Pseudo-gap features of intrinsic tunneling in (HgBr_2)-Bi2212 single crystals
a r
X i
v
:c o
n
d
-
m
a t
/
9
9
7
1
5
9
v
1
[
c o
n
d
-
m a
t
.
s u
p
r
-c
o
n
]
1
J u
l
1
9
9
9
International Journal of Modern Physics B,f c World Scienti?c Publishing Company PSEUDO-GAP FEATURES OF INTRINSIC TUNNELING IN (HgBr 2)-Bi2212SINGLE CRYSTALS ?AUGUST YURGENS,?DAG WINKLER,and TORD CLAESON Department of Microelectronics and Nanoscience,Chalmers University of Technology and G¨o teborg University,S-41296,G¨o teborg,Sweden SEONG-JU HW ANG and JIN-HO CHOY Department of Chemistry,Seoul National University,Seoul 151-742,Korea Received (received date)Revised (revised date)The c-axis tunneling properties of both pristine Bi2212and its HgBr 2intercalate have been measured in the temperature range 4.2-250K.Lithographically patterned 7-10unit-cell heigh mesa structures on the surfaces of these single crystals were investigated.Clear SIS-like tunneling curves for current applied in the c -axis direction have been ob-served.The dynamic conductance d I/d V (V )shows both sharp peaks corresponding to a superconducting gap edge and a dip feature beyond the gap,followed by a wide max-imum,which persists up to a room temperature.Shape of the temperature dependence of the c -axis resistance does not change after the intercalation suggesting that a coupling between CuO 2-bilayers has little e?ect on the pseudogap.1.Introduction The existence of a pseudogap in electronic excitation spectra is believed to be one of the very important features of high-T c (HTS)superconductors.1This gap was reported to exist in underdoped samples,although some experiments indicate that the pseudogap is present in overdoped samples also.2,3The tunneling spectroscopy is particular sensitive to the density of states (DOS)at the Fermi level and therefore can be used to study any gap in the quasiparticle excitation spectrum.The most common experimental methods,STM and point-contact techniques are however essentially surface probes.The surfaces of most HTS compounds deteriorates with time unless special measures are undertaken.That is why there is a great variety of data quality reported in the literature.4It is now experimentally established that highly anisotropic layered HTS exhibit intrinsic tunneling e?ects.5For example,in Bi2212(Bi 2Sr 2CaCu 2O 8)the ≈3?A thick metallic copper oxide sheets (CuO-bilayers]are separated by ≈12?A thick insulating layers.Out-of-plane (or c -axis)charge transport occurs via a sequential
2Pseudo-gap features of intrinsic tunneling in(HgBr2)-Bi2212single crystals
tunneling of electrons or Cooper pairs between these sheets.5The atomic perfection of the naturally occurring tunnel junctions in such materials provides a reliable basis for the tunneling spectroscopy inside the single crystal giving a high degree of homogeneity and reproducibility.
The phonon-mediated sub-gap structures of such intrinsic Josephson junctions (IJJ)has already been studied.6The pseudogap spectroscopy with intrinsic tunnel junctions has also been attempted.7,8
Here,we repeat experiments of Refs.7,8on HgBr2-Bi2212samples.Insertion of inert HgBr2-molecules in between adjacent BiO-layers results in a signi?cant stretching of Bi2212crystals in the c-axis direction,see Fig.1without a?ecting the superconducting critical temperature T c much.The intercalation leads to decrease in the c-axis critical current I c and increase in the normal state resistance R c.The Joule heating,which was considered to be a problem in experiments on IJJ at high bias current,7,8can therefore be signi?cantly suppressed.Moreover,the e?ect of the interlayer coupling on the pseudogap can be investigated.
Fig.1.Schematic picture of HgBr2-intercalation.
2.Experiment
Bi2212single crystals were synthesized by the traveling-zone method.The interca-lation of HgBr2was perfomed by heating the vacuum sealed tube containing pristine single crystals and HgBr2during about one week.9According to X-ray analysis,the lattice expansion after the intercalation was～12.6?A.
To investigate the c-axis transport properties,several mesas with areas10?600μm2and heights150?200?A were fabricated on the freshly-cleaved surfaces of the single crystals using a standard photolithography and Ar-ion milling.10The resulting mesas comprised～5?15IJJ in series.
Pseudo-gap features of intrinsic tunneling in (HgBr 2)-Bi2212single crystals
3
01
23R (T ) / R (250K ) TEMPERATURE (K)
Fig.2.The temperature dependence of the c -axis resistance of the pristine (dash line)and inter-calated samples (circles and lines).
-1.5-1.0
-0.5
0.0
0.5
1.0
1.5
I (m V ) Fig.3.Current-voltage characteristics for (HgBr 2)Bi2212-sample.The contact resistance of 5?has not been subtracted.
3.Results and discussion
Fig.2shows the temperature dependence of R c for both the pristine (a dash line)and intercalated samples (symbols and lines).The intercalation lead to a signi?cant increase of the c -axis resistivity ρ.At T ≈175K it is about ～10?cm for the former,and ～2k?cm for the latter samples.The pristine samples were slightly overdoped after annealing in oxygen,which is seen both from the reduced value of T c ≈75K,and from the temperature dependence of the resistance showing the linearly increasing with temperature part at high temperatures,see Fig.2.It can be also
4Pseudo-gap features of intrinsic tunneling in (HgBr 2)-Bi2212single crystals
seen that the linear part of R c (T )disappeared after the intercalation in one sample,while conserved in another.After intercalation,T c somewhat further decreased,down to about 62-65K.T c of Bi2212seems to be less prone to HgBr 2
-molecules as compared with other intercalates,despite larger stretching of the lattice.9,11
Recently,it has been shown that the onset of the upturn in R c (T )correlates with appearance of the non-linearities in the c -axis tunneling characteristics with decreasing temperature.7These non-linearities bear witness of pseudogap features in the c -axis transport.From our experiments we see that the intercalation does not in?uence the characteristic temperature T ?in a straightforward way.Despite signi?cant increase,the c -axis resistivity can preserve the form of the temperature dependence.This implies that the out-of-plane resistance and pseudogap features are determined mainly by properties of individual CuO-bilayers in Bi-2212.The coupling between di?erent bilayers seems has a little e?ect on pseudogap,as long as oxygen content is preserved.12
The main frame of Fig.3shows a typical I ?V -characteristic of a mesa with 10IJJ.The inset shows an enlarged central part of the I ?V -characteristic.The majority of junctions have I c ～0.1mA,which is more than an order of magnitude smaller than I c of the pristine-crystal mesas with the same area A (～200μm).10The normal state tunneling parts of the I ?V -characteristics are clearly seen at ～1mA.The dynamic conductance d I/d V vs.the average voltage per one IJJ in the mesa was deduced by numerical di?erentiation of the I ?V curve corresponding to all 10IJJ being in the quasiparticle state,Fig.4.d I/d V of 0.01S at ～100mV in our experiments would correspond to 10?10S for typical STM-contact areas (say,15×15?A 2),emphasizing a high quality of the tunneling barriers in IJJ.
05
10
1520
25
V/10 (mV)
d I /d V (m S )Fig.4.d I /d V (V )at di?erent temperatures for (HgBr 2)Bi2212-sample.
Sharp peaks seen in Fig.4likely correspond to the superconducting energy
Pseudo-gap features of intrinsic tunneling in (HgBr 2)-Bi2212single crystals 5
gap V
g of CuO 2layers.The dips at ～1.5V g were attributed to strong-coupling e?ects.13A gently sloping dip at the zero voltage at T >T c likely represents the pseudogap in the DOS.It has the almost temperature independent energy scale of ～100mV per each IJJ,which is somewhat larger than in experiments of Suzuki et al.7It is gradually smearing with temperature,although being seen in I ?V ’s at temperatures up to about 250K for some samples.
00.2
0.4
0.6
0.811.2
?(Τ) /?(4.2)TEMPERATURE (K)
Fig.5.The normalized energy gap parameter ?vs.temperature for the pristine (circles and triangles),and (HgBr 2)Bi2212-(squares)samples.All lines are guides for the eye.There were two distinct gap features for the pristine sample,?1and ?2.The empty symbols correspond to the disappearance of the c -axis critical current.
The temperature dependence of the normalized energy gap parameter ?is shown in Fig.5.The BCS-like temperature dependence is shown for comparison.For the pristine sample there were two distinct features in d I/d V (V )-curves,which may re-?ect either the in-plane anisotropy of the superconducting energy gap parameter,14or priximity e?ects in the complex multilayer system of Bi2212.15
The superconducting energy gap cannot be traced close to T c ,the correspond-ing feature of the d I/d V (V )-curves vanishes earlier,although the superconducting critical current can be seen clearly from measured I ?V -characteristics in the whole temperature range up to T c ,see closed and open symbols in Fig.5.
4.Conclusions
In conclusion,we observed both superconducting and pseudogap features of low-transparency intrinsic tunneling in Bi-2212single crystals intercalated with HgBr 2.The dramatic decrease of the coupling between CuO 2-bilayers in the c -axis direction after the intercalation does not in?uence much the shape of R c (T )-curves.The characteristic temperature when the pseudogap features set in in the current-voltage
6Pseudo-gap features of intrinsic tunneling in(HgBr2)-Bi2212single crystals
characteristics with decreasing of temperature stays also nearly unchanged.All this implies that it is properties of an individual CuO2-bilayer(plus surrounding charge reservoirs)which determine both R c(T)and the pseudogap in the single particle spectrum of Bi2212.
References
1.W.W.Warren et al.,Phys.Rev.Lett.62,1193(1989);G.V.M.Williams et al.,Phys.
Rev.Lett.78,721(1997);R.Nemetschek et al.,Phys.Rev.Lett.78,4837(1997);J.W.
Loram et al.,Phys.Rev.Lett.71,1740(1993);H.Ding et al.,Nature(London)382, 51(1996);M.R.Norman et al.,Nature(London)392,157(1998).See also a review paper by T.Timusk and B.Statt,cond-mat/9905219.
2.H.J.Tao,F.Lu,and E.L.Wolf,Physica C282-287,1507(1997).
3.Ch.Renner et al.,Phys.Rev.Lett.80,149(1998).
4.T.Hasegawa,H.Ikuta,and K.Kitazawa,in Physical Properties of High Temperature
Superconductors III,ed.by D.M.Ginsberg(World Scienti?c,Singapore,1992)p.525.
5.R.Kleiner et al.,Phys.Rev.Lett.68,2394(1992);
6.Ch.Helm et al.,Phys.Rev.Lett.79,737(1997).
7.M.Suzuki,T.Watanabe,and A.Matsuda,Phys.Rev.Lett.82,5361(1999).
8.I.F.G.Parker et al.,in Conference on Superlattices II:Native and Arti?cial,ed.I.
Bozovic and D.Pavuna,(SPIE,Bellingham,WA,1998),SPIE Vol.3480,p.11.
9.J.-H.Choy,S.-J.Hwang,and N.-G.Park,J.Am.Chem.Soc.119,1624(1997).
10. A.Yurgens et al.,Appl.Phys.Lett.70,1760(1997).
11.X.-D.Xiang et al.,Nature348,145(1990).
12.In experiments with hydrostatic pressure we also observed that despite signi?cant
change of the c-axis resistance with pressure,its temperature dependence did not change much,see A.Yurgens et al.,Phys.Rev.Lett.82,3148(1999).
13.Y.DeWilde et al.,Phys.Rev.Lett.80,153(1998).
14.J.Ma et al.,Science267,862(1995).
15. A.Yurgens et al.,Phys.Rev.B53,R8887(1996);