1: \documentclass{epl2}
2: \usepackage[latin1]{inputenc}
3: \usepackage{setspace}
4: \usepackage{amsmath}
5: \usepackage{amsfonts}
6: \usepackage{amssymb}
7: \usepackage{graphicx}
8: \usepackage{wrapfig}
9: \usepackage{subfig}
10:
11:
12: \title{Chiral d-wave superconductivity in the heavy-fermion compound $\rm CeIrIn_5$}
13: \author{K. Maki \and A. Raghavan \and S. Haas}
14: \institute{Department of Physics \& Astronomy, University of Southern California, Los Angeles, CA 90089-0484}
15: \abstract{Recent thermal conductivity measurements in the heavy-fermion compound
16: $\rm CeIrIn_5$ indicate that its superconducting order parameter
17: is very different from $\rm CeCoIn_5$. Here we show that these
18: experiments are consistent
19: with chiral d-wave symmetry, i.e. $\Delta(\vec{k})\sim e^{\pm i\phi}\cos(ck_z)$.}
20:
21: \pacs{74.70.Tx}{Heavy-fermion superconductors}
22: \pacs{74.25.Fy}{Transport properties}
23: \pacs{71.27.+a}{Strongly correlated electron systems; heavy fermions}
24:
25: \begin{document}
26: \maketitle
27:
28: The discovery of antiparamagnon mediated superconductivity in the 115 heavy-fermion compounds
29: $\rm CeTIn_5$, where T represents Co, Ir, Rh, or a mixture of these,
30: has recently opened up a new avenue to unconventional nodal superconductivity\cite{petrovic}. These
31: strongly interacting materials are
32: characterized by a plethora of competing ground states in addition to superconductivity, including conventional
33: and unconventional spin density wave (SDW) phases\cite{knebel}. Among the 115 compounds, the currently most well
34: studied is $\rm CeCoIn_5$ for which a d-wave superconducting order parameter
35: $\Delta(\vec{k})\sim\cos(2\phi)=\hat{k}_x^2-\hat{k_y^2}$ has been identified\cite{izawa,aoki,won,miclea}.
36: Indeed, there are many parallels between $\rm CeCoIn_5$ and the high-$T_c$ cuprates, including
37: (a) a layered quasi-two-dimensional Fermi surfaces\cite{maehira},
38: (b) d-wave superconductivity, and
39: (c) d-wave spin density wave order in the pseudogap phase\cite{hu,maki,won2}.
40:
41: Recent thermal conductivity measurements \cite{shakeripour,ohira} indicating an order parameter symmetry
42: in $\rm CeIrIn_5$ very different from the one in $\rm CeCoIn_5$ came as a big surprise.
43: An initial analysis of this data suggested a hybrid $E_g$ gap,
44: $\Delta(\vec{k})\sim Y_{2,\pm 1}(\theta,\phi)$,
45: based on the assumption that the Fermi surface is three-dimensional. However, the Fermi surface of $\rm CeIrIn_5$
46: is in fact quasi-two-dimensional, as known from band structure analysis\cite{maehira,ohira}. Therefore, one needs
47: to consider instead superconductivity in
48: layered structures, similar as discussed in Refs. \cite{won3,won4}. In this case, only
49: $f=e^{\pm i\phi}\sin(\chi)$ (chiral d-wave) with $\chi = ck_z$ or $f\sim \sin(\chi)$ (non-chiral p-wave) are
50: consistent with the observed thermal conductivity data \cite{shakeripour}.
51: The magnitudes of the d-wave and chiral d-wave/non-chiral p-wave
52: order parameters $|\Delta(\vec{k})|$ are shown in Fig. 1.
53:
54: \begin{figure}
55: \centering
56: \includegraphics[width=3cm]{dwave.pdf}
57: \includegraphics[width=2.5cm]{chiraldwave.pdf}
58: \caption{Magnitude of order parameters $|\Delta(\vec{k})|$ for d-wave (left)
59: and chiral d-wave/non-chiral p-wave (right) superconductors.}
60: \label{fig:delta_k}
61: \end{figure}
62:
63: In the following, we present a theoretical analysis based on a generalized BCS model
64: that properly accounts for a quasi-two-dimensional Fermi surface and a chiral d-wave superconducting order
65: parameter.
66: The thermal conductivity is computed following the receipe given in Refs. \cite{won4,yang}.
67: Here, we assume for simplicity that the quasiparticle scattering is due to impurities.
68: Furthermore, we consider the physically relevant limit $\Gamma/\Delta\ll 1$, where $\Gamma$
69: is the quasiparticle scattering rate in the normal state and $\Delta (=0.856K)$ is the maximum
70: value of the energy gap at $T=0K$. This $\Delta$ is the weak-coupling value for
71: nodal superconductors\cite{won3,wonmaki}.
72:
73: Let us begin by considering the
74: zero-temperature limit. For quasi-two-dimensional structures, the thermal
75: conductivity
76: strongly depends on the direction within the material.
77: Therefore we will discuss the cases
78: $\vec{q}\|\vec{a}$ (in-plane) and $\vec{q}\|\vec{c}$
79: (out-of-plane) separately.
80: Here $\vec{q}$ denotes the heat current.
81: For $\vec{q}\|\vec{a}$, one obtains
82: \begin{equation}
83: \dfrac{\kappa^a}{\kappa^a_n}=\dfrac{2\Gamma_a}{\pi\Delta}
84: \end{equation}
85: and similarly for $\vec{q}\|\vec{c}$
86: \begin{equation}
87: \dfrac{\kappa^c}{\kappa^c_n}=2\left(\dfrac{\Gamma_c}{\Delta}\right)^2,
88: \end{equation}
89: where $\Gamma_a$ and $\Gamma_c$ denote the in-plane and out-of-plane
90: scattering rates respectively.
91: Eq.1 describes the universal heat conduction as discovered by
92: P. Lee\cite{lee,sun}, whereas Eq.2 is very different. The strength
93: of the impurity scattering can be extracted directly from the
94: experimental data show in Fig. 2 of Ref.\cite{shakeripour}, from
95: which we can deduce that $\dfrac{\Gamma}{\Delta}=0.19635$. Furthermore,
96: from the observed anisotropy of the thermal conductivity, we can infer
97: the ratio of the Fermi velocities along the c-axis and the a-b plane,
98: i.e. $\dfrac{v_c}{v_a}=0.66$, which is very similar to $\dfrac{v_c}{v_a}=0.5$
99: extracted for $\rm CeCoIn_5$\cite{hu}. Then, for $T\neq 0K$ but
100: $\dfrac{T}{\Delta}\ll 1$, we obtain in the regime $T\gg\Gamma$,
101: \begin{equation}
102: \dfrac{\kappa^a(T)}{\kappa^a_n(T)}=\dfrac{27}{2\pi^2}\zeta(3)\left(\dfrac{T}{\Delta}\right)+O\left(\dfrac{T}{\Delta}\right)^3,
103: \end{equation}
104: and
105: \begin{equation}
106: \dfrac{\kappa^c(T)}{\kappa^c_n(T)} = \dfrac{45^2}{4\pi^2}\zeta(5)\left(\dfrac{T}{\Delta}\right)^3+O\left(\dfrac{T}{\Delta}\right)^5.
107: \end{equation}
108: This is consistent with the experimental observation of a dominant in-plane
109: heat conductivity propertional to the temperature, and a subdominant
110: out-of-plane conductivity.
111:
112: In order to connect these finite-temperature results with the above
113: equations for $T=0$, we use an interpolation formula which applies
114: in the regime for $T/\Delta(T)\ll 1$. The resulting low-temperature
115: thermal conductivities are then given by
116: \begin{equation}
117: \dfrac{\kappa^a(T)}{\kappa^a_n(T)} =\dfrac{2\Gamma_a}{\pi\Delta}\left(1+\left(\dfrac{27}{4\pi}\zeta(3)\dfrac{T}{\Gamma_a}\right)^2\right)^{1/2}
118: \end{equation}
119: and
120: \begin{equation}
121: \dfrac{\kappa^c(T)}{\kappa^c_n(T)} = 2\left(\dfrac{\Gamma_c}{\Delta}\right)^2\left(1+\left(\dfrac{45^2}{8\pi^2}\zeta(5)\left(\dfrac{T}{\Gamma_c}\right)^2\left(\dfrac{T}{\Delta}\right)\right)^2\right)^{1/2}
122: \end{equation}
123: respectively.
124:
125: In Fig. 2, we compare these dependencies with the experimental data
126: reported in Ref.\cite{shakeripour}. A fit of the low-temperature regimes
127: yields good agreement with $\dfrac{\Gamma_c}{\Gamma_a}=0.5592$.
128: Evidently, the quasi-particle scattering rate is somewhat anisotropic in
129: the present system. Here, the temperature dependence of the gap function
130: $\Delta (T)$, is approximated by
131: \begin{equation}
132: \Delta(T)=2.14T_c\left[1-\left(\dfrac{T}{T_c}\right)^3\right]^{1/2}
133: \end{equation}
134: with $T_c=0.4K$, which is known to be a very good approximation
135: for d-wave superconductors\cite{dora}.
136:
137: \begin{figure}
138: \centering
139: \includegraphics[width=6.0cm]{figure1.pdf}
140: \includegraphics[width=6.0cm]{figure2.pdf}
141: \caption{Thermal conductivity in the $\vec{q}\| \vec{a}$ (in-plane) and
142: $\vec{q}\| \vec{c}$ (out-of-plane) directions. $T_c=0.4 K$.
143: The symbols represent experimental data from \cite{shakeripour}, and the
144: solid lines are low-temperature fits using Eqs. 5 and 6.}
145: \label{fig:two}
146: \end{figure}
147:
148: Similarly, the ratio $\kappa^c(T)/\kappa^a(T)$ can be computed and
149: compared to the experiments. Within our model, it is given by
150: \begin{equation}
151: \kappa^c(T)/\kappa^a(T)=0.2703\left[\dfrac{1+\left(\dfrac{45^2}{8\pi^2}\zeta(5)\right)^2\dfrac{T^6}{\Gamma_c^4\Delta^2}}
152: {1+\left(\dfrac{27}{4\pi}\zeta(3)\right)^2\left(\dfrac{T}{\Gamma_a}\right)^2}\right]^{1/2}
153: \end{equation}
154: which is shown in Fig. 3 along with the thermal conductivity measurements of Ref. \cite{shakeripour}.
155: \begin{figure}
156: \centering
157: \includegraphics[width=8cm]{figure3.pdf}
158: \caption{Ratio of thermal conductivities of the $\vec{q}\| c$
159: and $\vec{q}\|\vec{a}$ direction, plotted as a function of $T/T_c$.
160: The symbols represent the experimental data from Ref.\cite{shakeripour}.}
161: \label{fig:three}
162: \end{figure}
163:
164:
165: These expressions give a very reasonable description of the thermal
166: conductivity for $T/T_c\leq 0.3$. We note that a similarly good description
167: of the thermal conductivity is given by the hybrid gap proposed in Ref.
168: \cite{shakeripour}. At higher temperatures, $T/T_c\geq 0.3$,
169: our simple model fails to
170: describe the measured thermal conductivity, possibly due to the fact that
171: phonons begin to play an important role as we approach $T\rightarrow T_c$.
172: Nevertheless, we can conclude that chiral d-wave SC is consistent with
173: the experimental data of Refs. \cite{shakeripour,ohira} in the relevant
174: low-temperature regime. Note also, that our calculations predict an
175: interesting upturn in the ratio $\kappa^c(T)/\kappa^a(T)$ as the temperature
176: is further lowered. This prediction can be scrutinized experimentally, and
177: may serve as a means to distinguish the present theory from the hybrid
178: gap model that was proposed earlier.
179:
180: In the present context, the unconventional superconducting order
181: in $\rm CeRhIn_5$ is of
182: great interest. Let us briefly contemplate on the doped
183: case. Inspecting Fig. 3 of Ohira-Kawamura et al\cite{ohira}
184: we may conclude that the order parameter in $\rm CeRh_{1-x}Co_xIn_5$
185: should be d-wave SC with an
186: angular dependence $f = \cos(2\phi)$, whereas the order
187: parameter in $CeRh_{1-x}Ir_xIn_5$ is consistent with
188: chiral d-wave superconductivity with an angular dependence
189: $f = e^{\pm i\phi}\cos(\chi)$. Therefore, the above approach will provide a
190: basis to identify the many competing
191: phases of the 115 compounds. Also, the
192: phase diagrams for $\rm CeRh_{1-x}Co_xIn_5$ and $\rm CeRh_{1-x}Ir_xIn_5$
193: in Ref.\cite{ohira} are of great interest for the perspective of the Gossamer
194: superconductivity, i.e. a phase with competing order parameters\cite{won2,won5,maki2}.
195: We observe that
196: (a) the incommensurate phases in both $\rm CeRh_{1-x}Co_xIn_5$ and
197: $\rm CeRh_{1-x}Ir_xIn_5$ are conventional spin-density waves,
198: (b) the commensurate phase in $\rm CeRh_{1-x}Co_xIn_5$ and
199: the incommensurate+commensurate phase in $\rm CeRh_{1-x}Ir_xIn_5$
200: have d-wave symmetry.
201: Therefore, there is a wide region where d-wave superconductivity
202: coexists with unconventional nodal spin density wave order.
203:
204: In summary, we have successfully applied a nodal weak-coupling BCS theory
205: to fit recent experimental
206: data on the directional thermal conductivity of $\rm CeRhIn_5$. We
207: find that in contrast to $\rm CeCoIn_5$, which has plain d-wave order,
208: this compound is consistent with chiral
209: d-wave superconductivity. Furthermore, this technique will allow us
210: to identify the many different phases which were recently discovered
211: in doped derivatives of these materials.
212:
213: \acknowledgements
214: We would like to thank Y. Matsuda for very informative discussions
215: on the superconductivity in $\rm CeIrIn_5$.
216:
217:
218: \newpage
219: \begin{thebibliography}{widest-label}
220: \bibitem{petrovic}C. Petrovic, R. Movshovich, M. Jaime, P. G. Pagliuso, M. F. Hundley, J. L. Sarrao, Z. Fisk, and J. D. Thompson, Europhys. Lett. \textbf{53}, 354 (2001).
221: \bibitem{knebel}G. Knebel, K. Izawa, F. Bourdarot, E. Hassinger, B. Salce, D. Aoki and J. Flouque, J. MMM \textbf{310}, 195 (2007).
222: \bibitem{izawa}K. Izawa, H. Yamaguchi, Yuji Matsuda, H. Shishido, R. Settai, and Y. Onuki, Phys. Rev. Lett\textbf{87}, 057002 (2001).
223: \bibitem{aoki}H. Aoki, T. Sakakibara, H. Shishido, R. Settai, Y. Onuki, P. Miranovic, and K Machida, J. Phys. Cond. Matt \textbf{16}, L13 (2004).
224: \bibitem{won}H. Won, K. Maki, S. Haas, N.Oreschler, F. Weickest and P. Gegenwart, Phys. Rev. B \textbf{61}, 180504(R) (2004).
225: \bibitem{miclea}C. Miclea, M. Nicklas, D. Parker, K. Maki, J. L. Sarrao, J. D. Thompson, G. Sparn, and F. Steglich, Phys. Rev Lett. \textbf{96}, 117001 (2006);ibid \textbf{97}, 039901 (2006).
226: \bibitem{maehira}T. Maehira, T. Hotta, K. Ueda and A. Hasegawa, J. Phys. Soc. Jpn \textbf{72}, 854 (2003).
227: \bibitem{hu}T. Hu, H. Xiao, T. A. Sayles, M. B. Maple, K Maki, B. Dora and C. C. Almasan, Phys. Rev. B \textbf{73}, 134509 (2006).
228: \bibitem{maki}K. Maki, B. Dora, A. Vanyolas and A. Virosztek, Physica C \textbf{460-462}, 226 (2007).
229: \bibitem{won2}H. Won, Y. Morita and K. Maki Phys. State Sol(B)\textbf{244}, 4371 (2007)
230: \bibitem{shakeripour}H. Shakeripour, M.A. Tanatar, S.Y. Li, C. Petrovic,
231: and Louis Taillefer, Phys. Rev. Lett. \textbf{99}, 187004 (2007).
232: \bibitem{ohira}S. Ohira-Kawamura, H. Shishido, A. Yoshida, R. Okazaki,
233: H. Kawano-Furukawa, T. Shibauchi, H. Harima, and Y. Matsuda,
234: Phys. Rev. B \textbf{76}, 132507 (2007).
235: \bibitem{won3}H. Won et al, AIP conference proceedings \textbf{789} (Melville 2005).
236: \bibitem{won4}H. Won, D. Parker, K. Maki, T. Watanabe, K. Izawa,
237: and Y. Matsuda, Phys. Rev. B \textbf{70}, 140509 (2004).
238: \bibitem{yang}G. Yang and K. Maki, Eur Phys. J B \textbf{21}, 61 (2001).
239: \bibitem{wonmaki}H. Won and K. Maki, Phys. Rev. B \textbf{49}, 1397 (1994).
240: \bibitem{lee}P. A. Lee, Phys. Rev. Lett. \textbf{71}, 1887 (1993).
241: \bibitem{sun}Y. Sun and K. Maki, Europhys. Lett \textbf{32}, 355 (1995).
242: \bibitem{dora}B. Dora and A. Virosztek, Eur. Phys. J. B \textbf{22}, 167 (2000).
243: \bibitem{won5}H. Won, S. Haas, D. Parker, and K. Maki, Phys. State Sol(B) \textbf{242}, 363 (2005).
244: \bibitem{maki2}K Maki, S. Haas, D. Parker, H. Won, B. Dora, and A. Virosztek, Phys. State Sol(C) \textbf{3}, 3156 (2006).
245: \end{thebibliography}
246: \end{document}
247: