43
Thermochimica Acta, 133 (1988) 4348 Elsevier Science Publishers B.V., Amsterdam
THERMAL
ANALYSIS
IN THE FIELD OF HIGH TEMPERATURE
E. Kaisersberger,
J. Janoschek,
NETZSCH-Geratebau
GmbH, Wittelsbacherstrasse
SUPERCONDUCTORS
W. Hadrich 42, D-8672
Selb FRG
ABSTRACT
Thermoanalytical metry
methods
were used to study the preparation,
and phase transitions
A considerable
increase
ting oxide-carbonate
and loss between
0
balance.
in density
during
powder mixture
in the dilatometer.
The oxygen and 0
6
7
oxygen
stoichio-
of superconductors
of the type (Y, Eu)Ba Cu 0 . 2 37 the sintering process for the star-
was achieved
content
by rate controlled
and the reversibility
was investigated
with a symmetric
sintering
of oxygen
gain
thermo-micro-
INTRODUCTION
The ceramic prepared
technology
powders
used for manufacturing
allows
dilatometric,
sintesed
bodies
thermogravimetric
from specially
and DSC/DTA
measure-
ments. Mixtures
of oxides,
atmosphere l-2-3
carbonates
to follow
oxides.
The starting
from the fine grinded resulting
and oxalates
the reactions powders
oxides
superconductors
during
were heated
the preparation
were stoichiometrical
of Ytterbium
had critical
or Europium
temperatures
in air and in oxygen of superconducting mixtures
(by weight)
and CuO and BaCO
at 90 K to 95 K.
3
; the
EXPERIMENTAL
The sintering in a single
of powder
pushrod
dilatometer.
compressed
to small rods
and heated
in oxygen
shrinkage
by linear
The carbonate 6 mm, length
precursor
in the quasi isothermal
measurements
for Y6a2Cu3U7 150-200
was N/mm')
increase was programed
For rate controlled
at the desired
shrinkage
7-8mm, pressure
The temperature
up to 975'C.
raise was programmed
to hold the linear
was followed
(diameter
atmosphere.
at the rate of 10 K/min ature
compacts
sintering
or stepwise
isothermal
rate level.
Thermal Analysis Proc. 9th ETA congress, Jerusalem, Israel,21-25 Aug. 1988 0040-6031/88/$03.50 Q 1988 Elsevier Science Publishers B.V.
linear
the tempermode
44 Thermogravimetric microbalance.
tests were performed
The symmetrical
of very small mass changes or in a static or dynamic (stoichiometrical mosphere, hydrogen
in a newly developed
of this instrument
over a wide
were heated
product
Powders
thermo-
under a high vacuum
of Eu 0 , CUD and BaCD
in PtRh crucible:
was tempered
(97% He, 3% H2) atmosphere
symmetrical
allows determination
range of temperature
gas atmosphere.
mixture)
the resulting
design
in oxygen,
and afterwards
Sn dynamic reduced
again
oxygzn
at-
in a helium-
reoxidized.
RESULTSAND DlS&USSlON Dilatometry tering
is used for a long time in ceramic
temperatures,
changes.
Also
to a "ceramic
on the reactivity
the sintered
product
bonate
technology",
reactions
and structural
superconductors,
difatometry
of the raw materials
sin-
provides
which are preuseful
and on structural
fn-
changes
of
of sintering
curves
for powder
compacts
of the car-
precursor
heated
for YBa Cu 0 . The main sintering process starts at 78O'C 2 37 in oxygen at 10 K/min (curve 2), Minor shrinkage steps are recor-
ded at 34D'C and SlO'C.
The shrinkage
shown in fig. 1. Rate controlled age for the .same material and chemical
Fig.
mineral
to determine
fref 1).
Fig. 1 shows a comparison
when
in,firing,
in the field of high temperature
pared according formation
shrinkage
technology
1
stability
is not completed
sintering
(curve l), thus providing
of the sintered
Sintering
at the final temperature
leads to a 50% higher linear higher mechanical
shrink-
strength
body.
of YBa Cu 0 - carbonate precursors with linear heating 2 37 rate (curve 2) or stepwise isothermal mode (rate controlled sintering,
curve
1) In oxygen
atmosphere
45 The sintering w&ght
temperature
loss for the CO
780-C corresponds
evolution)
changes
2 in the expansion
tigated
up to now. The stepwise
with an intercept ent mechanisms isothermal
Preparation,
isothermal
further
are controlling
of the type EuBa2Cu307
mode
the sintering
reduction
resolves
tests are necessary
in thermogravimetry
tempering,
The smaller
curve at 340-C and 610'C have not been further
at 82O'C,
heating
to the main TG step (start of
for the same powder mixture.
and oxidation
evolved
differ-
(e.g. stepwise
gas analysis).
of superconducting
in subsequent
ranges
to show whether
in these ranges
and coupled
was investigated
two shrinkage
inves-
material
thermogravimetric
ex-
periments. Fig. 2 shows the temperature pering,
starting
program
with the carbonate
and TG-OTG
curves
for preparation
and tem-
precursor. 7
Fig. 2
Total
graphic
symmetrical
for preparation
thermobalance
and tempering
in flowing
of EuBa2Cu307
It can be seen.from
the one graph that the final weight
tempering
is higher
(+ 0.42%)
er weight
gain is achieved
ermal
segments
The preparation
of the selected
was evaluated
The Eu203-CuO-&CO3 low temperature, Therefore tions.
mixture most
this weight
compared
to level
in the linear tempering
separately
of the sample after
after preparation,
cooling
segments
and the high-
compared
to the isoth-a
program.
according
shows an unexpected
probable
in a
oxygen.
due to impurities
to fi'g. 3.
weight
loss of 1.52 % at
in the starting
step is not used for the further
powders.
stoichiometric
calcula-
46 The main weight
loss starting
at 760-C us evluated
for the formation
of
from the carbonate. The theoretical mass change of 10.88 X Cu 0 2 3 6.5 is exceeded in the experiment by impurities and inaccuracy in the stoichi-
EL&
ometry
of the starting mixture.
PREPARATION
Fig. 3
Preparation
(6 = x = 6.5) from the carbonate-oxide
of EuBa2Cu30x
mixture,
Above
109O'C a sharp mass change
is recorded
to 0 * This TG step is reversible tempgrature exact
range
for heating
(see Fig. 2). With
stoichiometric
mixture
starting
shown in fig. 2.
tent after
to 0
stoichiometry
6.92' seems to be the reduction
taining
atmosphere.
strongly
depending
ison between change
during
results
The reaction
reduction,
evaluated
loss from 0
in oxygen to 6th?s
of high purity and an
the oxygen
stoichiometry
content
only
In above example the oxygen conproof for oxygen
of the superconductor in the reduction
of the atmosphere,
seem to be critical.
gen in the amount of 0
powders
But the better
temperatures
on the hydrogen
laboratories
and cooling
one could evaluate
from this type of experiment tempering
for the oxygen
in hydrogen process are therefore
compar-
Fig. 4 shows the total mass
to 7.51 %. This corresponds
With the general
assumption
to the re'ful: %%n
above,
to loss of oxy-
for the composition
3.42' BaO and Cu, the stoichiometry of the of the reduced material to be Eu Cu 2 3' before the reduction exsuperconductor is calculated as EuBa Cu 0 periment.
This corresponds
con-
after tempering.
Fig:4
Reduction
The reduction
covers
of the relatively gas, sharp
of EuBa Cu U 2 3x
The reoxidation
a broad temperature
low hydrogen
reduction
in hydrogen
content
mass changes
containing
range between
ZOO'C and 8UO'C because
in the atmosphere.
are achieved
of the same material
atmosphere,
around
Using pure hydrogen
400-C.
is shown in fSg. 5. The oxygen content
is calculated as FuBa Cu 0 after this run, but there 2 3 6.89 are intermediate stages with higher oxygen content (6UO'C to 8OO'C). Further
of the sample
tests are necessary
Fig, 5
for identificatSon
of these apparently
in flowing Reoxidatiofl Of Eu6a Cu 0 2 3 3.6
cxygen.
unstable
products.
48 The highest
reaction
imum at 570-C. tetragonal
rate for reoxidation
The latter
transition
A newly developed for reduction
corresponds
described
software
is found at 41O'C and a second max-
to the temperature
by Gallagher
of the orthorhombic
(ref. 2).
program will be used to study the reaction kinetics
and reoxidation.
CONCLUSION
Densification
during
for Y6a2Cu307
can be improved
Thermogravimetry ometry
sintering
intermediate
from the carbonate
sintering
tempering,
superconductors.
products
reduction
Impurities
may influence
and reoxidation
in starting
the exact
of oxygen
stoichi-
experiments
powders
stoichiometric
precursor
in dilatometers.
is shown to be useful for the determination
from preparation,
EuBa*Cu~07-type
of a powder compact by rate controlled
on
and unstable
calculations.
REFERENCES 1
P.K. Gallagher
2
Cr. Sageev (received
a.o. Mat. Re. Bull. 22 (7) 1987
Grader,
P.K. Gallagher
as manuscript
to be published
from the author)
ACKNOWLEDGEMENT We thank Prof. GUntherroth, and the fruitful
discussions
University
of Kdln, for the supply of raw materials
of the test program.