Skip to main content

Full text of "USPTO Patents Application 08479810"

See other formats


PHYiSICAL REVIEWS 



VOLUME 38. NUMBER 10 



1 OCTOBER 1988 



Model famUy of high-temperature superconductors: 'O.Ca.-.BazCu.OiO. + D+m 

(m-1,2; «-l,2,3) 

«; S P Parkin V Y. Lec. A. I. Nazzal. R. Savoy. T. C. Huang, G. Gorman and R. Beyers 
S. S. P. P""""-;^ 5^;,^^^;^ jt^search Center. 650 Harry Road. 

San Jose. California 95120-6099 
(Received 31 May 1988) 

We describe the structures and superconducting p«.perties of six ^^"""^^^^J]^-^;^ 

the perovskitehke block for both the \ , ix^nMon tempera- 

compounds (m-1.2) with .fit It^rials with single 11-0 layers. 

-atrnstS: — rr^ri ^ 
serS to correlate with increased densities of intergrowths of related structures. 



Recently'-' several new high-temperature supercon- 
durtS have been synthesized in the Tl-Ca-Ba-Cu-O sys- 
?em. including ThCa^Ba^CujO.oi,. whtch d^aJ^ jhe 
highest superconducting transition temperature yet found. 
T -125 K ' In this arUcle we present data on the struc- 
tu'res and superconducting properties of s« compounds of 
the form Tl„Ca„-,Ba2Cu„02(„ + i)+-.. " ' 

and n-1 2, or 3. The structures consist of copper 
perovskitelike blocks containing 1. 2. or 3 CuOi planes 
«parated by one or two Tl-O layers. P^.~i;P°""5! 
thus form a model family of structures m which both the 
size and separation of the copper blocks can be indepen- 
dently variS^. We present daU that establish that the su- 
perconducting transition temperature i«creas« with tihe 
Tumber of CuO: planes in the perovskitehke blc«k for 
both the Tl-O monolayer and bilayer compounds. For 
each pair of compounds (m -1. 2) with the same number 
of CuOz planes (same n\ the transition tempera ure is 
15-20 K lower in the material with single Tl-O layers. 
Variations in the transition temperatures in the double 
and triple CuOz layer compounds are obsenrcd to corre- 
late with increased densities of intergrowths of related 

^^'Se^ples were prepared by thoroughly mixing suit- 
able amounts of Tl^O,. CaO. BaO,. a»^^«^0 "d form- 
ing a pellet of this mixture under pressure. The pellet was 
then wrapped in gold foU. sealed in a quarU tube contain- 
ing slightly less than 1 atm of oxygen, and baked for ap- 
proximately 3 h at =880-C. A wide range of startmg 
compositions was studied. In most cases the resultmg pel- 
let was comprised of several phases. However, for certain 
ranges of starting compositions, the pellets contained Mily 
one superconducting phase of the form Tl"Ca,^iB^- 
Cu,O20,+i)+m together with minor amounts ( < --20%) 
of insulating oxides such as those of Cu. Ca-Ci^ Ba-Cu, 
and Tl-Ba. The relative amounts of each phase depended 
on the annealing time and temperature and the rate of 



cooUng from this temperature. In particular, for slow 
cooUng rates (=lOO»C/h) the composition of the major 
Tl„Ca.-iBa2Cu.O20,+.)+« more closely matched 
that of the storting composiUon. The composition and nai- 
crostructure of the peUets were determined from comple- 
mentary powder x-ray diffraction, electron microprobe. 
electron diffraction, and high-resolution transmission elec- 
tron microscopy (TEM) studies. The superconductmg 
properties of each pellet were examined by resistivity and 
dc Meissner susceptibility studies. The latter was mea- 
sured with a SHE SQUID magnetometer. Coolmg in a 
field of 100 Oe, the magnitude of the Meissner susceptibil- 
Hy at 5.5 K ranged from 10% to 35% of the suscep ibibty 
of a perfect diamagnet of the same volume. neglecUng 
small demagnetizing corrections. The 
diamagnetic shielding signal is very dependent on the dis- 
tribution of the normal and superconducting phases wiUim 
the multiphase pelleU and in most <^ did "ot g|vejiw- 
ful information. The susceptibility dato revealed that for 
some pelleu Uie presence of a minority superconducting 
phase resulUrf in tiie resUtance of the pellet dropping to 
iero at substiuitiaUy higher temperaturra than the ol 
the majority supereonducting phase. IHis type of behav- 
ior emjasizes the importance of determining the transi- 
tion temperature from a flux exclusion »easurement m 
this comjlex quinary system. These results are summa- 

rized in Table I. . . 

We have previously described the preparation and prop- 
erties of the Uiree members of the 
0,r family, namely T!2CaiBa2Cu30io « 

TbQ.Ba"cu20, (2:1:2:2).' and Tl.Ca^Ba^Cu,©, 
(1 2l-3) * wWch display superconducting transiUM tem- 
on25. loS. andTlO K. -^^^1^. 

SSS^rsJnth^SiVu-Jce^^^^^^^ 

©1988 The American Physical Society 



38 6531 



BEST AVAiLAbLfc UUf'Y 



PHYSICAL REVIEW B 



VOLUME 38, NUMBER 10 



1 OCTOBER 1988 



Model family of high-temperature superconductors: 11m Ca.- iBaiCunOaCji + 1 ) +m 

(m-1.2;/i-l,2,3) 

S. S. p. Parkin, V. Y. Lee, A. I. Nazzal, R. Savoy, T. C. Huang, G. Gorman, and R. Beyers 
IBM Research Division, Almaden Research Center, 650 Harry Road, 
San Jose, California 95120-6099 
(Received 31 May 1988) 

We describe the structures and superconducting properties of six compounds in the TI-Ca-Ba- 
Cu-O system of the genera! form. TUiCa*-iBa2Cu„O20i+i)+«. where m-1 or 2 and w-l. 2, or 
3. One of the compounds displays the highest known superconducting transition temperature, 
Tc — \2S K. The structures of these compounds consist of copper perovskiteiike blocks containing 
I, 2, or 3 Cu02 planes separated by one or two Tl-O layers and thus form a model family of 
structures in which both the size and separation of the copper oxide blocks can be independently 
varied. The superconducting transition temperature increases with the number of Cu02 planes in 
the perovskiteiike block for both the Tl-O monolayer and bilayer compounds. For each pair of 
compounds (m — 1,2) with the same number of Cu02 planes (same n), the transition tempera- 
tures are similar but are consistently 15-20 K lower in the materials with single Tl-O layers. 
Variations in the transition temperatures in the double and triple Cu02-layer compounds are ob- 
served to correlate with increased densities of intergrowths of related structures. 



Recently*"' several new high-temperature supercon- 
ductors have been synthesized in the Tl-Ca-Ba-Cu-O sys- 
tem, including Tl2Ca2Ba2Cu30io±x. which displays the 
highest superconducting transition temperature yet found, 
— 125 K.' In this article we present data on the struc- 
tures and superconducting properties of six compounds of 
the form TlmCa„-|Ba2Cu„O20i+i)+m. where m — 1 or 2 
and n"l, 2, or 3. The structures consist of copper 
perovskiteiike blocks containing 1, 2, or 3 Cu02 planes 
separated by one or two Tl-O layers. These compounds 
thus form a model family of structures in which both the 
size and separation of the copper blocks can be indepen- 
dently varied. We present data that establish that the su- 
perconducting transition temperature increases with the 
number of Cu02 planes in the perovskiteiike block for 
both the Tl-O monolayer and bilayer compounds. For 
each pair of compounds (m 1, 2) with the same number 
of Cu02 planes (same /i), the transition temperature is 
15-20 K lower in the material with single Tl-O layers. 
Variations in the transition temperatures in the double 
and triple Cu02 layer compounds are observed to corre- 
late with increased densities of intergrowths of related 
structures. 

The samples were prepared by thoroughly mixing suit- 
able amounts of TI2O3, CaO, Ba02, and CuO, and form- 
ing a pellet of this mixture under pressure. The pellet was 
then wrapped in gold foil, sealed in a quartz tube contain- 
ing slightly less than 1 atm of oxygen, and baked for ap- 
proximately 3 h at =880**C. A wide range of starting 
compositions was studied. In most cases the resulting pel- 
let was comprised of several phases. However, for certain 
ranges of starting compositions, the pellets contained only 
one superconducting phase of the form TlmCaii-iBa2- 
Cu„02(rt + i)+/n together with minor amounts (< =20%) 
of insulating oxides such as those of Cu, Ca-Cu, Ba-Cu, 
and Tl-Ba. The relative amounts of each phase depended 
on the annealing time and temperature and the rate of 

38 



cooling from this temperature. In particular, for slow 
cooling rates (==100**C^) the composition of the major 
TlmCan-iBa2Cu„02(rt+i)+m phase more closely matched 
that of the starting composition. The composition and mi- 
crostructure of the pellets were determined from comple- 
mentary powder x-ray diffraction, electron microprobe, 
electron diffraction, and high-resolution transmission elec- 
tron microscopy (TEM) studies. The superconducting 
properties of each pellet were examined by resistivity and 
dc Meissner susceptibility studies. The latter was mea- 
sured with a SHE SQUID magnetometer. Cooling in a 
field of 100 Oe, the magnitude of the Meissner susceptibil- 
ity at 5.5 K ranged from 10% to 35% of the susceptibility 
of a perfect diamagnet of the same volume, neglecting 
small demagnetizing corrections. The magnitude of the 
diamagnetic shielding signal is very dependent on the dis- 
tribution of the normal and superconducting phases within 
the multiphase pellets and in most cases did not give use- 
ful information. The susceptibility data revealed that for 
some pellets the presence of a minority superconducting 
phase resulted in the resistance of the pellet dropping to 
zero at substantially higher temperatures than the Tc of 
the majority superconducting phase. This type of behav- 
ior emphasizes the importance of determining the transi- 
tion temperature from a flux exclusion measurement in 
this complex quinary system. These results are summa- 
rized in Table I. 

We have previously described the preparation and prop- 
erties of the three members of the Tlff,Can-|Ba2Cun- 
02(#i+i)+m family, namely Tl2Ca2Ba2Cu30io (2:2:2:3),' 
Tl2CaiBa2Cu208 (2:1:2:2),^ and Tl|Ca2Ba2Cu309 
(1:2:2:3),* which display superconducting transition tem- 
peratures of 125, 108, and 110 K, respectively. By sys- 
tematically varying the starting composition of the pellets, 
the related compounds, Tl2CaoBa2CuiOx (2:0:2:1), 
TliCaoBa2CuiO, (1:0:2:1), and TliCaiBa2Cu20x 
(1:1:2:2) were synthesized. The unit cells for each phase 

653 1 © 1988 The American Physical Society 



6532 



S. S. P. PARKIN e/fl/. 
TABLEI. Summary of properties of Tl«Ca,-iBa2Cu„0x 



38 



Cone, 
ratio 


Tl 


Relative composition 
Ca Ba Cu 


o 


Symmetry 


Lattice parameters 
a (A) c (A) 


Superlattice 
wave vector (k) 


Tc (K) 














TliCan-i 










1 •rh>')* t 
1 Sj.Z, 1 


1 1 
1 .X 


0.0 


2 


0.7 


4.8 


PAlntmm 


3.869(2) 


9.694(9) 


a 


b 


1:1:2:2 


1.1 


0.9 


2 


2.1 


7.1 


P4/mmm 


3.8505(7) 


12.728(2) 


<0.29,0,0.5) 


65-85 


1:2:2:3 


1.1 


0.8 


2 


3.0 


9.7 


PAlmmm 


3.8429(6) 


15.871(3) 


(0.29,0,0.5) 


100-110 














TliCai,-! 


Ba2Cu«Ox 








2:0:2:1 


1.9 


0.0 


2 


l.l 


6.4 


Flmmm^ 


-5.445(2) 


23.172(6) 


(0:68,0.24, 1)*^ 


b 














A -5.492(1) 








2K):2:l** 


1.8 


0 


2 




6.4 


Flmmm^ 


a -5.4634(3) 


23.161(1) 


(008,0.24, 1>'= 


20 


2:0:2:1 


1.8 


0.02 


2 


1.1 


6.3 


J4/mmm 


3.8587(4) 


23.152(2) 


<aT6,0,08.I>* 


15-20 


2:1:2:2 


1.7 


0.9 


2 


2.3 


8.1 


I4/mmm 


3.857(1) 


29.39(1) 


(0.17,0,1) 


95-108 


2:2:2:3 


1.6 


1.8 


2 


3.1 


10.1 


lAimmm 


3.822(4) 


36.26(3) 


(0.17.0.1) 


118-125 



imSuc w Sj^cUlHc samples with no superconducting transition observed in resistivity and magnetic susceptibility studies 

n^h^^jl^mct^of^^^^ is ortAorWicif the observed superlattice is ignored. Taking the superlatUce into account lowers 

the symmetry to monoclinic, 

**Samp!e prepared from a Cu-rich starting composition, Tl2Ba2Cu2. 

*The superstructure is identical to that for the orthorhombic 2:0:2:1 polymorph. 



were determined from powder x-ray diffraction patterns 
extending from 2^-3'* to 70** and verified by electron 
diffraction studies. These studies showed that all of the 
TUiCa«-iBa2Cu«02(fl+i)+m compounds have tetragonal 
cells at room temperature. The Tl|Cart-iBa2CuB02ii+3 
compounds contain Tl-O monolayers, resulting in prima- 
tive tetragonal cells, whereas the Tl2Ca„-iBa2Cun02ii+4 
compounds contain Tl-O bilayers, resulting in body- 
centered tetragonal cells. The lattice parameters and 
symmetries of the various structures are included in Table 
I. As discussed later, the 2:0:2:1 compound also has an 
orthorhombic polymorph. As shown in Fig. I, each oxide 
has a single peak in the low-angle portion 
(3* < 2d< 10**) of its x-ray diffraction pattern which re- 
sults from the large da ratio in each structure. These 
peaks, (001) for m-1 and (002) for m-2, serve as 
fingerprints with which each of the compounds within the 




4 6 8 10 4 6 8 
ZB (deg) ZB (deg) 

FIG. 1. Low-angle section of the powder x-ray diffraction 
patterns for the six phases TU.Ca«-iBa2Cu«02Ci+i)+m (m — 1, 
2;n-l,2, 3). 



TlmCa„-iBa2Cu«02(«+i)+« famUy can be uniquely 
identified. The peak systematically shifts to lower angles 
as rt increases within both the TliCa„- iBa2Cu„02rt+3 and 
Tl2Cart-|Ba2Cu«02fl+4 families, consistent with an ex- 
pansion of the unit cell along the c axis by the addition of 
extra Cu02 and Ca planes. The peaks are in all cases at 
lower angles in the Tl2Ca«-iBa2CUrt02i,+4 compounds 
compared to the corresponding TliCa«-iBa2Cu„02n+3 
compound, consistent with the increased number of Tl-O 
layers in the Tl2Ca«-iBa2C:u„02i,+4 compounds. The 
peaks are asynmietrically broadened to low angles be- 
cause of geometrical aberrations in the focusing condition 
resulting from the flat specimens used.^ The arrangement 
of the cations in the various compounds is shown in Fig. 2. 
The positions of the oxygen atoms are inferred by compar- 
ison wjth related structures in the La2-xSrxCu04, 
YBa2Cu30x, and Bi2Sr2CaiCu20x families.*"" The six 
structures are comprised of Cu pcrovskitelike blocks con- 
taining one, two, or three Cu02 planes sandwiched be- 
tween Tl-O monolayers (1:0:2:1, 1:1:2:2, and 1:2:2:3 com- 
pounds) or bilayers (2:0:2:1, 2:1:2:2, 2:2:2:3 compounds). 
The Ba cations are located in planes adjacent to the Tl-O 
unit and the Ca cations form planes within the interior of 
the pcrovskitelike unit. 

Since the preparation, structure, and properties of the 
double and triple Cu02 layer oxides appear to be much 
less complex than those of the single Cu02 layer oxides 
for both the monolayer and bilayer Tl-O compounds, we 
will discuss these groups of compounds separately. As de- 
scribed earlier, for each of the /i — 2 and n - 3 compounds 
a single tetragonal structure was found. An important 
structural feature of these compounds observed by TEM, 
scanning electron microscopy (SEM). and electron mi- 
croprobe studies are intergrowths of structures related to 
the primary phase by the addition or removal of Cu02 or 
Tl-O layers. For some samples SEM images showed con- 
trast striations =5-10 itm in width within individual 



38 



MODEL FAMILY OF HIGH TEMPERATURE . 



6533 



o Tl 




TliCOn-l Ba2CUn02n+3 




TbCOn-l Ba2CUn02n+4 



n=l n = 2 n = 3 

FIG. 2. Nominal structures of the six TlmCa«-iBa2Cu«- 
Oiu+D+M phases for n -1, 2 and m — I, 2, 3. 



grains which result from intcrgrowths of regions with 
different proportions of heavy atoms. TEM studies re- 
vealed the existence of intergrowths on much finer length 
scales, as demonstrated in Fig. 3 for a sample prepared 
from a starting composition of TlossCaiBaaCua. Figure 
3(a) shows a selected area diffraction pattern along b* 
which indicates that this grain contains both 1:1:2:2 and 
1:2:2:3 phases. Indeed Meissner data on this sample [in- 
cluded in Fig 4(d)] indicate two superconducting transi- 
tions with Tc^ 65 and =105 K, consistent with the pres- 
ence of extended regions of two distinct phases. Coin- 
cidently, the c lattice parameters of the 1:1:2:2 and 1:2:2:3 
phases are almost exactly in the ration of 4/5 so that every 
fifth 1:2:2:3 hOl spot coincides with every fourth 1:1:2:2 
hOl spot in Figure 3(a). High-resolution TEM micro- 
graphs in Figs. 3(b) and 3(c) show intergrowths of the 




FIG. 3. (a) lOlO] selected area diffraction (SAD) pattern 
and (b) corresponding image of crystallites containing regions of 
1:2:2:3 and 1:1:2:2. The arrows in (b) denote unit-cell thick in- 
tergrowths of 1:1:2:2 in 1:2:2:3. (c) High-resolution transmis- 
sion electron micrograph of one unit-cell thick 1:1:2:2 inter- 
growth in 1:2:2:3. 



6534 



S. S. P. PARKIN €t al 



38 




E 20 40 60 80 100 
X 20, ■ ■ — 



60 



80 



100 120 



1,0 



2223 



80 



100 



120 140 40 
Temperature (K) 




FIG. 4. Mcissncr susceptibility vs temperature for an applied 
field of too Oe for materials with starting cation composition, 
(a) TlaBaiCui (•), TljCaaosBajCui.oj (0, + ), and TliCao^s- 
BaiCuMs (▼); (b) TliCaiBazCus (•), TlTCaiBajCu: (o). and 
TltisCaiBaiCui (■); (c) TliCa2Ba2Cu3 (■). TliCawBaiCuj (o). 
and TliCa3BaiCu3 (•); (d) TlojsCaiBaiCui (o). TliCaiBaiCuj 
(#,+), and TlossCazBajCuj (▼). The phases present in the pel- 
let giving rise to the diamagnetic susceptibility are (a) 2:0:2:1 
and 2:1:2:2, (b) 2:1:2:2. (c) 2:2:2:3, and (d) 1:1:2:2 and 1:2:2:3. 



1:1:2:2 and 1:2:2:3 phases on length scales extending from 
= 1/im down to one unit cell. The intergrowths are ran- 
domly distributed along the stacldng axis. Isolated inter- 
growths comprising four CuOj planes were found in some 
samples (see Fig. 5) but no evidence was found for ex- 
tended intergrowths comprising greater than three Cu02 
layers in these or other samples especially prepared from 
Cu- and Ca-rich starting compositions. A second type of 
intergrowth was observed in samples of the 1:2:2:3 phase 
in which an extra Tl-O plane was occasionally inserted be- 
tween the Cu perovskitelike units, creating local regions of 
the 2:2:2:3 phase. Microprobe data show that the Tl con- 
tent is systematically high in the compounds containing 
single Tl-O layers and systematically low in those com- 




FIG. 5. High-resolution TEM image of an isolated four- 
CuOj-layer intergrowth. The markers denote the positions of 
the Cu columns. 



pounds with Tl-O bilayers (see Table I) thus suggesting 
that intergrowths of Tl-O monolayers in the Tl-O bilayers 
materials and Tl-O bilayers in the Tl-O monolayer com- 
pounds are a general feature of these materials. 

Meissner data (see Fig. 4) established that Tc can take 
a range of values for all of the double and triple CuOi lay- 
er compounds— rc=95-108 K for 2:1:2:2, ^-118-125 
K for 2:2:2:3, r,=65-85 K for 1:1:2:2, and 7^ = 100 
-110 K for 1:2:2:3. For a given compound, x-ray 
diffraction and microprobe studies did not detect any ob- 
vious difference between the samples with different transi- 
tion temperatures. TEM studies, however, showed a clear 
correlation between the density of intergrowths and . 
For the 2:1:2:2, 2:2:2:3, and 1:2:2:3 phases the material 
with no intergrowths displayed the highest transition tem- 
perature, whereas for the 1:1:2:2 compound the sample 
with the lowest density of intergrowths had the lowest T^. 
As the density of intergrowths increased we observed that 
Tc systematically decreases or increases, respectively. It 
is possible that the structural or electronic modifications 
caused by the intergrowths are directly responsible for the 
decreased transition temperatures. Alternatively the pres- 
ence of the intergrowths may simply reflect a means 
whereby the system accommodates, to some extent, off- 
stoichiometry in the cation sites which in turn may 
influence Tc- It is difi&cult to determine whether it is the 
local change in structure or composition which is responsi- 
ble for the decrease in Tc since these are concurrent 
changes. 

A second important structural feature found in all of 
the double and triple CuOa layer compounds is the pres- 
ence of weak superlattice reflections in the selected area 
electron diffraction patterns. These reflections are consid- 
erably weaker than those previously found in the 
Bi2Sr2CaiCu20x compound'^"'* and indicate different 
structural modulations than those in the Bi2Sr2CaiCu20x 
compound. The patterns can be described by a set of 
symmetry-related wave vectors, k. Each wave vector de- 
scribes a pair of reflections symmetrically disposed a re- 
ciprocal distance | k | along k on either side of each Bragg 
peak, which would be consistent with a sinusoidal modula- 
tion of the charge density along this direction. *^ The pos- 
sibility that each k corresponds to a different crystal vari- 
ant with lowered symmetry is unresolved. The Tl-O 
monolayer and bilayer families each display a distinctive 
pattern of superlattice reflections, shown schematically in 
Figs. 6(a) and 6(b). One example of electron diffraction 
patterns showing the superlattice reflections is given in 
Fig. 7 for the 1:1:2:2 phase. 

The structure and properties of the single Cu02 layer 
compounds are more sensitive to the preparation condi- 
tions than those of the double and triple Cu02 layer com- 
pounds. When prepared from a Tl2Ba2Cui starting com- 
position, the 2:0:2: L compound has a face-centered ortho- 
rhombic cell and is not superconducting. The material is 
heavily twinned with twin planes of {l 10} type in the or- 
thorhombic cell. This cell is related to the tetragonal cell 
by a rotation of —45** about the c axis with a and b in- 
creased in size by a factor of ^>/2. However when the 
2:0:2:1 compound is prepared from a Cu-rich starting 
composition, Tl2Ba2Cu2, the compound is supcrconduct- 



38 



MODEL FAMILY OF HIGH-TEMPERATURE 



6535 




(b) 




010 



(c) 




• 0 Q O 



[T30] 



FIG. 6. Schematic diagram of the arrangement of superlat- 
tice reflections about the fundamental reflections for (a) the 
1:1:2:2 and 1:2:2:3 phases, (b) the 2:1:2:2 and 2:2:2:3 phases, (c) 
the 2K):2:1 phase. The fundamental reflections are shown as 
solid circles, and those which are systematically absent are 
shown as dashed circles. The superstructure is shown by open 
circles and the corresponding wave vectors by bold arrows. 



ing at =20 K. While x-ray data indicate the structure is 
pseudotetragonal, transmission electron inicrographs re- 
veal a tweed pattern which is consistent with local ortho- 
rhombic distortion. A tetragonal polymorph with no evi- 
dence from TEM studies of either an average or local or- 
thorhombic distortion can be formed by preparing the 
compound from a pellet containing a small amount of Ca 
{Tl:Ca:Ba:Cu-2:y:2:l+>^, with 3?= 0.05-0. 15). This po- 
lymorph is also superconducting with a Tc which is in- 
dependent of the amount of Ca in the starting composition 
but weakly dependent on the annealing time— Tc =15 
and 20 K for anneal times at 880 **C of 3 and 9 h. respec- 
tively. As suggested by the Meissner data in Fig. 4(a) 
these pellets contain, in addition to the tetragonal 2:0:2:1 
phase, a substantial amount of the 2:1:2:2 phase which in- 
creases as the proportion of Ca in the starting composition 
is increased. There is a sufficient amount of this phase 
that the resistance of these pellets actually drops to zero at 



. 4 . . 



FIG. 7. (a) llOO) and (b) lOOl] selected area diffraction pat- 
terns from crystallites of 1:1:2:2 showing superlattice reflections. 

100 K. The Meissner data in Fig. 4(a) show that for 
I' =0.05 the ratio of 2:1:2:2 to 2.0:2:1 is about 8% and for 
y=0.15 the ratio is increased to =30%. Electron mi- 
croprobe analysis shows that only a small amount of Ca 
(=0.2 at.%) is incorporated into the 2:0:2:1 grains and 
consequently the role of the Ca doping in changing the 
structure and properties of the 2:0:2:1 material is unclear. 
Moreover there are reports that the 2:0:2:1 phase can be 
prepared without Ca with a transition temperature as high 
as =85 K.^ Both polymorphs of the 2:0:2:1 structures 
display a si milar superlattice with an approximate wave 
vector, k- [0.08,0.24,1] in the orthorhombic setting. 
Taking the superlattice into account lowers the symmetry 
of both the orthorhombic and tetragonal structures to 
monoclinic with the c axis being unique. As shown in Fig. 
8 this superstructure is different from those found in the 
double and triple CuOj layer compounds. 

The other member of the TlmCa„-iBa2Cu„O20i + i)+m 
family which contains single CuOi layers, the 1:0:2:1 
phase, has a primitive tetragonal cell and is not supercon- 
ducting for the wide range of preparative conditions con- 
sidered in this study, including growth from Cu-rich or 
Ca-doped starting compositions. No superstructures have 
been observed in these crystals so far. 



6536 S. S. P. PARKIN et al. 



38 




(b) 




(C) 



• # • O' 



T T 



# f 

G : 



i 

— I 1 

0 12 3 4 

Number of CU-O2 Layers 

FIG. 9. Dependence of Tc on the number of CuOj planes 
within the Cu perovskitelike unit for the 'niCa„-iBa2Cu,02i,+j 
(■) and Tl2Ca„-iBa2Cu,02i.+4 this work; O, Ref. 5) scries of 
compounds. The dashed vertical lines correspond to the varia- 
tions in Tc found for each phase. □ corresponds to data for (H, 
Bi)i(Ca,Sr)2CuiOx (Ref. 21). 



5 120 



t 80 



40 



FIG. 8. (a) [1001, (b) IllO]. and (c) [OOll selected area 
diffraction patterns from a crystallite of 2:0:2:1. 



As shown in Table I there is no obvious correlation of 
superlatticc structure with the superconducting properties 
of the T!„Ca„-|Ba2Cu„02(i,+o+m compounds. Note 
that in the closely related compound, Bi2SriCa2Cu20x, it 
has recently been proposed that the observed incommens- 
urate superlatticc corresponds to a distortion of both the 
Bi-O and CUO2 planes resulting from ordered vacancies 



on the Sr sites. The vacancies are postulated to deter- 
mine the carrier density on the CUO2 planes and so 
influence the TV in a manner similar to that first noted by 
Schafer, Penney, and Olsen for the La2-xSrxCu04-^ 
compounds. The number of different superlatticc struc- 
tures found in the Tl-Ca-Ba-Cu-O system provides a more 
extensive basis with which to test such hypotheses. Indeed 
it may be significant that, as shown in Table I, there are 
important variations in stoichiometry away from the ideal 
stoichiometries expected for the various TlmCan-iBaa- 
Cu„O20i+i)+m phases. In particular, the [TU/lBal ratio 
is higher for the n*l compounds compared to those for 
rt — 2 and « -3. Band-structure calculations of both the 
TUCa„-iBa2Cu„O20i+i)+m compounds and Bi2Sr|Ca2- 
Cu20x indicate that the stoichiometry of the Tl-O and 
Bi-O layers would have a profound impact on the carrier 
density in these materials.*''^ The extent of off- 
stoichiometry on the cation or the oxygen sites in the Tl- 
Ca-Ba-Cu-O phases requires further study. Note also 
that one group has recently prepared a complex material 
of the form (Tl,Bi)i(Ca,Sr)2CuiOx with the 1:0:2:1 struc- 
ture which appears to superconduct at temperatures of up 
to 50 K (Ref. 21). The variation of properties of the sin- 
gle Cu02 layers compounds provides a fertile area for fur- 
ther study and highlights the difficulties in preparing these 
multicomponent oxides in a controlled manner. 

In conclusion, these studies have shown that the super- 
conducting transition temperature increases with the 
number of CUO2 planes in the perovskitelike unit for both 
the TliCa„-|Ba2Cu„02«+3 and Tl2Cai,-iBa2Cu„02fl+4 
structures (Fig. 9). A similar dependency is found in both 
series of compounds with an increased spread of Tc as the 
number of CUO2 planes is reduced. The range of Tc in the 
double and triple CUO2 layer compounds correlates with 
the density of intergrowth defects. No such defects have 
ben observed so far in the single CUO2 layer compounds, 



38 



MODEL FAMILY OF HIGH-TEMPERATURE . 



6537 



even when doped with Ca. One might speculate that in 
this case the variation in transition temperature may re- 
sult from variations in cation or oxygen site occupancy. 
The increase in Tf as n increases may be accounted for by 
various theories, including several based on the BCS 
theory and others invoking more exotic mechanisms 
such as the resonating-valencc-bond model The variety 
of structures and properties in the Tl-Ca-Ba-Cu-O system 
provides a model family of compounds with which various 



theories of high-temperature superconductivity can be 
evaluated. 

We are indebted to S. J. La Placa. F. Herman, and J. B. 
Torrance for many useful discussions. We thank C. C. 
Torardi. R. B. Flippen, and R. M. Hazcn for discussions 
regarding the 2:0:2:1 compound. We are grateful to Pro- 
fessor Sinclair at Stanford for the use of his electron mi- 
croscope. 



*Z. Z. Shcng and A. M, Hermann, Nature (London) 332, 138 

(1988); 332, 55 (1988). 
2R. M. Hazen. L. W. Finger, R. J. Angel, C. T. Prewitt, N. L. 

Ross, C. G. Hadidiacos, P. J. Hcancy. D. R. Vcblcn, Z. Z. 

Sheng, A. El Ali, and A. M. Hermann, Phys. Rev. Lett. 60, 

1657 (1988). 

3$. S. P. Parkin, V. Y. Lcc, E. M. Englcr. A. I. Nazzal, T. C. 

Huang, G. Gorman, R. Savoy, and R. Beyers, Phys. Rev. Lett. 

60, 2539 (1988). 
^. S. P. Parkin, V. Y. Lec. A. L Nazzal. R. Savoy, R. Beyers. 

and S. J. La Placa, Phys. Rev. Lett 61, 750 (1988). 
5C. C. Torardi. M. A. Subramanian, J. C. Calabrese, J. 

Gopalakrishnan, K. J. Morrissey, T. R. Askew. R. B. Flippen. 

U. Chowdry, and A. W. Sleight. Science 240, 631 (1988). 
6M. a. Subramanian. J. C. Calabrese. C C. Torardi. J. 

Gopalakrishnan, T. R. Askew, R. B. Hippen. K. J. Morrissey, 

U. Chowdry. and A. W. Sleight, Nature (London) 332. 420 

(1988). 

^S. Kondoh, Y. Ando. M. Onoda, and M. Sato. Solid State 

Commun. 65, 1329(1988). 
8 W. Parrish, X-ray and Electron Methods of Analysis (Plenum. 

New York, 1968). 
9j. M. Tarascon. Y. U Page. P. Barboux, B. G. Baglcy, L. H. 

Greene, W. R. McKinnon. G. W. Hull, M. Giroud, and D. M. 

Hwang. Phys. Rev. B 37. 9382 (1988). 
>0S. A. Sunshine, T. Siegrist. L. F. Schneemcyer. D. W. Mur- 
phy, R. J. Cava. B. Batlogg, R. B. van Dover. R. M. Fleming. 

S. H. Glarum, S. Nakahara, R. Farrow, J. J. Krajcwski, S. M. 

Zahurak. J. V. Waszczak, J. H. Marshall. P. Marsh, L. W. 



Rupp, and W. F. Peck, Phys. Rev. B 38. 893 (1988). 
»>J. B. Torrance, Y. Tokura, S. J. LaPlaca, T. C. Huang. R. J. 
Savoy, and A. L Nazzal, Solid State Commun. 66. 703 
(1988). 

>2t. M. Shaw. S. A. Shivanshankar. S. J. LaPlaca, J. J. Cuomo. 

T. R. McGuire. R. A. Roy, K. H. Kellehcr, and D. S. Yec. 

Phys. Rev. B 37, 9856 (1988). 
*3d. R. Veblcn. P. J. Heancy, R. J. Angel, L. W. Finger. R. M. 

Hazen, C. T. Prewiit. N. L. Ross. C. W. Chu, P. H. Hor, and 

R. L. Meng, Nature (London) 332, 334 (1988). 
»^E. A. Hewat. M. Dupuy, P. Bordet, J. J. Capponi. C. Chail- 

lout, J. L, Hodeau, and M. Marczio. Nature (London) 333. 

53 (1988). 

>5h. W. Zandbergcn. Y. K. Huang, M. J. V. Menken, J. N. Li. 

K. Kadowaki, A. A. Mcnovsky, G. van Tendcloo, and S. 

Amelinckx. Nature (London) 332, 620 (1988). 
**P. L. Gai and P. Day (unpublished). 
l^P. M. De Wolff. Acta Crystallogr. Sec. A 30, 777 (1974). 
»»M. W. Shafer, T. Penney, and B. L. Olsen. Phys. Rev. B 36, 

4047 (1987). 

>9M. S. Hybcrtson and L. F. Mattheiss. Phys. Rev. Lett. 60, 
1661 (1988). 

20F. Herman. R. V. Kasowski. and W. Y. Hsu. Phys. Rev. B 38, 
204 (1988); F. Herman, R. V. Kasowski. S, J. La Placa, and 
S. S. P. Parkin (unpublished). 

2»P. Haldar, A. Roig-Janicki, S. Sridhar, and B. C. Gtessen (un- 
published). 

"J. M. Wheatley. T. C. Hsu, and P. W. Anderson. Nature 
(London) 333, 121 (1988). 




Tl2Ca„H Ba2 CUn02n*4 



n=1 n~2 n=3 

FIG; 1 Nominal structures of the six 7l,iCa«i-iBa3Cu4t 
OiOi4-i)4» i^ases for i}* !, 2aiid iit <-K2« 3. 



Fia 3. (a) lOlOl selected area diffraction (SAD) pattern 
and (b) conespwiding image of crystallites containing regions of 
1:2:2:3 and 1:1:2:2. The arrows in (b) denote unit-cell thick in- 
tergrowths of 1:1:2:2 in 1:2:2:3, (c) High-resolution transmis- 
sion electron micrograph of one unit-cell thick 1:1:2:2 inter- 
growth in 1:2:2:3. 



FI6. 5. High*^rdK>liilkm TEM image id dn isdated fc^- 
CuOa-laycF intergtowtis. Tlit marlcers detiote the positmos 6f 
the Cu ootumns* 




FIG. 7. (a) IlOOl and (b) [00 U selected area diffraction pat- 
terns from crystalfites <tf 1 : 1 :2;2 showing superlatticc refiec^ons. 




(b) 




(C) 



.020^220, 

• m- ^ m • 



FIG. 8. (a) [1001. (b) [IIOJ. and (c) JOOlJ selected 
diffraction patterns from a crystallite ot 2^.2: 1 . 



This Page is Inserted by IFW Indexing and Scanning 
Operations and is not part of the Official Record 

BEST AVAILABLE IMAGES 

Defective images within this dGcument are accurate representations of the original 
documents submitted by the applicant. 

Defects in the images include but are not limited to the items checked: 

□ BLACK BORDERS 

□ IMAGE CUT OFF AT TOP, BOTTOM OR SIDES 

□ FADED TEXT OR DRAWING 
□ffiLURRED OR ILLEGIBLE TEXT OR DRAWING 

□ SKEWED/SLANTED IMAGES 

/ 

El COLOR OR BLACK AND WHITE PHOTOGRAPHS 

□ GRAY SCALE DOCUMENTS 

□ LINES OR MARKS ON ORIGINAL DOCUMENT 

□ REFERENCE(S) OR EXHIBIT(S) SUBMITTED ARE POOR QUALITY 

□ OTHER: 



IMAGES ARE BEST AVAILABLE COPY. 
As rescanning these documents will not correct the image 
problems checked, please do not report these problems to 
the IFW Image Problem Mailbox.