$
422: \cite{36}. Furthermore, from the unusual scaling relation seen in the thermal
423: conductivity in CePt$_{3}$Si we can deduce anomalous impurity scattering in
424: this system lacking crystalline inversion symmetry. Indeed the scaling
425: relations in nodal superconductors provide a unique way to characterize
426: this new class of superconductors.
427:
428: \begin{thebibliography}{99}
429:
430: \bb{1} F. Steglich et al, Phys. Rev. Lett. {\bf 43}, 1892 (1979).
431: \bb{2} A. Damascelli, Z. Houssain and Z.X. Shen, Rev. Mod. Phys. {\bf 75},
432: 473 (2003).
433: \bb{3} D.J. van Harlingen, Rev. Mod. Phys. {\bf 67}, 515 (1995).
434: \bibitem{4} C. C. Tsuei and J. R. Kirtley, Rev. Mod. Phys. {\bf 72}, 969 (2000).
435: \bibitem{5} K. Izawa, H. Takahashi, H. Yamaguchi, Yuji Matsuda, M. Suzuki, T. Sasaki, T. Fukase, Y. Yoshida,
436: R. Settai and Y. Onuki, Phys. Rev. Lett. {\bf 86}, 2653 (2001).
437: \bibitem{6} K. Izawa, H. Yamaguchi, Yuji Matsuda, H. Shishido, R. Settai and Y. Onuki,
438: Phys. Rev. Lett. {\bf 87}, 57002 (2001).
439: %68
440: \bibitem{7} K. Izawa, H. Yamaguchi, T. Sasaki and Yuji Matsuda, Phys. Rev. Lett. {\bf 88}, 027002 (2002).
441: %69
442: \bibitem{8} K. Izawa, K. Kamata, Y. Nakajima, Y. Matsuda, T. Watanabe, M. Nohara, H. Takagi,
443: P. Thalmeier and K. Maki, Phys. Rev. Lett. {\bf 89}, 137006 (2002).
444: %70
445: \bibitem{9} K. Izawa, Y. Nakajima, J. Goryo, Y. Matsuda, S. Osaki, H. Sugawara, H. Sato, P. Thalmeier and
446: K. Maki, Phys. Rev. Lett. {\bf 90}, 117001 (2003).
447: \bibitem{10} K. Maki, S. Haas, D. Parker, H. Won, K. Izawa and Y. Matsuda, Europhys. Lett. {\bf 65},
448: 720 (2004).
449: \bibitem{11} T. Watanabe, K. Izawa, Y. Kasahara, Y. Haga, Y. Onuki, P. Thalmeier, K. Maki
450: and Y. Matsuda, Phys. Rev. B {\bf 70}, 184502 (2004).
451: \bibitem{12} H. Won, D. Parker, K. Maki, T. Watanabe, K. Izawa and Y. Matsuda, Phys. Rev. B
452: {\bf 70}, 140509 (2004).
453: \bb{13} G.E. Volovik, JETP Lett. {\bf 58}, 496 (1993).
454: \bb{14} H. Won, S. Haas, D. Parker, S. Telang, A. Vanyolos, and K. Maki in
455: AIP Proceedings of Ninth Training School at Vietri sul Mare, Italy, 2005;
456: also cond-mat/0501463.
457: \bb{15} K. Deguchi, Z.Q. Mao and Y. Maeno, J. Phys. Soc. Jpn. {\bf 75},
458: 1315 (2004).
459: \bb{16} H. Won and K. Maki, Europhys. Lett. {\bf 54}, 246 (2001).
460: \bb{17} S.H. Simon and P.A. Lee, Phys. Rev. Lett. {\bf 78}, 1548 (1997).
461: \bb{18} C. K\"{u}bert and P.J. Hirschfeld, Solid State Comm.
462: {\bf 105}, 459 (1998).
463: \bb{19} C. K\"{u}bert and P.J. Hirschfeld, Phys. Rev. Lett.
464: {\bf 80}, 4963 (1998).
465: \bb{20} H. Won and K. Maki, cond-mat/0004105.
466: \bb{21} H. Won and K. Maki, Phys. Rev. B {\bf 49}, 1397 (1994).
467: \bb{23} K. Maki, S. Haas, D. Parker and H. Won, Chinese J. Phys. {\bf 43},
468: 532 (2005).
469: \bb{24} B. Dora, K. Maki and A. Virosztek, Europhys. Lett. {\bf 62}, 426
470: (2003).
471: \bb{25} B. Revaz, J.-Y. Genoud, A. Junod, K. Neumaier, A. Erb, and E. Walker,
472: Phys. Rev. Lett. {\bf 80}, 3364 (1998).
473: \bb{26} L.P. Kadanoff and I.I Falko, Phys. Rev. {\bf 136}, A1170 (1964).
474: \bibitem{27} V. Ambegaokar and A. Griffin, Phys. Rev. {\bf 137}, A1151 (1965).
475: \bb{28} K. Maki and H. Won, SCES'05 proceeding
476: \bb{29} K. Izawa et al, Phys. Rev. Lett. {\bf 94}, 197002 (2005).
477: \bb{30} H. Suderow, J. P. Brison, A. Huxley, and J. Flouquet, Phys. Rev. Lett. {\bf 80}, 165 (1998).
478: \bb{31} F. Yu, M.B. Salamon, A.J. Leggett, W.C. Lee and D.M. Ginzburg,
479: Phys. Rev. Lett. {\bf 74}, 5136 (1995).
480: \bb{32} H. Aubin, K. Behnia, M. Ribault, R. Gagnon and T. Taillefer, Phys.
481: Rev. Lett. {\bf 78}, 2624 (1997).
482: \bb{33} R. Oca\~{n}a and P. Esquinazi, Phys. Rev. B {\bf 66}, 064525 (2002).
483: \bb{34} H. Won and K. Maki, Curr. Appl. Phys. {\bf 1}, 291 (2001).
484: \bb{35} R. Oca\~{n}a and P. Esquinazi, Phys. Rev. Lett. {\bf 87}, 167006
485: (2001).
486: \bb{36} H. Won, S. Haas, D. Parker and K. Maki, cond-mat/0503350.
487: \end{thebibliography}
488:
489: \end{document}
490: