ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ

We have measured the temperature dependence of gap features revealed by Andreev reflection (Ds) and by tunneling (D) in the ab-plane of optimally and slightly overdoped microcrystals of (BiPb)₂Sr₂Ca₂Cu₃O₁₀₊δ (Bi2223) with critical temperature Tc = 110-115 K, and Bi₂Sr₂CaCu₂O₈₊δ (Bi2212) with Tc = 80...

Full description

Saved in:
Bibliographic Details
Published in:Физика низких температур
Date:2003
Main Authors: D`yachenko, A.I., Tarenkov, V.Yu., Szymczak, R., Szymczak, H., Abal`oshev, A.V., Lewandowski, S.J., Leonyuk, L.
Format: Article
Language:English
Published: Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України 2003
Subjects:
Online Access:https://nasplib.isofts.kiev.ua/handle/123456789/128786
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Cite this:ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ / A.I. D`yachenko, V.Yu. Tarenkov, R. Szymczak, H. Szymczak, A.V. Abal`oshev, S.J. Lewandowski, L. Leonyuk // Физика низких температур. — 2003. — Т. 29, № 2. — С. 149-155. — Бібліогр.: 27. назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id nasplib_isofts_kiev_ua-123456789-128786
record_format dspace
spelling D`yachenko, A.I.
Tarenkov, V.Yu.
Szymczak, R.
Szymczak, H.
Abal`oshev, A.V.
Lewandowski, S.J.
Leonyuk, L.
2018-01-13T20:04:57Z
2018-01-13T20:04:57Z
2003
ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ / A.I. D`yachenko, V.Yu. Tarenkov, R. Szymczak, H. Szymczak, A.V. Abal`oshev, S.J. Lewandowski, L. Leonyuk // Физика низких температур. — 2003. — Т. 29, № 2. — С. 149-155. — Бібліогр.: 27. назв. — англ.
0132-6414
PACS: 74.25.Jb, 74.50.+r, 74.72.-h
https://nasplib.isofts.kiev.ua/handle/123456789/128786
We have measured the temperature dependence of gap features revealed by Andreev reflection (Ds) and by tunneling (D) in the ab-plane of optimally and slightly overdoped microcrystals of (BiPb)₂Sr₂Ca₂Cu₃O₁₀₊δ (Bi2223) with critical temperature Tc = 110-115 K, and Bi₂Sr₂CaCu₂O₈₊δ (Bi2212) with Tc = 80-84 K. The tunneling conductance of a Bi2223-insulator-Bi2223 junction shows peaks at the 2D gap voltage, as well as dips and broad humps at other voltages. In Bi2223, similarly to the well-known Bi2212 spectra, the energies corresponding to 2D, to the dip, and to the hump structure are in the ratio 2:3:4. This confirms that the dip and hump features are generic to the high-temperature superconductors, irrespective of the number of CuO₂ layers or the BiO superstructure. On the other hand, in both compounds the D(T) and Ds(T) dependences are completely different, and we conclude that the two entities are of different natures.
This work was supported by Polish Government (KBN) Grant No PBZ-KBN-013/T08/19.
en
Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
Физика низких температур
Свеpхпpоводимость, в том числе высокотемпеpатуpная
ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
Article
published earlier
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
title ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
spellingShingle ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
D`yachenko, A.I.
Tarenkov, V.Yu.
Szymczak, R.
Szymczak, H.
Abal`oshev, A.V.
Lewandowski, S.J.
Leonyuk, L.
Свеpхпpоводимость, в том числе высокотемпеpатуpная
title_short ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
title_full ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
title_fullStr ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
title_full_unstemmed ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ
title_sort ab-plane tunneling and andreev spectroscopy of superconducting gap and pseudogap in (bi,pb)₂sr₂ca₂cu₃o₁₀₊β and bi₂sr₂cacu₂o₈₊δ
author D`yachenko, A.I.
Tarenkov, V.Yu.
Szymczak, R.
Szymczak, H.
Abal`oshev, A.V.
Lewandowski, S.J.
Leonyuk, L.
author_facet D`yachenko, A.I.
Tarenkov, V.Yu.
Szymczak, R.
Szymczak, H.
Abal`oshev, A.V.
Lewandowski, S.J.
Leonyuk, L.
topic Свеpхпpоводимость, в том числе высокотемпеpатуpная
topic_facet Свеpхпpоводимость, в том числе высокотемпеpатуpная
publishDate 2003
language English
container_title Физика низких температур
publisher Фізико-технічний інститут низьких температур ім. Б.І. Вєркіна НАН України
format Article
description We have measured the temperature dependence of gap features revealed by Andreev reflection (Ds) and by tunneling (D) in the ab-plane of optimally and slightly overdoped microcrystals of (BiPb)₂Sr₂Ca₂Cu₃O₁₀₊δ (Bi2223) with critical temperature Tc = 110-115 K, and Bi₂Sr₂CaCu₂O₈₊δ (Bi2212) with Tc = 80-84 K. The tunneling conductance of a Bi2223-insulator-Bi2223 junction shows peaks at the 2D gap voltage, as well as dips and broad humps at other voltages. In Bi2223, similarly to the well-known Bi2212 spectra, the energies corresponding to 2D, to the dip, and to the hump structure are in the ratio 2:3:4. This confirms that the dip and hump features are generic to the high-temperature superconductors, irrespective of the number of CuO₂ layers or the BiO superstructure. On the other hand, in both compounds the D(T) and Ds(T) dependences are completely different, and we conclude that the two entities are of different natures.
issn 0132-6414
url https://nasplib.isofts.kiev.ua/handle/123456789/128786
citation_txt ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)₂Sr₂Ca₂Cu₃O₁₀₊β and Bi₂Sr₂CaCu₂O₈₊δ / A.I. D`yachenko, V.Yu. Tarenkov, R. Szymczak, H. Szymczak, A.V. Abal`oshev, S.J. Lewandowski, L. Leonyuk // Физика низких температур. — 2003. — Т. 29, № 2. — С. 149-155. — Бібліогр.: 27. назв. — англ.
work_keys_str_mv AT dyachenkoai abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT tarenkovvyu abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT szymczakr abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT szymczakh abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT abaloshevav abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT lewandowskisj abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
AT leonyukl abplanetunnelingandandreevspectroscopyofsuperconductinggapandpseudogapinbipb2sr2ca2cu3o10βandbi2sr2cacu2o8δ
first_indexed 2025-11-26T00:18:49Z
last_indexed 2025-11-26T00:18:49Z
_version_ 1850600107646386176
fulltext Fizika Nizkikh Temperatur, 2003, v. 29, No. 2, p. 149–155 ab-plane tunneling and Andreev spectroscopy of superconducting gap and pseudogap in (Bi,Pb)2Sr2Ca2Cu3O10+� and Bi2Sr2CaCu2O8+� A.I. D’yachenko1, V.Yu. Tarenkov1, R. Szymczak2, H. Szymczak2, A.V. Abal’oshev2, S.J. Lewandowski2, and L. Leonyuk 3 1A. Galkin Donetsk Physical and Technical Institute of the National Academy of Sciences of Ukraine 72 R. Luxemburg Str., Donetsk 83114, Ukraine 2Institute of Physics, Polish Academy of Sciences, 32/46 Al. Lotnik�w, 02-668 Warsaw, Poland E-mail: abala@ifpan.edu.pl 3Moscow State University, Moscow, 118899 Russia Received May 7, 2002, revised July 29, 2002 We have measured the temperature dependence of gap features revealed by Andreev reflection (�s) and by tunneling (�) in the ab-plane of optimally and slightly overdoped microcrystals of (BiPb)2Sr2Ca2Cu3O10+� (Bi2223) with critical temperature Tc = 110–115 K, and Bi2Sr2CaCu2O8+� (Bi2212) with Tc = 80–84 K. The tunneling conductance of a Bi2223—insulator—Bi2223 junction shows peaks at the 2� gap voltage, as well as dips and broad humps at other voltages. In Bi2223, similarly to the well-known Bi2212 spectra, the energies corresponding to 2�, to the dip, and to the hump structure are in the ratio 2:3:4. This confirms that the dip and hump features are generic to the high-temperature superconductors, irrespective of the number of CuO2 layers or the BiO su- perstructure. On the other hand, in both compounds the �(T) and �s(T) dependences are com- pletely different, and we conclude that the two entities are of different natures. PACS: 74.25.Jb, 74.50.+r, 74.72.–h 1. Introduction Along with the usual coherence gap �s, in the spec- trum of quasiparticle excitations in high-Tc supercon- ductors there appears a gap �p (pseudogap), which persists above the superconducting transition tempera- ture Tc [1,2]. The pseudogap has the same d symmetry as �s, but disappears (more accurately: becomes indis- tinct) at some temperature T* > Tc [3]. The relation- ship between the pseudogap and superconductivity is far from clear [1,2]. One of the reasons appears to be that the most popular methods of investigating the ex- citation spectrum in cuprates, like tunneling and an- gle-resolved photoemission (ARPES), cannot distin- guish between �p and �s without recourse to various theoretical models. However, it is known that in the process of Andreev reflection of an electron from a normal metal–superconductor (N–S) interface, a Coo- per pair is created in the superconductor [4]. This oc- curs only in the presence of a nonzero energy gap �s in the superconductor. In other words, the process of Andreev transformation of an electron–hole pair into a Cooper pair is possible only for a reflection from the superconducting order parameter �s. In marked con- trast, the tunneling effect is sensitive to any singu- larity in the quasiparticle excitation spectrum [5]. Therefore, the tunneling characteristics at T < Tc in general depend on joint contributions of the energy gap and pseudogap. d-wave symmetry of the energy gap introduces some additional complications. The dominant con- tribution to the junction conductivity in classical (Giaever) tunneling comes from electrons with wave vectors forming a narrow, only a few degrees wide, cone [5]. Accordingly, tunnel junctions yield informa- tion on the gap anisotropy �( )k , and the gap revealed in tunneling experiments can be expressed as � � �( ) [ ) )]k k k� �s p /2 2 1 2( ( [6]. In the case of And- reev reflection from a clean N–S interface, the situa- © A.I. D’yachenko, V.Yu. Tarenkov, R. Szymczak, H. Szymczak, A.V. Abal’oshev, S. J. Lewandowski and L. Leonyuk, 2003 tion is different. The incident electron is not scat- tered, but reflected back along the same trajectory. This is true for any angle of incidence. It can be said that all incident electrons participate in Andreev re- flection on an equal footing. Therefore, measurement of a single Andreev N–S junction is in principle suffi- cient to determine the maximal value of the supercon- ducting gap �s(k). In this paper we employ the above discussed cha- racteristic features of tunneling and Andreev spectro- scopy to investigate the temperature dependence of the energy gaps � and �s in Bi2Sr2CaCu2O8+� (Bi2212) and (Bi,Pb)2Sr2Ca2Cu3O10+� [(BiPb)2223] cuprates. The well-studied Bi2212 has two CuO2 lay- ers per unit cell and strong incommensurate modu- lation in the BiO layer [7], which complicates the interpretation of tunneling and ARPES data. The sub- stitution of Bi by Pb in the (BiPb)2223 compound completely erases the superstructure in the BiO layers. Tunneling measurements were carried out on «break junctions» with the barrier surface practically normal to the crystallographic axes in the base ab plane of the material. In the c direction, the influence of the BiO layer on the tunneling spectra is much more pronounced. Andreev experiments were per- formed on S–N–S junctions. In both cases we retained only the samples showing pure tunneling or Andreev characteristics. The temperature dependences of the energy gaps obtained in the two types of experiments are completely different and attest to fundamental dif- ferences between the «superconducting» gap �s and the a,b-axis quasiparticle gap �. 2. Sample preparation The tunnel junctions were elaborated from Bi2223 and Bi2212 single crystals. Textured (Bi1.6Pb0.4)Sr2Ca2Cu3O10+� and Bi2Sr2CaCu2O8 sam- ples in the form of 10�1�0.1 mm rectangular bars were prepared [8–10] by compacting powdered (BiPb)2223 and Bi2212 compounds, respectively, at 30–40 kbar between two steel anvils. The powder was contained between two thin copper wires, whose deformation provided uniform pressure distribution in sample vo- lume. In this manner the powder was compacted into dense plane-parallel bars about 0.1 mm thick. The bars were then pre-annealed at T = 845�C for 16 h, compressed again, and finally annealed at T = 830�C for 14 h, producing a well-pronounced texture. Usual- ly the Bi2212 samples were slightly overdoped and ex- hibited a critical temperature Tc = 80–84 K. The dop- ing level of the oxygen content was controlled by annealing optimally doped samples in flowing gas ad- justed for different partial pressures of oxygen. The samples emerging from this procedure were highly textured, composed of tightly packed microcrystals aligned in one direction. Sample quality was controlled by transport measurements. We used for further pro- cessing only those samples which were showing a criti- cal current density Jc (T = 4.2 K) � 4�104 A/cm2. The superconducting transition temperature Tc was deter- mined from the midpoint of the resistive R(T) transi- tion (see Fig. 1). The S–I–S and S–N–S junctions were made by breaking specially prepared Bi2212 and Bi2223 sam- ples. Each sample was hermetically sealed by insula- ting resin and glued to an elastic steel plate, which was then bent until a crack occurred, running across the sample width and detected by monitoring the sam- ple resistance. The hermetic seal remained unbroken in this process. After the external load was relieved, the sample returned to its initial position with the crack closed and the microcrystals once again tightly pressed to each other along the line of the fracture. The best alignment is expected in the sample region in which the shear deformation was minimal. This is ap- parently one of the reasons why such a procedure re- sults in the realization of one effective junction of the microcrystal–microcrystal type. The selection of a sin- gle junction with minimal tunneling resistance from among the competing junctions is further assisted by the nature of the tunneling effect, which decreases ex- ponentially with the barrier thickness. A small sample thickness (< 100 �m) and a relatively large size of the microcrystals (> 10 �m) are also important factors en- hancing junction quality. Such break junctions on microcrystals were found to be particularly effective in the investigation of high-Tc superconductors [9]. The typical normal-state resistance of junctions used 150 Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 A.I. D’yachenko et al. � � Fig. 1. Temperature dependence of the ab-plane resistance of (BiPb)2223 samples with different oxygen doping. in the present study was between a few ohms and a few tens of ohms, and were remarkably stable. The surface of our Bi2212 and (BiPb)2223 break junctions was perpendicular to the CuO2 plane, and the direction of tunneling formed only a very small an- gle with one of the crystallographic axes (a or b) in this plane, as is attested to by the presence of Andreev bound states, seen in the tunneling S–I–S characteris- tics as a characteristic peak of conductivity at zero bias (cf. Fig. 2). Numerical calculations based on a simplified theoretical model [9,11] and taking into ac- count the d-wave mechanism of pairing show that the appearance of such a narrow zero-bias peak in the tun- neling conductance occurs at 6�. In high-quality break junctions the zero-bias conductance peak (ZBCP) was reported to coexist with the Josephson effect [12], but we have to rule out this possibility be- cause of the wrong signature: the ZBCP was insensi- tive to magnetic field and did not reflect on the I–V characteristics. The spectra �(V) = dI/dV show the quasiparticle peaks at 2�, where � is defined as a quar- ter of the peak-to-peak separation (Fig. 2). We use this � value as a measure of the gap, since there is no exact method of extracting the energy gap from the tunneling spectra, given that the exact functional form of the density of states for high-Tc superconduc- tors is not known. In general, the type of the junction was determined ex post facto from their conductance �(V) spectra. We retained for further investigation only the junc- tions conforming to either S–I–S or S–N–S types. For example, the �(V) curve in Fig. 2 reveals all the cha- racteristic features of a superconducting tunnel S–I–S junction: an almost flat region around zero bias fol- lowed by a sharp increase in the tunneling current, peaking around � 60 meV (2�); at still higher bias voltages V the conductance depends parabolically on V. The junction shown in Fig. 3, on the other hand, behaves as a typical Andreev S–N–S junction. First, there is a low-resistance region at low bias volt- ages, seen as a broad pedestal spanning the coordinate origin. The next indication is the excess current, which was observed in all S–N–S junctions included in this study. Finally, the differential conductivity of the junction at eV > 2�s coincides with the nor- mal-state conductivity at T > Tc [see inset in Fig. 3,b], i.e., for T > Tc practically all of the bias voltage is applied directly to the junction. ab-plane tunneling and Andreev spectroscopy Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 151 Fig. 2. Tunneling conductance of a Bi2223–I–Bi2223 junction at T = 77.4 K. The zero-bias peak is due to the Andreev bound state. The spectra clearly show dip and hump structures. Arrows indicate the 3� and 4� positions. C o n d u ct an ce , ar b . u n it s c � � (T ) / S m ax a b Fig. 3. Conductance � of S–N–S (Andreev) Bi2212–N–Bi2212 break junction. (a) Temperature depen- dence of �. The individual plots are shifted vertically for clarity. (b) Temperature dependence of the energy gap �s. The inset shows the � plots in their original position. One may inquire about the mechanism which might produce in an apparently random manner either S–I–S or S–N–S junctions. The insulating layer in S–I–S junctions is most probably caused by oxygen deple- tion. As to the normal barrier, we speculate that the CuO2 planes are harder to fracture than the buffer layers. After the sample is broken, they penetrate slightly into the buffer layers (see left inset in Fig. 4). In this manner, the coupling between the CuO2 planes belonging to the separated sample parts would be stronger than the normal coupling across the buffer layers, and it could assist in creating a constriction, which would act as a normal 3D metal. This hypothe- sis is in agreement with the scanning microscope study of the fracture surfaces of Bi2212 single-crystal break junctions, which revealed rough, but stratified frac- ture surfaces [13]. 3. Experimental results The temperature dependence of the energy gap �s(T) obtained from Andreev S–N–S measurements for Bi2212 exhibited a BCS-like form (see Fig. 3). We used two methods to determine �s for Andreev junc- tions. The first one is shown in Fig. 4 and relies on measuring the distance between the points of maximal slope changes of the �(V) plot, which is taken as the measure of 4�s. The details of the second one are shown in top inset in Fig. 4. The rationale for both methods is in recent calculations [14], based on the Klapwijk, Blonder, and Tinkham [15] treatment of multiple Andreev reflections between two supercon- ductors, which indicate that 2�s is determined by the separation of the extrema in d�/dV. The results ob- tained by both methods are plotted together in Fig. 3,b. It is seen that these results differ slightly, but both outline essentially the same �s(T) depen- dence. The �(T) gap dependence determined from tunnel- ing measurements performed on the same compounds diverged considerably from the BCS relation (Fig. 5). In fact, �(T) depends on temperature very weakly for T Tc. According to an ARPES investigation [16], such behavior of �(T) in Bi2212 near optimal doping is expected for the a (or b) direction in the CuO2 plane. This result agrees with our assumption about the direction of tunneling in our Bi2212–I–Bi2212 junctions. As mentioned above, further confirmation is provided by the presence of an Andreev bound state, 152 Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 A.I. D’yachenko et al. Fig. 4. Geometrical construction for the determination of �s from Andreev measurements. The top inset shows the corresponding d�/dV plot. The left inset shows the hypo- thetical inner structure of the Andreev break junction. Fig. 5. Conductance of S–I–S (tunneling) Bi2223–I–Bi2223 break junction. The inset shows the tem- perature dependence of the tunneling gap �(T). Some struc- tural details of the spectra have been blurred by the speed of recording needed to overcome temperature instabilities of the experimental setup. seen in the spectra of S–N–S and S–I–S junctions as a characteristic peak of conductivity at zero bias (cf. Fig. 3 and Fig. 5). According to the ARPES data [16], near optimal doping the �(T) gap becomes tem- perature dependent only when is of the order of 15�. For technological reasons, the formation of break junctions with the crystal broken at such an angle is improbable. As a result, the tunneling characteristics at T > Tc relate to the gap in (100) or (010) direction. In full agreement with the ARPES results [16], with increasing temperature the gap � of Bi2212 be- comes filled with quasiparticle excitations, and the conductance peaks at 2� become less distinct. The dis- tance between the still-discernible conductance peaks does not decrease, and the �(T) gap is seen to persist into the region T > Tc. Similar behavior is also ob- served for the (BiPb)2223 compound. The temperature dependence of the proper coherent gap �s(T) behaves in a completely different manner (Fig. 3). The gap narrows with increasing tempera- ture and at T = Tc it closes completely. The high cur- vature of the Andreev conductance dip at eV � 2�s is evidence both of the good quality of the investigated junctions and of the long lifetime of quasiparticles in the gap region. This was confirmed by the analysis of spectra of the normal metal–constriction–supercon- ductor (N–c–S) junctions [9]. For Bi2212 Andreev N–c–S junctions, the Blonder—Tinkham—Klapwijk [17] parameter Z used to obtain the theoretical fit was small, Z � 0.5, a value characteristic for very clean N–S contacts. For energies beyond the gap � value, tunneling in the ab plane of (BiPb)2223 S–I–S junction revealed the so-called dip and hump structures, as shown in Fig. 2 and Fig. 6. In Fig. 6, the voltage axis is norma- lized to the voltage eVp = �, and the conductance axis is normalized to the background; the spectra are shifted vertically for clarity. The dip and hump fea- tures roughly scale with the gap � for different oxygen doping levels (see Fig. 7). There is, however, a slight deviation of the data from a straight line. 4. Theoretical implications The considerable interest in pseudogap investiga- tion is stimulated to a great extent by the theoretical models of high-Tc superconductivity, in which a pseu- dogap appears as a precursor of the superconducting gap [18,19], e.g., the bipolaron model [20]. In ano- ther group of models, the appearance of pseudogap is related to some sort of magnetic pairing [21]. How- ever, the domains of applicability of these models are not very strictly defined and it is quite possible that the pseudogap (like high-temperature superconducti- vity) is caused by several simultaneously acting me- chanisms. For example, in the Emerson—Kilverson—Zachar (EKZ) theory [19] the crucial role in the formation of high temperature superconductivity is ascribed to the separation of spin and charge, arising as a result of partitioning of the CuO2 planes into narrow conduct- ing and dielectric stripes. «Pairing» at T* > Tc in the EKZ model means the formation of a spin gap. A wide ab-plane tunneling and Andreev spectroscopy Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 153 Fig. 6. S–I–S tunneling conductance in the ab plane for the Bi2223 samples of Fig. 1 at T = 77.4 K. The voltage axis has been rescaled in units of �. Each curve has been rescaled and shifted for clarity. � Fig. 7. � (dip) and Ep (hump) positions as a function of energy gap �, determined from the tunneling data of Figs. 2 and 6. spin gap (or pseudogap �p) is indeed formed in a spa- tially limited hole-free region, such as the region between the conducting stripes. A phase-coherent (i.e., actually superconducting) state is created only at T < Tc. The model explains well the smooth transi- tion of the pseudogap into the tunneling gap � when the temperature decreases below Tc. However, the ob- served temperature dependence of the order parameter gap �(T) at T < Tc is fundamentally different from that of the gap �s(T), as is seen in Figs. 3 and 5. It is not clear how the BCS-like �s(T) dependence arises in the phase-fluctuation picture. Such a situation would be possible, e.g., in the generation of charge (and spin) density waves, with the superconducting gap and pseudogap competing for the same region of the Brillouin zone [22]. Then the transition to the super- conducting state could occur in the presence of a pseudogap in normal excitations, opening, e.g,. in the electron–hole channel (i.e., a pseudogap, which would not transform directly into the superconducting gap, as in the Emery—Kivelson model). There are numerous experiments that confirm the essentially different nature of the superconducting gap �s and the gap (pseudogap) � [23–25]. The most convincing are intrinsic c-axis tunneling experiments (in stacked layers) [26]. However, they yield diffe- rent results from the point contact, scanning tunne- ling spectroscopy (STM), and break junction experi- ments: the hump was observed at an energy of 2� instead of 4�. The authors note a similarity between the observed c-axis pseudogap and Coulomb pseu- dogap for tunneling into a two-dimensional electron system. In our case, the tunneling and Andreev reflec- tion were realized in the ab plane, and together with the �(T) dependence (Fig. 4), we clearly observed the peak–dip–hump structure (Figs. 2 and 3). The posi- tion of the dip and hump for S–I–S junctions was at 3� and 4� (Fig. 2). This suggests that the observed dip–hump structure may originate from short-range magnetic correlations in the ab plane [27]. Then the gap � would be the fermionic excitation gap and �s — the mean-field order parameter. It should be empha- sized, finally, that the observed �s(T) dependence ex- hibits non-BCS behavior at T � 0 (Fig. 3). In summary, our ab-plane tunneling and Andreev spectroscopy studies of normal and slightly overdoped (BiPb)2223 and Bi2212 compounds show the presence of both a superconducting energy gap �s, corres- ponding to d-wave Cooper pairing, and a dip–hump structure at 3� and 4� (for the S–I–S junction). This suggests that the high-energy pseudogap, which is as- sociated with the dip and hump, could be magnetic in origin. The gap � is nearly temperature independent and becomes blurred above Tc, being continuously transformed with increasing temperature into the pseudogap. In contrast, the order parameter gap �s(T) has a strong temperature dependence and for T � 0 re- veals a non-BCS mean field behavior. Our findings are in general agreement with those of Deutscher [25], al- though it must be emphasized again that we have con- sidered the slightly overdoped case. Acknowledgments This work was supported by Polish Government (KBN) Grant No PBZ-KBN-013/T08/19. 1. T. Timusk and B. Statt, Rep. Prog. Phys. 62, 61 (1999). 2. T. Tohoyama and S. Maekawa, Supercond. Sci. Tech- nol. 13, R17 (2000). 3. Ch. Renner, B. Revaz, J.-Y. Genoud, K. Kadowaki, and �. Fischer, Phys. Rev. Lett. 80, 149 (1998). 4. A.F. Andreev, Sov. Phys. JETP 19, 1228 (1964). 5. E.L. Wolf, Principles of Electron Tunneling Spectro- scopy, Oxford University Press, New York (1985). 6. J.L. Tallon and G.V.M. Williams, Phys. Rev. Lett. 82, 3725 (1999). 7. M.A. Subramanian, C.C. Torardi, J.C. Calabrese, J. Gopalakrishnan, K.J. Morrissey, T.R. Askew, R.B. Flippen, U. Chowdhry, and A.W. Sleight, Science 239, 1015 (1988). 8. A.I. Akimenko, T. Kita, J. Yamasaki, and V.A. Gu- dimenko, J. Low. Temp. Phys. 107, 511 (1997). 9. A.I. D’yachenko, V.Yu. Tarenkov, R. Szymczak, A.V. Abal’oshev, I.S. Abal’osheva, S.J. Lewandowski, and L. Leonyuk, Phys. Rev. B61, 1500 (2000). 10. V.M. Svistunov, V.Yu. Tarenkov, A.I. Dyachenko, and R. Aoki, Physica C314, 205 (1999). 11. Y. Tanaka and S. Kashiwaya, Phys. Rev. Lett. 74, 3451 (1995). 12. A.M. Cucolo, A.I. Akimenko, F. Bobba, and F. Giubi- leo, Physica C341-348, 1589 (2000). 13. A.I. Akimenko, R. Aoki, H. Murakami, and V.A. Gu- dimenko, Physica C319, 59 (1999). 14. A.I. D’yachenko, private communication (2002). 15. T.M. Klapwijk, G.E. Blonder, and M. Tinkham, Phy- sica B109–110, 1657 (1982). 16. M.R. Norman, H. Ding, M. Randeria, J.C. Cam- puzano, T. Yokoya, T. Takeuchi, T. Takahashi, T. Mochiku, K. Kadowaki, P. Guptasarma, and D.G. Hinks, Nature 392, 157 (1998). 17. G.E. Blonder, M. Tinkham, and T.M. Klapwijk, Phys. Rev. B25, 4515 (1982). 18. V.J. Emery and S.A. Kivelson, Phys. Rev. Lett. 74, 3253 (1995); Nature 374, 434 (1995). 19. V.J. Emery, S.A. Kivelson, and O. Zachar, Phys. Rev. B56, 6120 (1997). 20. A.S. Alexandrov, Philos. Trans. R. Soc. London A356, 197 (1998) (and references therein). 154 Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 A.I. D’yachenko et al. 21. P.W. Anderson, The Theory of Superconductivity in the High-Tc Cuprate Superconductors, Princeton Uni- versity Press, Princeton (1997). 22. R.S. Markiewicz, C. Kusko, and V. Kidambi, Phys. Rev. B60, 627 (1999). 23. Q. Chen, K. Levin, and I. Kosztin, Phys. Rev. B63, 184519 (2001). 24. V.V. Kabanov, J. Demsar, B. Podobnik, and D. Mi- hailovic, Phys. Rev. B59, 1497 (1999). 25. G. Deutscher, Nature 397, 410 (1999). 26. V.M. Krasnov, A. Yurgens, D. Winkler, P. Delsing, and T. Claeson, Phys. Rev. Lett. 84, 5860 (2000). 27. N. Miyakawa, J.F. Zasadzinski, L. Ozyuzer, P. Gup- tasarma, D.G. Hinks, C. Kendziora, and K.E. Gray, Phys. Rev. Lett. 83, 1018 (1999). ab-plane tunneling and Andreev spectroscopy Fizika Nizkikh Temperatur, 2003, v. 29, No. 2 155