Synthesis and transport properties of solid solutions Sr1−xKxPbO3−yFy (0  x, y  0.20)

https://doi.org/10.1016/j.jallcom.2007.10.129Get rights and content

Abstract

Polycrystalline solid solutions Sr1−xKxPbO3−yFy (0  x, y  0.20) were synthesized. Their structural properties were determined by X-ray diffraction. Resistivity and thermoelectric power were measured as functions of temperature. It was shown that a simultaneous substitution of potassium for strontium and fluorine for oxygen results in a transformation of conductivity from semiconductive to metallic at low temperatures. The results are discussed in connection with possible superconductivity in new families of doped perovskites.

Introduction

Strontium plumbate, SrPbO3, is a structural analog of another compound BaPbO3, the latter being a basis of superconducting systems BaPb1−xBixO3 (the maximal critical temperature Tc max  13 K) [1] and BaPb1−xSbxO3 (Tc max  3.5 K) [2]. Since the ionic radius of Sr2+ is smaller in comparison to that of Ba2+, the perovskite-type crystal lattice of SrPbO3 suffers an extremely strong orthorhombic distortion [3]. The ceramics BaPbO3 demonstrates metallic conductivity due to the overlap of O (2p) nonbonding and Pb–O (spσ) antibonding bands at the Fermi level [4]. In the case of SrPbO3, the distances between Pb and O ions increase and the neighboring O–Pb–O angles deviate from the value of 90° appropriate to the undistorted perovskites. These structural changes reduce the overlap between 6s and 2p orbitals [3]. Hence, electrical resistivity ρ of SrPbO3 is expected to be larger than that for BaPbO3 and this trend has been actually observed, the former material no longer being a semimetal as BaPbO3 but revealing a semiconductor-like conductivity [5].

As is well known, the search for new oxide superconductors is mostly carried out on the basis of empirical recipes [6], [7] (nevertheless, intuitively justified by an existing deep theoretical background [8], [9]!) rather than first-principles approaches. These prescriptions are similar to Matthias rules for superconducting intermetallides and alloys [10] and more recent superconductivity criteria [11], [12], [13], [14], [15], [16], [17], [18], [19].

If one considers phonons as the main bosons, intermediating in Cooper pairing, then there should be certain restrictions on the strength of the required electron–phonon interaction in actual materials. We note, that for oxides the exact nature of the inter-electron attraction is not known with confidence [8], [9], [20], [21]. Nevertheless, there is much evidence that the electron–phonon interaction is very important in high-Tc cuprates [22], [23], [24], [25] despite the observed predominantly d-wave pairing [26], [27], more appropriate to the interaction repulsive at short distances [28]. For oxides with lower Tc, there are sound reasons to consider the Cooper pairing isotropic and boson mediators as ubiquitous phonons [29], [30].

Therefore, in all cases important in the context of this study the corresponding considerations are actually, although sometimes implicitly, based on a fundamental simple idea that the electron–phonon pairing interaction determines superconductivity and is strong enough to ensure high Tc's but weak enough to avoid any reconstruction of the parent crystal lattice followed by a concomitant shrinkage of a Fermi surface (FS) and an inevitable drop of Tc [31], [32]. Since a current-carrier screening reduces the matrix elements Vel–ph of the electron–phonon interaction (see, e.g., [33]), criteria often reduce to a peculiar balance between them and the electronic densities of states (electronic DOSes). Magnetic components should be included into estimates, when necessary, so that the interplay becomes more involved and interesting [16]. On the other hand, the complexity of the problem leads to the inability of existing microscopic theories either to predict new superconducting compounds or to calculate corresponding Tc's with a sufficient accuracy [34]. It is clear that this reasoning is valid for oxide and related materials as well.

Thus, according to general considerations based on the crucial role of the strong electron–phonon interaction, one might expect an appearance of superconductivity in those multi-component oxide systems with variable compositions (bertollides [35]) that occupy phase diagram areas near semiconducting phases. It means that the Cooper pairing should develop in the vicinity of the boundary beyond which a parent electron spectrum is rebuilt and the crystal lattice distorted [36], [37]. In particular, it is reasonable to seek for superconducting instability in chemically modified and electrically doped SrPbO3. Interest in studying of SrPbO3 is also stimulated by resistive and magnetic peculiarities, suspiciously similar to manifestations of superconductivity and observed in complex lead oxides. In particular, we mean some evidence in favor of room-temperature superconductivity in Cu24Pb2Sr2Ag2Ox [38] and AgxPb6CO9+β [39], as well as resistive anomalies in the temperature range 190 K < T < 270 K for a composition Cu24Pb2Sr2Ag2Ox [40]. Recently, our group has demonstrated the existence of a diluted superconducting phase in solid solutions Sr1−xLaxPbO3−δ hidden for bulk measurements but uncovered by tunnel studies [41].

Besides the strontium plumbate as a basic material, a number of its derivatives with various kinds of substitutions were studied. Thus, a replacement of lead by bismuth, contrary to what happens in the BaPbO3 case [1], results in a current-carrier localization for pristine [42] and doped SrPbO3 [43]. At the same time, doping by antimony almost does not influence electrical conductivity for low concentrations of substituting ions [44]. Heterovalent substitution of La for Sr transfers the system into the metallic state but does not lead to the appearance of the superconducting phase [43], [45], [46]. Finally, insertion of K+ instead of Sr2+ results in an extremely narrow homogeneity region and an electron spectrum transformation, which may reflect either dielectrization or localization effects [47].

All presented doping cases for strontium plumbate are connected to the cation sub-lattice modification. On the contrary, substitution of the oxygen anions by halogens, especially by fluorine, has not been adequately studied, although such substitutions turned out to be quite effective for superconducting cuprates. Thus, upon inserting of fluorine into a semiconducting La2CuO4 halogen ions intercalate into interstitial sites and CuO2 layers oxidize. As a result a superconducting phase emerges with Tc  40 K [48]. Fluorination of reduced HgBa2Ca2Cu3O8+δ leads to an enhancement of Tc up to 138 K [49], which is conspicuously higher than the maximal Tc  134 K achieved in the oxygen flow. When Nd2CuO4 is doped, fluorine substitutes oxygen ions in the crystal lattice sites that leads to the reduction of CuO2 layers and superconducting pairing of negatively charged current carriers [50]. It is worthwhile to indicate the existence of other F-containing superconducting phases Sr2CuO2F2+δ (Tc  46 K) [51] and YBa2Cu3O6F2 (Tc  94 K) [52].

To fluorinate oxide compounds, various agents are used: gaseous F2, XeF2, NH4F and others, which make it possible to carry out a process at the room temperature or by heating up to relatively low temperatures 100–200 °C. Fluorination using transition-metal difluorides MF2 (M = Cu, Ni, Zn, Ag) requires higher T and results in a contamination of the final product by oxides of these metals. High-T synthesis under large pressure is also an effective method of oxide fluorination. For example, a superconducting phase Sr2CuO2F2+δ has been observed using such a technique [53].

Taking all the aforesaid into account, we formulated our goal as a synthesis of the system Sr1−xKxPbO3−yFy with a subsequent investigation of its structural and transport properties. One might expect that a fluorination would reduce the orthorhombic distortion of the crystal lattice for SrPbO3. While a tolerance factor for SrPbO3 calculated using a Sr ionic radius and a coordination number 12 [54] comprises 0.92, its value becomes 0.94 for the composition Sr0.8K0.2PbO2.8F0.2. Therefore, one may expect an increase of the metallic character of resistivity and a possible development of a superconducting state for low enough T.

Section snippets

Experimental

Polycrystalline samples Sr1−xKxPbO3−yFy (x = 0.00, 0.05, …, 0.30) were prepared using a solid state synthesis with Sr(NO3)2, KF and PbO as starting reagents. Salts and lead oxide (II) weighted in the needed proportion were ground in an agate mortar under an ethanol layer and were thereafter heated in an aluminum crucible while steadily increasing T up to 700 °C. Then the reacting mixture was sustained several hours at the indicated temperature, once more grounded and heated in air at 900 °C for 48 

Results and discussion

It is known that in fluorinated oxide materials, superconducting cuprates including, one or several coexisting mechanisms of the oxygen for fluorine substitution may be realized [53]. Specifically, (i) one O atom may be substituted by one F atom; (ii) one O atom may be substituted by two F atoms; (iii) F atoms may be located in interstitial sites. All three mechanisms lead to substantial structural {(ii) and (iii)} and electronic-configuration {(i) and (iii)} changes for a fluorinated compound.

Conclusions

We carried out studies, which made it possible to find limits for the existence of solid solutions Sr1−xKxPbO3−yFy while substituting K for Sr and F for O. These limits constitute up to 20 at%. Substitutions in cationic and anionic sub-lattices transform a semiconductor-like conductance into a metallic one in the low-T region when substitute concentrations increase. The thermoelectric power changes its sign to positive upon doping. The totality of data argues for the existence of several

Acknowledgements

V.A.D. and A.M.G. are grateful to Kasa im. Józefa Mianowskiego and Fundacja na Rzecz Nauki Polskiej for support of their visits to Warsaw University.

References (62)

  • W.T. Fu et al.

    Solid State Commun.

    (1995)
  • M. Itoh et al.

    Physica C

    (1992)
  • W.E. Pickett

    Physica B

    (2001)
  • P.W. Anderson

    Physica C

    (2007)
  • G. Gusman et al.

    J. Phys. Chem. Sol.

    (1965)
  • A.M. Gabovich et al.

    Phys. Rep.

    (2002)
  • D. Djurek et al.

    Physica C

    (2001)
  • M.V. Lobanov et al.

    Mater. Res. Bull.

    (1997)
  • W.T. Fu et al.

    Physica C

    (1998)
  • A. Yamamoto et al.

    Mater. Res. Bull.

    (1999)
  • V.A. Drozd et al.

    J. Alloys Compd.

    (2004)
  • V.A. Drozd et al.

    J. Alloys Compd.

    (2002)
  • B. Chevalier et al.

    Physica C

    (1990)
  • C. Greaves et al.

    Curr. Opin. Solid State Mater. Sci.

    (1998)
  • A.M. Gabovich et al.

    Sov. Phys. Usp.

    (1986)
  • R.J. Cava et al.

    Nature

    (1989)
  • L.F. Mattheiss et al.

    Phys. Rev. B

    (1983)
  • A.W. Sleight

    Chemtronics

    (1987)
  • A.J. Leggett

    Nature Phys.

    (2006)
  • G. Gladstone et al.
  • I.O. Kulik

    Zh. Eksp. Teor. Fiz.

    (1964)
  • I.O. Kulik

    Zh. Eksp. Teor. Fiz.

    (1974)
  • H. Stern

    Phys. Rev. B

    (1973)
  • H. Stern

    Phys. Rev. B

    (1975)
  • I.O. Kulik

    Fiz. Nizk. Temp.

    (1976)
  • A.M. Gabovich

    Ukr. Fiz. Zh.

    (1977)
  • A.M. Gabovich et al.

    Fiz. Tverd. Tela

    (1981)
  • C. Buzea et al.

    Supercond. Sci. Technol.

    (2005)
  • J.D. Dow et al.

    J. Vac. Sci. Technol. B

    (2006)
  • A. Cho

    Science

    (2006)
  • A. Lanzara et al.

    Nature

    (2001)
  • Cited by (2)

    View full text