On-line version ISSN 1678-4448
Braz. J. Phys. vol.33 no.4 São Paulo Dec. 2003
John D. DowI; Dale R. HarshmanI,II
IDepartment of Physics, Arizona State University, Tempe, Arizona 85287 U.S.A
IIPhysikon Research Corporation, P. O. Box 1014, Lynden, WA 98264 U.S.A*
The widely held notion that high-temperature superconductivity originates in the cuprate-planes is proven to be faulty. In the cuprates such as YBa2Cu3O7, we argue that the superconductivity resides in the BaO layers. This superconductivity is s-wave, not d-wave, in the bulk. The trio of ruthenate compounds, doped Sr2YRuO6, GdSr2Cu2RuO8, and Gd2-zCezSr2Cu2RuO10 all superconduct in their SrO layers, which is why they have almost the same ~49 K onset temperatures for superconductivity.
1 Faulty evidence for cuprate-plane superconductivity
Two of the most-cited papers in contemporary physics claim to show that the cuprate-planes of YBa2Cu3Ox superconduct [1,2]. However, there are problems with both papers that have not been well-recognized: (i) the evidence of superconductivity in the cuprate-planes comes from a jump in charge which was evident in the work of Cava et al., but not in the work of Jorgensen et al.; and (ii) the studies of Jorgensen et al. claim to confirm the data of Cava et al., but actually do not in the most important way: Jorgensen does not have the Cava jump. In fact, a closer examination of the data reveals that only one datum is responsible for the jump in charge that is purportedly evidence for cuprate-plane superconductivity, and this datum was not reproduced in the data of Jorgensen et al. or (to our knowledge) elsewhere. This startling fact has been missed by many people because a continuous line was drawn through rather sparse data and conveyed the impression that the data are much denser than they are. In other words, the concept of cuprate-plane superconductivity has not been confirmed and rests on only one unreproduced datum. (See Fig. 1.)
This is important to realize because the superconducting layers in most (and perhaps all) high-temperature superconductors are not the cuprate-planes, as was implied by the now-infamous fictitious jump in cuprate-plane Cu charge .
Perhaps an independent confirmation of the problem with cuprate-plane superconductivity comes from the scanning tunneling microscopy data on Bi2Sr2CaCu2O8 of the Illinois group [4,5]. (See Fig. 2). The surface layer of this compound, when it is cleaved, is the BiO layer. Underneath that is the SrO layer, and then a CuO2 plane; after that comes a Ca layer and a second CuO2 plane. Apparently the Illinois workers have imaged the BiO surface layer, and then a CuO2 layer that is exposed by a step protruding from the side of the sample, apparently the second CuO2 layer beneath the surface. An examination of their data reveals that their BiO layer looks like most others, with a U-shaped feature that is indicative of a layer nearby a superconducting layer. But their protruding CuO2 plane does not look at all like a superconductor, and instead of a U-shaped feature, seems to have a band-gap with no density of states in the gap. The Illinois workers have interpreted their observations as evidence of d-wave superconductivity in the cuprate-planes, but this interpretation is based primarily on the facts that (i) they assume that the cuprate-planes superconduct, and (ii) faced with the fact that their data do not exhibit the expected U-shape (with a sharper U than BiO's expected for s-wave superconductivity), they postulate that they have evidence of d-wave superconductivity. But the work of Klemm  and Li  provides a convincing demonstration that the superconductivity in Bi2Sr2CaCu2O8 is s-wave in character, and we have found , using the bulk probe of muon spectroscopy, that the superconductivity of YBa2Cu3O7 is also s-wave in character with an uncertainty of less than 4×10-6. It is exceedingly unlikely that, in the bulk, YBa2Cu3O7 could have s-wave pairing, while Bi2Sr2CaCu2O8 has d-wave  (which cannot be reconciled with other measurements ). Hence the better explanation of the facts about the CuO2 layers of the Illinois group is that they feature a band-gap in the CuO2 layers, and those layers do not initiate superconductivity.
2 Charge transfer in cuprates
The layer charges of the three layers of YBa2Cu3Ox, namely the CuO chain layers, the BaO layers, and the cuprate-planes (combined with the rare-earth charge) have been extracted from the data (using the bond-valence-sum method ), and have the following features: (1) they are all virtually linear in x and almost the same as the layer charges for PrBa2Cu3Ox; (2) the CuO chain layers have decreasing charge as oxygen content x increases; (3) the BaO layers and the combined CuO2/Rare-earth/CuO2 layers have increasing charge with oxygen content x; and (4) only the BaO layers have charges that appear to change sign as the oxygen content x increases beyond x=6.4, and the superconductivity sets in . These facts cause us to assign the superconductivity of YBa2Cu3O7 to the BaO layers.
The same facts also caused us to believe that PrBa2Cu3O7 would superconduct , which was later show [13,14].
2.1. YBa2Cu3O7 superconductivity: s-wave
We can ask if our data indicate that the superconductivity of YBa2Cu3O7 is s-wave or d-wave in character. Until rather recently, there was a nearly unanimous opinion that the bulk superconductivity is s-wave [15-18], but more recently the data have been reinterpreted as having d-wave character  (although there has been no evidence of a quantitative fit of the d-wave theory to the bulk YBa2Cu3O7 data ).
Figure 3 shows fits to the muon penetration depth extracted from m+SR data and extrapolated to H=0. Those data for H=0.05, 1.0, 3.0, and 6.0 Tesla were all fit with a single strong-coupling London model s-wave curve, after the flux-flow was accounted for. (In earlier 1989  work, which described the data with a similar strong-coupling s-wave model, the flux was pinned.) Moreover, the probability that the d-wave model fits the data as well a the s-wave model was found to be of order one in a million . In the bulk (which is what the muons probe), the superconductivity is definitely s-wave in character.
2.2. Other cuprates than YBa2Cu3O7
Independent evidence indicating that the superconducting layers in the cuprate materials are the BaO or SrO layers, not the cuprate-planes, is afforded by the HgBa2Can-1CunO2n+2 superlattices as functions of n, and by the n=1 compound of this class as a function of pressure. The layer charges behave the same for the superlattices versus n and for the n=1 compound versus pressure p [21-23]: The charges of the BaO layers increase with increasing number of layers n and pressure p, while the Hg layer-charges remain constant, and the charges of the cuprate-planes decrease. Since Tc increases with the number of layers n and with pressure p, the superconductivity must originate in the BaO layers (whose charges also increase), not in the cuprate-planes (whose charges decrease) .
3 Sr2YRuO6 doped with Cu
Sr2YRuO6 is an interesting compound because, upon doping with Cu, it begins to superconduct around ~49 K and becomes fully superconducting at ~23 K . It also has two superconducting sister compounds that contain cuprate-planes, which superconduct slightly below about ~49 K : GdSr2Cu2RuO8 and Gd2-zCezSr2Cu2RuO10 . The near coincidence of the onset temperatures for superconductivity in the three ruthenate compounds, which we term the O6's (as in Cu-doped Sr2YRuO6), the O8's (as in GdSr2Cu2RuO8), and the O10's (as in Gd2-zCezSr2Cu2RuO10), makes this class of three types of compounds especially worthwhile to investigate.
The simplest of these compounds is Sr2YRuO6 which is a two-layer compound with each pair of (SrO)2 layers having a YRuO4 layer in between. (See Fig. 4). Since this compound has only two kinds of layers, it is rather straightforward to select the superconducting layer: namely the SrO layer, which is the one without a strong magnetic field, and the one analogous to BaO. The YRuO4 layer is ferromagnetic in its a-b plane, with its magnetic moments stacked antiferromagnetically along the c-axis, and oriented in the ±(1,1) directions of the a-b plane. Muons in this material stop at one of two nearly identical sites: (i) the mO(1,2) site which is actually two sites (due to the difference between Y and Ru) near the center of the YRuO4 layer; and (ii) the mO(3) site, which is about midway between two oxygen ions on the edge of a SrO plane. Clearly the YRuO4 layer is highly magnetic (and hence rather hostile to superconductivity), while the SrO layer has an average magnetic field of zero and is the locus of superconductivity.
The muon data for the SrO-plane site (mO(3)) show a time-dependence that reveals (i) an onset of superconductivity near ~49 K , (ii) the onset of spin-glass behavior at »29.3 K , and (iii) the onset of diamagnetism at lower temperatures . Complementing the muon data are resistivity data, which show that the resistance vanishes at Tc »23 K , at which temperature the superconductivity becomes complete. Clearly Sr2YRu1-uCuuO6 is a superconductor, whose superconductivity originates at ~49 K and becomes complete at ~23 K when the Ru librations turn off.
Doped Sr2YRuO6 has a sister compound with Gd+3, Cu-doped Ba2GdRuO6, which does not superconduct. We attribute its non-superconductivity to the pair-breaking by Gd+3, which has L=0 and J ¹ 0, and breaks Cooper pairs located in the adjacent SrO layers. Hence doped Ba2GdRuO6 does not superconduct , but its sister compound, doped Sr2YRuO6, does superconduct in its SrO layers. For similar reasons, Gd2-zCezCuO4 does not superconduct, but many other (Rare-earth)2-zCezCuO4 compounds do superconduct.
4 GdSr2Cu2RuO8 and Gd2-zCezSr2Cu2RuO10
Sr2YRu1-uCuuO6 has two superconducting sister compounds, GdSr2Cu2RuO8 and Gd2-zCezSr2Cu2RuO10. They superconduct at nearly the ~49 K onset temperature for Sr2YRu1-uCuuO6's superconductivity - raising the possibility that the superconductivity of all three materials originates in the same physics .
Of course this raises the question of what roles do the cuprate-planes have in the superconductivity of these compounds? Obviously the answer is none" for doped Sr2YRuO6, which has no cuprate-planes and fewer than 1% Cu atoms even capable of participating in any cuprate-plane-like behavior. This, of course, raises the question of whether the cuprate-planes even superconduct in these other materials, GdSr2Cu2RuO8 and Gd2-zCezSr2Cu2RuO10. (We shall see that they do not).
First we examine GdSr2Cu2RuO8, which does superconduct. Do its cuprate-planes superconduct? Or do its SrO layers superconduct? Or both?
To address these questions, we analyze magnetic resonance data taken  on GdSr2Cu2RuO8 at T=120 K (above Tc »49 K) at a frequency of 20 GHz, and with fields Hrf^Hdc. (Hrf is the radio-frequency field; we also refer to the direct-current field Hdc as H). These data can be decomposed into two peaks, one associated with the Gd electron spin resonance (ESR), and the other associated with a Cu signal which is either an anti-ferromagnetic resonance or a weak ferromagnetic resonance (the equipment does not determine which) . See Fig. 5 . Our point is that, if the Cu in this compound is magnetic and resonating, the cuprate-planes almost certainly do not superconduct.
There should be no dispute over the well-known Gd ESR peak, so we shall discuss the identification of the Cu signal" primarily. The signal we term the Cu signal" is certainly due to Cu and not due to Ru (the only other possible magnetic ion that could resonate in GdSr2Cu2RuO8). A similar Cu signal is observed in GdSr2Cu2NbO8, which has no Ru (and does not superconduct). In that material, the resonance must be associated with Cu, since the Gd resonance is identified and there is no other magnetic ion.
The Cu signal" is Gd2-zCez in GdSr2Cu2RuO10 , and persists up to above 150 K although Ru does not give a magnetic resonance above ~133 K, which mean that the Cu signal" is due to Cu, not Ru.
To strengthen this argument, we have studied nine related compounds containing Ru, and we see no Ru magnetic resonance in any of them: these include the superconductors Gd1.5Ce0.5Sr2Cu2RuO10, Eu1.5Ce0.5Sr2Cu2RuO10, GdSr2Cu2RuO8, EuSr2Cu2RuO8, and Sr2YRu1-uCuuO6, and the magnetically-ordered non-superconductors SrRuO3, Sr3Ru2O7, GdSr2Cu2NbO8, and Ba2GdRu1-uCuuO6 .
Gd2-zCezSr2Cu2RuO10 behaves similarly to GdSr2Cu2RuO8, and also superconducts in its SrO layers, not in its cuprate-planes .
We have challenged the widely held notion that the cuprate-planes are the carriers of high-temperature superconductivity, and have instead proposed that the superconducting elements of most superconductors are not the cuprate-planes, but instead the BaO or SrO layers. (Some materials may also involve interstitial oxygen , which we have not discussed here.)
We have examined the three materials, doped Sr2YRuO6, GdSr2Cu2RuO8, and Gd2-zCezSr2Cu2RuO10, and we have found extremely powerful evidence that all three superconduct in their SrO layers, and not in their cuprate-planes.
A complete microscopic theory of high-temperature superconductivity will be presented elsewhere.
We are grateful to the U. S. Office of Naval Research for their financial support (Contract No. N00014-03-1-0375).
 R. J. Cava, A. W. Hewat, E. A. Hewat, B. Batlogg, M. Marezio, K. M. Rabe, J. J. Krajewski, W. F. Peck, Jr., and L. W. Rupp, Jr., Physica C, 165, 419 (1990); [ Links ]R. Cava, Synthesis and crystal chemistry of high-Tc oxide superconductors, in Processing and Properties of High-Tc Superconductors, Vol. 1, edited by S. Jin (World Scientific, Singapore, 1993), p. 1 et seq., especially p. 12. [ Links ]
 J. D. Jorgensen, B. W. Veal, A. P. Paulikas, L. J. Nowicki, G. W. Crabtree, H. Claus, and W. K. Kwok, Phys. Rev. B 41, 1863 (1990); [ Links ]J. D. Jorgensen, Phys. Today, 34 (June, 1991). [ Links ]
 In some superconductors, such as HgBa2Cun-1CunO2n+2, the holes are clearly accumulating in the BaO layers rather than in the cuprate planes. See H. A. Blackstead and J. D. Dow, Solid State Commun. 95, 613 (1995). [ Links ]
 S. Misra, S. Oh, D. J. Hornbaker, T. DiLuccio, J. N. Eckstein, and A. Yazdani, Phys. Rev. Lett. 89, 087002 (2002). [ Links ]
 Misra et al. claim to image a cuprate-plane protruding from the side of a superconductor. For a discussion of scanning tunneling microscopy of the surfaces of bulk superconductors, see S. H. Pan, J. P. O'Neal, R. L. Badzey, C. Chamon, H. Ding, J. R. Engelbrecht, Z. Wang, H. Eisaki, S. Uchida, A. K. Gupta, K.-W. Ng, E. W. Hudson, K. M. Lang, and J. C. Davis, Nature 413, 282 (2001). [ Links ]
 Q. Li, Y. N. Tsay, M. Suenaga, R. A. Klemm, G. D. Gu, and N. Koshizuka, Phys. Rev. Lett. 83, 4160 (1999); [ Links ]Q. Li, Y. N. Tsay, M. Suenaga, G. D. Gu, and N. Koshizuka, Physica C282-287, 1495 (1997). [ Links ]
 D. R. Harshman, W. J. Kossler, X. Wan, A. T. Fiory, A. J. Greer, D. R. Noakes, C. E. Stronach, E. Koster, A. Erb, and J. D. Dow, Nodeless pairing state in single-crystal YBa2Cu3O7." in press. [ Links ]
 D. R. Harshman, R. N. Kleiman, M. Inui, G. P. Espinosa, D. B. Mitzi, A. Kapitulnik, T. Pfiz, and D. L. Williams, Phys. Rev. Lett. 67, 3152 (1991). [ Links ]
 I. D. Brown and D. Altermatt, Acta Crystallogr., sect. B: Struct. Sci. B41, 244 (1985); [ Links ]D. Altermatt and I. D. Brown, Acta Crystallogr., Sect. B: Struct. Sci. B41, 241 (1985); [ Links ]I. D. Brown, J. Solid State Chem. 82, 122 (1989); [ Links ]I. D. Brown and K. K. Wu, Acta Crystallogr., Sect. B: Struct. Crystallogr. Cryst. Chem. B32, 1957 (1976); [ Links ]I. D. Brown, Structure and Bonding in Crystals, Vol. II, edited by M. O'Keefe and A. Navrotsky, pp. 1-20 (Academic Press, New York, 1980). [ Links ]
 The charges on the various layers are given in J. D. Dow, H. A. Blackstead, and D. R. Harshman, Physica C364-365, 74 (2001). [ Links ]
 H. A. Blackstead and J. D. Dow, Phys. Rev. B51, 11830 (1995). [ Links ]
 H. A. Blackstead, D. B. Chrisey, J. D. Dow, J. S. Horwitz, A. E. Klunzinger, and D. B. Pulling, Phys. Lett. A207, 109 (1995). [ Links ]
 H. A. Blackstead, J. D. Dow, D. B. Chrisey, J. S. Horwitz, P. J. McGinn, M. A. Black, A. E. Klunzinger, and D. B. Pulling, Phys. Rev. B54, 6122 (1996). [ Links ]
 D. R. Harshman, G. Aeppli, E. J. Ansaldo, B. Batlogg, J. H. Brewer, J. F. Carolan, R. J. Cava, M. Celio, A. C. D. Chaklader, W. N. Hardy, S. R. Kreitzman, G. M. Luke, D. R. Noakes, and M. Senba, Phys. Rev. B 36, 2386 (1987). [ Links ]
 B. Pümpin, H. Keller, W. Kündig, W. Odermatt, I. M. Savic, J. W. Schneider, H. Simmler, P. Zimmermann, J. G. Bednorz, Y. Maeno, K. A. Müller, C. Rossel, E. Kaldis, S. Rusiecki, W. Assmus, and J. Kowalewski, Physica C162-164, 151 (1989). [ Links ]
 B. Pümpin, H. Keller, W. Kündig, W. Odermatt, I. M. Savic, J. W. Schneider, H. Simmler, P. Zimmermann, E. Kaldis, S. Rusiecki, Y. Maeno, and C. Rossel, Phys. Rev. B 42, 8019 (1990). [ Links ]
 A. G. Sun, D. A. Cajewski, M. B. Maple, and R. C. Dynes, Phys. Rev. Lett. 72, 2267 (1994). [ Links ]
 J. E. Sonier, J. H. Brewer, and R. F. Kiefl, Rev. Mod. Phys. 72, 769 (2000). [ Links ]
 D. R. Harshman, L. F. Schneemeyer, J. V. Waszczak, G. Aeppli, R. J. Cava, B. Batlogg, L. W. Rupp, E. J. Ansaldo, and D. Ll. Williams, Phys. Rev. B 39, 851 (1989). [ Links ]
 H. A. Blackstead and J. D. Dow, Solid State Commun. 95, 613 (1995). [ Links ]
 J. D. Dow and D. R. Harshman, J. Phys. Chem. Solids 63, 2309 (2002). [ Links ]
 J. D. Dow and D. R. Harshman, Phil. Mag. B82, 1055 (2002). [ Links ]
 For a variety of opposing views, most relying on surface-sensitive experiments, see D. A. Wollman, D. J. Van Harlingen, J. Giapintzakis, and D. M. Ginsberg, Phys. Rev. Lett. 74, 797 (1995); [ Links ]A. Lanzara, P. V. Bogdanov, X. J. Zhou, S. A. Kellar, D. L. Feng, E. D. Lu, T. Yoshida, H. Eisaki, A. Fujimori, K. Kishio, J.-I. Shimoyama, T. Noda, S. Uchida, Z. Hussain, and Z. X. Shen, Nature 412, 510 (2001); [ Links ]and J. R. Kirtley, C. C. Tsuei, K. A. Moler, J. Z. Sun, A. Gupta, Z. F. Ren, J. H. Wang, Z. Z. Li, H. Raffy, J. Mannhart, H. Hilgenkamp, B. Mayer, and Ch. Gerber, Czech. J. Phys. 46, Suppl. S6, 3169 (1996), [ Links ]and subsequent work.
 D. R. Harshman, W. J. Kossler, A. J. Greer, D. R. Noakes, C. E. Stronach, E. Koster, M. K. Wu, F. Z. Chien, J. P. Franck, I. Isaac, and J. D. Dow, Phys. Rev. B67, 054509 (2003). [ Links ]
 D. R. Harshman, J. D. Dow, W. J. Kossler, D. R. Noakes, C. E. Stronach, A. J. Greer, E. Koster, Z. F. Ren, and D. Z. Wang, Muon spin rotation study of mSR of GdSr2Cu2RuO8. Phil Mag., Submitted. [ Links ]
 D. R. Harshman, W. J. Koessler, A. J. Greer, C. E. Stronach, D. R. Noakes, E. Koster, M. K. Wu F. Z. Chien, H. A. Blackstead D. B. Pulling, and J. D. Dow, Physica C364-365, 392 (2001). [ Links ]
 J. D. Dow, H. A. Blackstead, Z. F. Ren, and D. Z. Wang, Magnetic resonance of Cu and of Gd in insulating GdSr2Cu2NbO8 and in superconducting GdSr2Cu2RuO8. Submitted. [ Links ]
 A. Fainstein, E. Winkler, A. Butera, and J. Tallon, Phys. Rev. B 60, 12597 (1999). [ Links ]
 H. A. Blackstead, J. D. Dow, et al., Magnetically ordered copper in superconducting rutheno-cuprates: evidence against spin-fluctuation models of high-temperature superconductivity," in preparation. [ Links ]
 J. D. Dow and M. Lehmann, Phil. Mag. 83, 527 (2003). [ Links ]
Received on 23 May, 2003
* Permanent address.