1: \documentstyle[preprint,aps,prb,psfig]{revtex}
2: \begin{document}
3: \def\alt{\mathrel{\mathpalette\vereq<}}
4: \def\agt{\mathrel{\mathpalette\vereq>}}
5: \def\ep{{\epsilon}}
6: \def\tg{{\tilde{\cal G}}}
7: \def\bk{{{\bf k}}}
8:
9: \title{Infrared absorption in superconducting MgB$_{2}$}
10:
11: \author{C.S.Sundar, A. Bharathi, M. Premila, T.N. Sairam,
12: S. Kalavathi, G.L.N. Reddy,
13: V.S. Sastry, Y.Hariharan and T.S. Radhakrishnan}
14:
15:
16: \address
17: {
18: Materials Science Division,
19: Indira Gandhi Centre for Atomic Research\\
20: Kalpakkam, India 603 102
21: }
22:
23: \maketitle
24:
25: \begin{abstract}
26: {
27: Infrared absorption measurements in the range of 125 to 700 cm$^{-1}$
28: have been carried out as a function of temperature upto 5 K
29: in MgB$_{2}$. The absorption spectrum is
30: characterised by a broad band centred at 485 cm$^{-1}$, with
31: shoulders at 333 and 387 cm$^{-1}$. Studies on the temperature
32: dependence of absorption, indicate
33: that these modes initially harden with the lowering of
34: temperature, and this trend is arrested at $\sim $ 100 K, below
35: which they soften. Further, in the case of the mode at 333 cm
36: $^{-1}$, there is a distinct softening associated with the
37: superconducting transition at 39 K. The implications of these
38: experimental results in the context of superconductivity in
39: MgB$_{2}$ are discussed. } \\
40:
41:
42:
43: \noindent {PACS numbers: 74.70.Ad, 74.25.Kc, 78.30.-j \\}
44: \end{abstract}
45: %]
46: %\narrowtext
47:
48: \section{Introduction}
49:
50: The recent discovery of superconductivity \cite{akimitsu} at 39 K
51: in the binary
52: intermetallic MgB$_{2}$, having a simple
53: hexagonal structure consisting of alternating honeycomb layers
54: of B and closed packed layers of Mg, has
55: evoked a widespread interest. Initial experiments suggest that the
56: superconductivity in this system arises due to phonon-mediated
57: interaction, as supported by experiments on isotope effect
58: \cite{budko}; measurements on the superconducting gap by
59: tunnelling \cite{schmidt} and optical spectroscopy
60: \cite{gorshunov}, as also the decrease of T$_{c}$ with the application
61: of pressure \cite{chu}. There have been several theoretical
62: calculations \cite{kortus,pickett,andersen,satta} that emphasise
63: the importance of electron-phonon coupling, though alternative
64: mechanisms \cite{hirsch,baskaran} for superconductivity have also
65: been proposed. Electronic structure
66: calculations in MgB$_{2}$ indicate that Mg is completely ionised and
67: the bands at the Fermi level are derived from $\sigma $ orbitals
68: of boron. There are four distinct zone centre vibrational modes:
69: a silent mode B$_{1g}$, the doubly degenerate Raman mode,
70: E$_{2g}$, and two infrared active modes of A$_{2u}$ and
71: E$_{1u}$ symmetry. While there is a general agreement with
72: regard to electronic structure and vibrations, the details
73: differ, both with respect to the calculated frequencies and the
74: relative importance of these modes with respect to
75: superconductivity in the system. For example, in the
76: calculations of An and Pickett \cite{pickett}, using the
77: deformation potential approach, it is the E$_{2g}$ mode that is
78: shown to have a dominant coupling with the electrons, whereas in
79: the calculations of Kong et al \cite{andersen}, the electron
80: phonon interaction is spread over all the modes. Inelastic
81: neutron scattering experiments measuring the vibrational density
82: of states \cite{osborne,sato,yildrim} indicate acoustic modes at
83: less than 40 meV and peaks in the phonon density of states at
84: 54, 78, 89 and 97 meV
85: corresponding to the optic modes. Raman scattering measurements
86: \cite{bohnen,chen,goncharov} indicate a mode at $\sim $ 600
87: cm$^{-1}$, which is characterised by very large width $\sim $
88: 200 cm$^{-1}$, indicative of strong electron phonon interaction.
89:
90:
91: In this paper, results of infrared absorption measurements on
92: superconducting MgB$_{2}$, covering a spectral range of 125 -
93: 700 cm$^{-1}$, in which the optic modes are predicted to exist,
94: are reported. In addition supportive experiments have also been
95: carried out in the mid infrared range, upto 2000 cm$^{-1}$.
96: Based on these experiments and comparison with theoretical
97: calculations \cite{kortus,pickett,andersen,satta}, the optic
98: modes of MgB$_{2}$ are identified. From studies on the
99: temperature dependence of infrared absorption, it is seen that
100: the optic modes show an interesting temperature variation in
101: that the hardening behaviour at low temperatures is arrested
102: below 100 K, below which they soften. In addition, the IR mode
103: at 333 cm$^{-1}$ shows a distinct softening below T$_{c}$.
104:
105:
106: \section{Experimental Details}
107:
108: The MgB$_{2}$ sample used in the present experiments was
109: prepared from Mg (99.99 $\% $) powder of 50 mesh and B (99.98
110: $\% $) powder of 325 mesh. The starting materials were
111: thoroughly mixed and put in Ta tube that was sealed in Ar
112: atmosphere. This was subsequently sealed in a quartz tube that
113: was heat treated at 1223 K for 2 hours. After cooling down to
114: room temperature, the polycrystalline lumps were crushed using
115: agate mortar and pestle and subsequently used for
116: experimentation. X - ray diffraction measurements were carried
117: out using Cu-K$_{\alpha } $ radiation in the Bragg-Brentano
118: geometry. AC susceptibilty measurements were carried out with a
119: mutual inductance bridge and lock-in amplifier, using a dipstick cryostat operating
120: in the 4 to 300 K range. The diamagnetic signal corresponding to
121: the sharp superconducting transition at 39 K, and the results of the
122: x-ray diffraction measurements are shown in Fig.1. The
123: diffraction pattern of MgB$_{2}$ can be indexed to hexagnal
124: structure (P6/mmm) with lattice parameters: a= 3.0864 A and
125: c=3.5253 A, in agreement with earlier studies \cite{jorgensen}.
126: The diffraction pattern of crystalline $\beta $ rhombohedral
127: boron is also
128: shown, and it is seen that boron peaks are not discernable in the
129: diffraction pattern of MgB$_{2}$. However, there are a few
130: peaks that could be associated with MgO, whose concentration is
131: estimated to be $\sim 1 \% $. From the powder ac-susceptibilty
132: measurements, using Pb powder as standard, the superconducting
133: volume fraction was estimated to be $75 \% $.
134:
135: Infrared absorption measurements were carried out on finely
136: ground MgB$_{2}$ sample pelletized along with KBr, using a
137: BOMEM -DA8 spectrometer operating with a resolution of 2
138: cm$^{-1}$. Measurements in the range of 125 - 700 cm $^{-1}$
139: were carried out using a mylar beam splitter and a DTGS
140: detector. Experiments in the mid infrared range of 400 to 2000
141: cm$^{-1}$ have been carried out using the combination of KBr
142: beam splitter and MCT detector. To study the temperature
143: variation of IR modes, the sample was mounted inside a JANIS
144: continuous flow cryostat in which temperature variation of 300
145: to 5 K could be achieved.
146:
147: \section{Results and Discussion}
148:
149: Fig.2 shows the IR absorbance of MgB$_{2}$ in the range of 300
150: to 650 cm$^{-1}$ . The region below 300 cm$^{-1}$ is supressed,
151: as it is dominated by KBr absorption. The absorption spectrum in
152: MgB$_{2}$ is characterised by a broad band centred at 485
153: cm$^{-1}$ with shoulders at 333, 387 cm$^{-1}$. Further sharp
154: features are also noted at 542, 592 and 633 cm$^{-1}$. The
155: latter modes match with those of $\beta $ rhombohedral boron
156: \cite{nogi}, whose absorption spectrum is also shown. The occurrence of these
157: modes, while the x-ray diffraction pattern does not indicate the presence
158: of any B (cf. Fig. 1), p points to the activation of B-like modes in MgB$_2$
159: due to disorder (see below). Information on the
160: frequencies, widths and intensity of the various phonon modes
161: have been obtained by fitting the absorption curves to a sum of
162: Gaussians and a linear background. The resulting fits along
163: with the components is shown in the right panel of Fig.2, for
164: the two representative temperatures of 297 K and 5 K.
165:
166: Factor group analysis predicts for MgB$_2$ ( space group P6/mmm,
167: z=1) B$_{1g}$ + E$_{2g}$ + E$_{1u}$+ A$_{2u}$ zone centre optic
168: modes, of which E$_{1u}$ and A$_{2u}$ are IR active and E$_{2g}$
169: is Raman active. There have been several calculations of these
170: mode frequencies \cite{kortus,pickett,andersen,satta,yildrim} with a
171: general agreement. Kortus et al \cite {kortus} have calculated
172: the two IR active modes of E$_{1u}$ and A$_{2u}$ symmetry to be
173: at 320 and 390 cm$^{-1}$ respectively and a Raman mode of
174: E$_{2g}$ symmetry at 470 cm$^{-1}$. In the calculations of Kong
175: et al \cite{andersen}, carried out using LMTO method, the
176: infrared modes are at 335 and 401 cm$^{-1}$ and the Raman mode
177: is at 585 cm$^{-1}$. Yildrim et al \cite{yildrim} have
178: calculated the infrared modes to be $\omega$(E$_{1u}$) = 40.7
179: meV, $\omega$(A$_{2u}$) = 49.8 meV and the Raman mode
180: $\omega$(E$_{2g}$) = 74.5 meV. Comparing these theoretical
181: calculations \cite{kortus,andersen,satta,yildrim} with our
182: experimental results, we identify the absorption features at 333
183: and 387 cm$^{-1}$ with E$_{1u}$ and A$_{2u}$ infrared modes.
184:
185: As for the absorption band centred at 485 cm$^{-1}$, we first
186: note that it is different from the Raman mode identified to be
187: at 560 cm$^{-1}$ in the experiments by Bohnen et al
188: \cite{bohnen} and at 620 cm$^{-1}$ in the experiments of Chen et
189: al \cite{chen}, and Goncharov et al \cite{goncharov}. In all
190: these experiments, the Raman mode is observed to be very broad
191: $\sim 200 $ cm$^{-1}$. Chen et al \cite{chen} have suggested that
192: this broad feature arises due to disorder which relaxes the momentum
193: selection rule resulting in phonons in the entire Brillouin
194: zone being sampled in the Raman experiment. Taking cue from
195: this, we note that the broad feature at 485 cm$^{-1}$, observed
196: in the present infrared absorption experiments (cf. Fig.2) can
197: be associated with the peak in the phonon density of states at
198: $\sim $ 54 meV, that is seen in the theoretical calculations and
199: neutron scattering experiments\cite{osborne,yildrim}. In effect
200: the broad absorption band centred at 485 cm$^{-1}$ arises
201: due to sampling of the phonons in this energy range over the
202: entire Brillouin zone. The exact nature of disorder that is
203: being invoked to account for the absorption spectrum is not clear
204: at present, but could be off-stoichiometry or disorder in the arrangement
205: of layered structure. We reiterate that the present infrared
206: absorption measurements have been carried out on a sample
207: characterised by sharp x-ray diffraction pattern and
208: superconducting transition (cf. Fig.1).
209:
210:
211: The results of additional infrared absorption experiments in MgB$_2$,
212: carried out over an extended range upto 2000 cm$^{-1}$, are shown
213: in Fig.3. The absorption spectrum is characterised by a linear background
214: with broad humps centred at 485, 1040, 1442 and 1635 cm$^{-1}$. While the band
215: at 1040 cm$^{-1}$ matches with crystalline B (also shown), we discount the
216: possibilty of attributing this feature in MgB$_2$ to the presence of a second
217: phase of B in our sample, since our x-ray diffraction pattern does not indicate
218: the presence of unreacted B. To substatntiate this, we also show in Fig. 3b
219: the absorption spectrum of superconducting MgB$_2$ synthesised
220: from amorphous B. This is also characterised by an absorption band
221: at 1040 cm$^{-1}$, a feature that is absent in the
222: starting amorphous B. This could not have arisen due to crystallisation of
223: amorphous B itself, a process that is known \cite{talley} to occur only beyond
224: 1500 K. From these studies, as also several control infrared and x-ray diffraction
225: experiments on MgB$_2$ samples in which intentionally crystalline
226: B has been added, and studies \cite{bharathi} on Cu doped MgB$_2$,
227: we infer that the absorption spectrum of
228: MgB$_2$ shown Fig.3, viz., as characterised by a linear background with
229: broad bands centred at 485, 1040, 1442 and 1635 cm$^{-1}$ is intrinsic to
230: the system. We note that the absorption bands occurring at 1040,
231: 1442 and 1635 cm$^{-1}$ are
232: beyond the range of optic phonons predicted by theoretical
233: calculations \cite{kortus,pickett,andersen,satta}. This once again points to the
234: important role of disorder in this system which may be activating the
235: the high frequency B-like modes \cite{nogi} in MgB$_2$. We also note that
236: these high frequency features appear like replication of the absorption
237: band at 485 cm$^{-1}$ and hence may be due to combination modes. The
238: relatively large intensity of these combination modes could be due to
239: the large anharmonicity, which has been shown to exist
240: \cite{yildrim} in this system.
241:
242:
243: \section{Temperature dependence of IR absorption}
244:
245: The temperature dependence of IR absorption in the range of
246: 300-650 cm$^{-1}$ has been followed across the superconducting
247: transition. Fig.4 summarises the results on the temperature
248: variation of the frequencies and widths of the phonon features
249: at 333, 387 and 485 cm$^{-1}$. These are seen to harden
250: initially with the lowering of temperature, but this trend is
251: arrested at $\sim $ 100 K, below which they soften. An anomalous
252: variation is also noticed in the widths, in that they increase
253: with the lowering of temperature. In contrast, it is seen from
254: Fig.5 that the phonon modes at 542, 592 and 634 cm$^{-1}$,
255: which are B-like modes(cf. Fig.2), indicate a regular behaviour, viz., a
256: hardening of the modes and a decrease in phonon width with the
257: lowering of temperature.
258:
259: The observed softening of the mode frequencies below $\sim $ 100
260: K in Fig.4 point to a structural anomaly. While the presence of an
261: incipient structural instability is well known in the case of
262: strong electron phonon coupled superconductors \cite{sinha},
263: this is not clearly established in the case of MgB$_{2}$. In
264: fact the experiments by Jorgensen et al \cite{jorgensen} on the
265: variation of lattice parameters of MgB$_{2}$ with temperature,
266: down to 11 K, indicate only a continuous decrease in the a and c
267: axis parameters with no anomalies. At the same time, experiments
268: on the suppression of superconductivity with small decrease in
269: c-parameter, obtained by Al doping \cite {slusky}, is indicative
270: of the fact that MgB$_2$ is near a structural instabilty. The
271: present infrared absorption measurements, which shows softening
272: of the phonon modes suggest that there could be a minor
273: increase in c axis parameter below $\sim $100 K. This calls for
274: further detailed structural investigations using techniques such
275: as EXAFS.
276:
277: In the case of mode at 333 cm$^{-1}$, there is also a distinct
278: softening below T$_{c}$. While the underlying reason for this
279: behaviour is not clear, we would like to point out that in the
280: experiments of Jorgensen et al \cite{jorgensen}, a distinct
281: increase in the Debye-Waller factor, U$_{33} (B)$, has been
282: observed below T$_{c}$. This increase in the thermal factor of
283: B along the z axis may have a bearing on the softening of the
284: optic mode, observed in the present investigations. Earlier
285: neutron diffraction measurements by Sato et al \cite{sato}
286: indicated an anomolous behaviour of a mode at 17 meV,
287: subsequently discounted in other experiments \cite{yildrim}. In
288: the studies by Yildrim et al \cite{yildrim}, no substantial changes in the
289: phonon density of states has been observed with the lowering of
290: temperature from 200 to 7 K. While this is contrary to the
291: present observations, this may be pointing to the sensitivity
292: of infrared absorption measurements to pick up small changes in
293: the mode frequencies with the lowering of temperature. We also
294: note that while we have seen a small softening of the 333
295: cm$^{-1}$ mode, there is no corresponding change in the width,
296: which is expected to be modified with the appearence of
297: electronic energy gap in the superconducting phase
298: \cite{shirane}.
299:
300:
301: \section{conclusions}
302:
303: To summarise, through infrared absorption measurements, optic
304: modes in MgB$_{2}$ have been identified. Infrared modes are seen
305: at 333 and 387 cm$^{-1}$, in accordance with theoretical
306: calculations \cite{kortus,pickett,andersen,satta}. While the
307: present experiments have been carried out on MgB$_2$ sample
308: characterised by sharp diffraction peaks and superconducting
309: transition, it is noted that the infrared absorption shows broad
310: features that can be accounted by invoking disorder and the
311: consequent relaxation of momentum selection rule. The broad band
312: centred at 485 cm$^{-1}$, has been identified with the peak in
313: the phonon density of states and additional broad features
314: corresponding to combination modes are also seen in the present
315: experiments. In effect, the measured absorption spectrum is an
316: indication of the phonon density of states as sampled by the
317: infrared absorption. These optic modes in the far infrared range
318: exhibit interesting temperature dependence involving a softening
319: below $\sim $ 100 K. The implications of this on a possible
320: structural anomaly needs to be investigated.
321:
322: It is seen that the infrared modes at 333 and 387 cm$^{-1}$
323: are characterised by a
324: widths of $\sim $ 40 cm$^{-1}$, which is considerably smaller
325: than the width of the Raman mode \cite{bohnen,chen,goncharov}
326: $\sim $ 200 cm$^{-1}$. This may be taken as an evidence for the
327: strong electron phonon interaction with the E$_{2g}$ mode, as
328: has been suggested by theoretical calculations
329: \cite{pickett,andersen}. However, given that in this system
330: disorder and dispersion effects seem to play a considerable
331: role, to dileneate the electronic contribution to the width and
332: to estimate the electron phonon interaction from the measured
333: widths, using Allen formula \cite{allen}, experiments need to be
334: carried out on single crystals of MgB$_2$. This will also help
335: to clarify the possible role, if any, of disorder on the
336: superconductivity in MgB$_{2}$.
337:
338: \section{acknowledgements}
339: The authors thank Prof. A.K. Sood for discussions.
340:
341: \begin{figure}
342: \centerline{\psfig{figure=fig1.eps,width=16cm,height=16cm}}
343: \caption{X-ray diffraction pattern of MgB$_2$ indexed to
344: hexagonal P6/mmm structure. The impurity lines due to MgO are
345: indicated by asterix. Also shown is the diffraction pattern of
346: crystalline B to indicate that no B features are discernable in the
347: diffraction pattern of MgB$_{2}$. The inset shows the
348: superconducting transition at 39 K.}
349: \label{Fig 1}
350: \end{figure}
351:
352: \begin{figure}
353: \centerline{\psfig{figure=fig2.eps,width=16cm,height=16cm}}
354: \caption{Panel (a) shows the absorption spectrum of MgB$_2$ at
355: 297 and 5 K respectively. Also shown are the absorption spectra
356: of the crystalline B. Panel (b) shows the
357: fits of the absorption spectrum at 297 K and 5 K in terms of sum
358: of six Gaussians and a linear background. }
359: \label{Fig 2}
360: \end{figure}
361:
362: \begin{figure}
363: \centerline{\psfig{figure=fig3.eps,width=16cm,height=16cm}}
364: \caption{Mid-infrared absorption spectrum of superconducting MgB$_2$
365: (a) synthesised from crystalline B, and (b) from amorphous B. For
366: comparison, the absorption spectra of the starting amorphous and
367: crystalline B are also shown. The absorption spectrum of MgB$_2$
368: is characterised by broad absorption bands at 485, 1040, 1442
369: and 1635 cm$^{-1}$.}
370: \label{Fig 3}
371: \end{figure}
372:
373: \begin{figure}
374: \centerline{\psfig{figure=fig4.eps,width=16cm,height=16cm}}
375: \caption{ Temperature dependence of the frequency and width of
376: the modes at 333, 387 and 485 cm$^{-1}$. The lines are guide to
377: the eye, and the arrow corresponds to the superconducting
378: transition. Notice the softening of the infrared modes below
379: $\sim $100K and a distinctive softening below T$_c$ for the 333
380: cm$^{-1}$ mode. }
381: \label{Fig 4}
382: \end{figure}
383:
384: \begin{figure}
385: \centerline{\psfig{figure=fig5.eps,width=16cm,height=16cm}}
386: \caption{Temperature dependence of the frequency and
387: width of the modes at 542, 594 and 634 cm$^{-1}$. In
388: contrast to the modes shown in Fig.4, these show a regular
389: hardening behaviour with the lowering of temperature. The filled
390: squares are the mode frequency and width for crystalline B.
391: The lines are guide to the eye. }
392: \label{Fig 5}
393: \end{figure}
394:
395:
396:
397:
398: \begin{references}
399: \bibitem{akimitsu} J. Nagamutsu, N. Nagakawa, T. Muranaka, Y.
400: Zenitani, and J. Akimitsu, Nature {\bf 410}, 63 (2001).
401:
402: \bibitem{budko} S.L. Bud'ko, G. Lapertot, C. Petrovic,
403: C.E.Cunningham, N. Anderson and P.C. Canfield, Phys. Rev. Lett.
404: {\bf 86}, 1877 (2001).
405:
406: \bibitem{schmidt} H. Schmidt, J.F. Zasadzinski, K.E. Gray and
407: D.G. Hinks, cond-mat / 0102389; G. Karapetrov, M. Iavarone,
408: W.K. Kwok, G.W. Crabtree and D.G. Hinks, cond-mat /0102312
409:
410: \bibitem{gorshunov} B. Gorshunov, C.A. Kuntscher, P. Haas, M.
411: Dressel, F.P. Mena, A.B. Kuz'menko, T. Muranaka and J. Akimitsu,
412: cond-mat / 0103164
413:
414: \bibitem{chu} B. Lorenz, R.L. Meng and C.W. Chu, cond-mat / 0102264
415:
416: \bibitem{kortus} J. Kortus, I.I. Mazin, K.D. Balaschenko, V.P.
417: Antropov, and L.L. Boyer, cond-mat / 0101446
418:
419: \bibitem{pickett} J.M. An and W.E. Pickett, cond-mat / 0102391
420:
421: \bibitem{andersen} Y. Kong, O.V. Dolgov, O. Jepsen and O.K. Andersen,
422: cond-mat / 0102499
423:
424: \bibitem{satta} G. Satta, G. Profeta, F. Bernardini, A.
425: Continenza, and S. Massida, cond-mat / 0102358
426:
427: \bibitem{hirsch} J.E. Hirch, Cond-mat /0102115
428:
429: \bibitem{baskaran} G. Baskaran, Cond-mat /0103308
430:
431:
432: \bibitem{osborne} R. Osborne, E.A. Goremychkin, A.I. Kolesnikov,
433: and D.G. Hinks, Cond-mat /0103064
434:
435: \bibitem{sato} T.J. Sato, K. Shibata, and Y. Takano, cond-mat /
436: 0102468
437:
438: \bibitem{yildrim} T. Yildrim, O. Gulseren, J.W. Lynn, C.M.
439: Brown, T.J. Udovic, H.Z. Qing, N. Rogado, K.A. Regan, M.A.
440: Hayward, J.S. Slusky, T. He, M.K. Haas, P. Khalifah, K. Inumaru
441: and R,J, Cava, Cond-mat /0103469
442:
443: \bibitem{bohnen} K.P. Bohnen, R. Heid and B. Renker, cond-mat /
444: 0103319
445:
446: \bibitem{chen} X.K. Chen, M.J. Konstantinovic, J.C. Irwin, D.D.
447: Lawriem and J.P. Franck, cond-mat / 0104005
448:
449: \bibitem{goncharov} A.F. Goncharov, V.V. Struzhkin, J. Hu, R.J.
450: Hemley, H.K. Mao, G. Lapertot, S.L. Budko, and P.C. Canfield, cond-mat /
451: 0104042
452:
453: \bibitem{jorgensen} J.D. Jorgensen D.G. Hinks and S. Short,
454: cond-mat / 0103069
455:
456: \bibitem{nogi} N. Nogi, S. Tanaka, T. Noda, and T. Hirata, Solid
457: State Commun. {\bf 111}, 447 (1999)
458:
459:
460: \bibitem{talley} C.P. Talley, L.E. Line, Jr and Q.D.
461: Overman, Jr, in Boron, Synthesis, Structure and Properties, Ed.
462: J.A. Kohn, W.F. Nye, and G.K. Gaule, (Plenum, New York, 1960),
463: p. 94.
464:
465: \bibitem{bharathi} A. Bharathi et al (to be published)
466:
467: \bibitem{sinha} S.K. Sinha and B.N. Harmon, in "
468: Superconductivity in d and f band metals", Ed. D.H, Douglass
469: (Plenum, Newyork, 1976), p. 269
470:
471: \bibitem{slusky} J.S. Slusky, N. Rogado, K.W. Reagan, M.A.
472: Hayward, P. Khalifah, T. He, K. Inumark, S. Loureiro, M.K. Haas,
473: H.W. Zandbergen and R.J. Cava, Nature, {\bf 410}, 343 (2001)
474:
475: \bibitem{shirane} J.D. Axe and G. Shirane, Phys. Rev. B {\bf 8},
476: 1965 (1973)
477:
478:
479: \bibitem{allen} P.B. Allen, Solid State Commun. {\bf 14}, 937
480: (1974).
481:
482:
483: \end{references}
484: \end{document}
485:
486: