Rapid Communications
The whole arrangement was moved towards the samples
surface with a speed of 2 mm/s with a scanning interval
of 0.5 mm. The critical current density, Jc, was investi-
gated by a conventional pulsed current four-probe
method with a criterion of 10 V cm−1 at 77 K at zero
external field. The sample size for this Jc measurement
was 0.5 × 0.5 mm2 with a contact distance of 6 mm. The
microstructure of the YBCO monoliths was character-
ized by optical microscopy (OM) and scanning electron
microscopy (SEM). The size of Y-211 particles the
TSMG monoliths was determined by a linear intercept
method from SEM photographs.7
The determination of a suitable temperature profile is
crucial for the preparation of large single-domain mono-
liths with a high quality of the textured matrix. First, the
presintered samples are heated under normal atmosphere
above the peritectic temperature of Y-123 up to 1045 °C
with a heating rate of 70 °C/h. The samples are main-
tained at this temperature for 0.2 h to ensure the complete
reaction of the precursor material. This dwell time has to
be optimized with respect to reduction of Ostwald rip-
ening of Y-211 particles as well as the minimization of
the loss of liquid. The latter leads to a depletion of Cu
which results in the suppression of the growth of the
Y-123 phase.8 These first steps of the temperature profile
also determine the morphology of the pores of the mono-
liths.9 Using optimized heating conditions, the pores are
homogeneously distributed over the interior part of the
monoliths with an average size of approximately 47 m.
A fast cooling step down to the temperature where the
crystallization of Y-123 starts follows the high tempera-
ture dwell. The solidification window for the heteroge-
neous formation of the Y-123 phase at the Sm-123 seed
crystal was determined by quenching experiments. At
998 °C, the undercooling of the melt is large enough for
the cyrstallization of the Y-123 phase at the Sm-123 seed
without the formation of any other crystallites at the sur-
face of the sample. On reaching this temperature, the
bottom zone heater is set 20 °C higher than the other
heaters to generate a thermal gradient of approximately
1 °C/cm over the samples height during the following
cooling steps. The equilibrium temperature at the liquid–
solid interface depends on the initial composition and the
use of Y2O3 admixtures instead of excess Y-211 to the
starting powder increases the size of the processing win-
dow of stable growth conditions for the Y-123 phase
from the melt (Fig. 1). Unlike in the 211-123 case, the
reaction 211 + L + O2 → 123 is no longer univariant. As
Fig. 1 depicts, crystallization commences by reaching the
critical supersaturation (p) and solidification of the 211 +
L mixture can proceed stable down to 940 °C following
the p-p1 line.10 Thus, the texturing process for the Y2O3-
123 system is more resistant against fluctuations in the
stoichiometric composition than for the 211 + 123 sys-
tem in which loss of melt occurs at high temperatures.
FIG. 1. Polythermic YO1.5−Ba0.4Cu0.6O section in the Y–Ba–Cu–O
phase diagram at an oxygen partial pressure of 0.21 bar.
Several slow cooling steps of 0.5–1 °C/h interrupted
by several dwell times follow. Once the furnace tempera-
ture of 940 and 960 °C for the bottom zone is reached, a
uniform cooling of all heating zones, ramped down to
room temperature with 18 °C/h, completes the thermal
profile of the melt-texturing process. Finally, the textured
YBCO monoliths are annealed at 485 °C under flowing
oxygen for five days in order to obtain fully oxigenated
superconductors.
Another important factor for the fabrication of high
quality single-domain YBCO monoliths is the use of an
optimized substrate material. The Y-211 phase possesses
a thermal expansion similar to that of the Y-123 phase
and makes an excellent substrate material, minimizing
the mechanical stresses caused by thermal mismatch dur-
ing the texturing process.11 Thus, the amount of cracks
formed during the cooling steps can be reduced signifi-
cantly by employing high-density Y-211 substrates,
while all other materials tested however, such as MgO,
Al2O3, and BaZrO3, induced severe cracking during the
cooling process. In addition, the loss of liquid is further
reduced to less than 2% for samples melt-textured on
Y-211 substrates with a density of more than 70%.
Due to this very low amount of liquid loss the shape
of the melt-textured samples undergoes only slight
changes during the texturing process. The melt-textured
monoliths had dimensions of about 38 × 38 × 12 mm.
The sample shrinkage is mainly due to an increase of
the density from 4.5 to 5.7 g/cm3 during the texturing
process.
The grain size and the amount of Y-211 particles,
which are trapped in the growing Y-123 monodomain
material, strongly influence the superconducting prop-
erties of the melt-textured monoliths. These particles act
as flux pinning centers which determine the supercon-
ducting properties of the sample such as the critical cur-
1232
J. Mater. Res., Vol. 15, No. 6, Jun 2000
Downloaded: 13 Mar 2015
IP address: 129.78.139.28