Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.issp.ac.ru/lhpp/PapersBashkin/122.pdf Äàòà èçìåíåíèÿ: Fri Oct 26 12:43:44 2007 Äàòà èíäåêñèðîâàíèÿ: Tue Oct 2 06:39:34 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: http astrokuban.info astrokuban

PHYSC 51807 16 January 2006 Disk Used

ARTICLE IN PRESS

No. of Pages 3, Model 5+

1

Physica C xxx (2006) xxx­xxx www.elsevier.com/locate/physc

2 3 4 5 6 7

I.O. Bashkin *, V.G. Tissen, M.V. Nefedova, E.G. Ponyatovsky
Institute of Solid State Physics, Russian Academy of Sciences, Chernogolovka, Moscow 142432, Russia Received 19 December 2005; accepted 26 December 2005

8 Abstract 9 10 11 12 13 14
The superconducting transition temperature of the equiatomic Tc(P) dependence is found to achieve a maximum around 42 GPa pressure range 5­10 GPa. This behavior of Tc(P) is similar to that calculations suggests that the features of the Tc(P) behavior both in the high-pressure anomalies in their phonon spectra. ñ 2006 Published by Elsevier B.V.

15 PACS: 74.62.Fj

16 Keywords: Superconducting transition temperature; High-pressure; ZrNb alloy 17

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

The electronic band structure calculations rather early resulted in a conclusion on the pressure-induced transfer of the s-electrons to the d-band in the metals opening the series of the periodic table [1,2]. As the accuracy of the model calculations increased, the pressure dependence of the band structure and the properties of metals as well as a correlation between the band structure and the stability of the crystal lattice have been a subject of many works. Transitions to the body-centered cubic phase in zirconium and hafnium under pressure have also been related to the s ! d electron transfer [3­5]. Formation of the bcc phase in Zr [6] and Hf [7] is concomitant with an abrupt increase of the superconducting transition temperature, Tc. Akahama et al. [8] noticed that the bcc b-Zr phase formed at P = 30 GPa had approximately the same values of the specific volume and Tc as bcc Nb at atmospheric pressure. They assumed therefore that the Group IV metal, Zr, becomes a close analogue of the Group V metal, Nb, due to increasing occupancy of the d-band on compression. The
*

UNCORRECTED

ZrNb alloy is measured in dependence on pressure to 56.4 GPa. The and to have an anomalous behavior with a shallow minimum in the earlier observed on pure metal Nb. A comparison with recent model pure Nb and in the ZrNb alloy are of the same nature and result from

Corresponding author. Fax: +7 096 5249701. E-mail address: bashkin@issp.ac.ru (I.O. Bashkin).

experimental data on the high-pressure behavior of the crystal structure and the superconducting temperature of the Zr­Ti [9,10] and Zr­Hf [11] alloys were also discussed in relation to the s ! d transfer in Zr and Hf under pressure. These speculations, however, are not well supported by the theoretical estimate of the s ! d transfer in Zr [5] indicating that the number of the electrons transferred in the relevant pressure range is considerably less than unity for all phases. A correlation between the number of the electrons transferred and Tc was not theoretically considered for the Group IV metals as well as for their alloys until recently. The first estimation of this kind was given for the equiatomic bcc ZrNb and TiV alloys whose band structure, the d-band occupancy and Tc were calculated at several pressure values [12]. The calculation based on few experimental points measured on ZrNb in a pressure range limited to 25 GPa [13] therefore a more detailed experimental test of the correlation between the actual and estimated Tc values in a broader pressure range was of interest. Further, accurate data for bcc ZrNb was expected to extend the experimental basis for conclusions on the Tc(P) behavior of other related bcc phases.

PROOF

Superconducting transition temperature of the equiatomic ZrNb alloy at pressures to 56.4 GPa

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

0921-4534/$ - see front matter ñ 2006 Published by Elsevier B.V. doi:10.1016/j.physc.2005.12.065


PHYSC 51807 16 January 2006 Disk Used 2

ARTICLE IN PRESS
I.O. Bashkin et al. / Physica C xxx (2006) xxx­xxx

No. of Pages 3, Model 5+

16

ZrNb

UNCORRECTED

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94

This work presents the Tc(P) dependence of the bcc ZrNb alloy measured up to P = 56.4 GPa and compared to the available model calculations. To prepare the ZrNb alloy, initial Zr and Nb metals of 99.95 at.% purity were cut to chips, mixed up and compacted to a cylindrical ingot that was subjected to multiple electron-arc melting in vacuum. The final composition of the alloy was tested using the JXA-5 local X-ray microanalyzer, it was stoichiometric within the experimental accuracy of ±1 at.%. The bcc single-phase state of the alloy was checked using the X-ray diffraction. High-pressure was generated with a diamond-anvil cell. The pressure-transmitting medium was the 4:1 methanol­ ethanol mixture. The superconducting transitions were recorded as jumps in the thermal dependence of the magnetic susceptibility, v(T), measured on heating. To eliminate the effects due to the mechanical relaxation of the press upon cooling to helium, the pressure was determined from the shift of the ruby luminescence line after recovery of the press to room temperature at the end of the cooling/ heating cycle. Other details of the device and the experimental procedure are described elsewhere [7,10]. The measured Tc values are plotted in dependence on pressure in Fig. 1 (black circles). The Tc values are determined from the experimental v(T) curves as intersection points between the steepest tangent to the curve and the linear extension of the high-temperature section of the curve, as illustrated in the inset. The widths of the superconducting transitions are about 0.5 K at low pressures and increase twice at highest pressures. The prominent features of the Tc(P) curve are a drastic decrease of the slope in a narrow pressure interval of 5­10 GPa and a broad maximum around P = 42 GPa. Above the maximum, Tc(P) is decreasing with increasing pressure. Earlier data measured on a nearly stoichiometric Zr­Nb alloy in the range to 25 GPa [13] are shown in Fig. 1 in

3.3

14

12

(arb.un.)

14.3 GPa

Tc

10 0 10

present [13] [12]

3.2
12 13 14 T (K) 15

20

30 P (G Pa)

40

50

60

Fig. 1. Pressure effect on the superconducting transition temperature in the ZrNb alloy (black circles). Open circles are for the earlier data to 25 GPa [13]. Calculated data [12] are shown with crosses and the dotted line. The dashed line is for the d-band occupancy, Nd (right axis). Determination of the Tc values from the v(T) curves is illustrated for the 14.3 GPa isobar.

Nd

Nd

open circles. The agreement between the present and earlier data is mainly quantitative, but the low-pressure Tc(P) anomaly was not found by Kawamura et al. [13]. This seems to be due to scanty experimental points in the relevant pressure interval in Ref. [13]. The dotted line in Fig. 1 is drawn through the calculated Tc values (crosses) presented in Table 3 of Ref. [12], the dashed line represents the pressure dependence of the d-band occupancy, Nd, of the ZrNb alloy taken as the average of the d-electron numbers listed for the constituent Zr and Nb atoms in Table 1 [12]. One can see from Fig. 1 that the calculated Tc(P) and Nd(P) dependencies demonstrate an approximately linear increase with pressure. The only deflection from a smooth dependence is the Tc value calculated at P = 21.72 GPa; this deflection was not discussed in Ref. [12]. On the basis of this relationship between Tc and Nd, Selvamany et al. [12] concluded that the continuous increase in Tc of the ZrNb alloy is related to the continuous electron transfer from the sp-states to the d-states. To support this conclusion, Selvamany et al. [12] indicated that both the s­d electron transfer and the Tc increase in the ZrNb alloy under pressure are more rapid than in the TiV analogue. A comparison of the calculated and present data shows their proximity in the pressure range 8­25 GPa, but the experimental Tc(P) curve as a whole has a more complex shape. The calculation does not reproduce a distinct anomaly at 5­13 GPa in the experimental curve resulting in a shallow minimum near P = 10 GPa. We do not think that the lacking anomaly in the calculated data [12] is due to a usual simplifying assumption that the ZrNb alloy is an atomically ordered bcc phase whereas the actual Zr­Nb alloys are disordered solid solutions; rather this is due to insufficient accuracy of the calculation. Finally, calculation [12] is limited to P = 26.29 GPa therefore it is not clear whether this calculation method can reproduce the highpressure Tc(P) maximum. Further insight into the Tc(P) behavior of the ZrNb alloy can be achieved from comparison with the experimental and theoretical data obtained for pure Nb [14­17]. The experimental Tc(P) curve for Nb has similar features, i.e., a distinct shallow minimum at pressures around 5 GPa and decreasing Tc(P) above P =60GPa [14]. Struzhkin et al. [14] suggested that these anomalies arose from changes in the Fermi surface topology at 5 and 60­70 GPa. Subsequent theoretical works [15­17] presented calculation of the band structure of Nb at atmospheric and high pressures in order to explain the anomalous superconducting behavior. All three calculations reported no marked changes in the Fermi surface topology below 40­50 GPa and an essential change in the topology at higher pressures. The latter effect was related to the decreasing Tc(P) behavior above 60 GPa in Refs. [15,16]. The Tc(P) anomaly around 5 GPa either remained unexplained [15] or was attributed to very subtle changes in the topology of the Fermi surface [16]. Wierzbowska et al. [17] supplemented the band structure calculation with calculation of the phonon spectra of Nb. They found that the phonon spectra had an anomalous pressure

95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151

T c (K )

PROOF


PHYSC 51807 16 January 2006 Disk Used

ARTICLE IN PRESS
I.O. Bashkin et al. / Physica C xxx (2006) xxx­xxx

No. of Pages 3, Model 5+

3

176 Acknowledgements

177 This work was supported by the RFBR Grant No. 03178 02-17005 and by the RAS program ``Thermal physics and 179 mechanics of the intense impulse effects''.

UNCORRECTED

152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175

dependence: low-frequency phonon peaks close to the Cpoint of the Brillouin zone (Kohn anomalies) were observed in some pressure range below 60 GPa, and the phonon linewidths increased for some modes in the same pressure interval. The occurrence of the Kohn anomalies is connected with the increase of the electron­phonon coupling constant and the Tc values in the pressure range 5­60 GPa [17]. This interpretation is rather uncommon; it can also explain the high-pressure superconducting behavior of the ZrNb alloy. So far as the phonon spectrum is determined primarily by the bcc lattice, one may think that the spectrum anomalies similar to those found in Ref. [17] also occur if the electron concentration is somewhat varied by alloying. It is expected therefore that the Tc(P) dependence of many bcc Zr­Nb alloys and, probably, of the virtual bcc Zr phase in the range of its instability at P < 30 GPa is a curve with a maximum below 60 GPa and an anomaly at markedly lower pressures. So, the Zr­Nb system where the electron concentration and the position of the Fermi level with respect to the d-band features are readily varied by a change of the alloy composition is a convenient test for the present-day models dealing with the effect of the electron band structure, the d-band occupancy and the phonon spectrum on the superconducting transition temperature.

References
[1] G.M. Gandelman, V.M. Ermachenko, Ya.B. Zeldovich, Zh. Exp. Theor. Phys. 44 (1963) 386. [2] J.C. Duthie, D.G. Pettifor, Phys. Rev. Lett. 38 (1977) 564. [3] H. Xia, S.J. Duclos, A.L. Ruoff, Y.K. Vohra, Phys. Rev. Lett. 64 (1990) 204. [4] H. Xia, A.L. Ruoff, Y.K. Vohra, Phys. Rev. B 44 (1991) 10374. [5] R. Ahuja, J.M. Wills, B. Johansson, O. Eriksson, Phys. Rev. B 48 (1993) 16269. [6] Y. Akahama, M. Kobayashi, H. Kawamura, J. Phys. Soc. Japan 59 (1990) 3843. [7] I.O. Bashkin, M.V. Nefedova, V.G. Tissen, E.G. Ponyatovsky, JETP Lett. 80 (2004) 655. [8] Y. Akahama, M. Kobayashi, H. Kawamura, J. Phys. Soc. Japan 60 (1991) 3211. [9] I.O. Bashkin, V.G. Tissen, M.V. Nefedova, A. Schiwek, W.B. Holzapfel, E.G. Ponyatovsky, JETP Lett. 73 (2001) 75. [10] I.O. Bashkin, V.K. Fedotov, M.V. Nefedova, V.G. Tissen, E.G. Ponyatovsky, A. Schiwek, W.B. Holzapfel, Phys. Rev. B 68 (2003) 054401. [11] I.O. Bashkin, M.V. Nefedova, E.G. Ponyatovsky, V.G. Tissen, JETP Lett. 78 (2003) 80. [12] P. Selvamani, G. Vaitheeswaran, V. Kanchana, M. Rajagopalan, Physica C 370 (2002) 108. [13] H. Kawamura, J. Wittig, B. Bireckoven, W. Buckel, Physica B 126 (1984) 485. [14] V.V. Struzhkin, Y.A. Timofeev, R.J. Hemley, H.K. Mao, Phys. Rev. Lett. 79 (1997) 4262. [15] S.F. Ostanin, V.Yu. Trubitsin, S.Yu. Savrasov, M. Alouani, H. ´ Dreysse, Comp. Mater. Sci. 17 (2000) 202. [16] J.S. Tse, Z. Li, K. Uehara, Y. Ma, R. Ahuja, Phys. Rev. B 69 (2004) 132101. [17] M. Wierzbowska, S. Gironcoli, P. Giannozzi, .

180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213

PROOF