New phases in the ZrO2-Bi2O3 and HfO2-Bi2O3 systems

Article abstract of DOI:10.1016/S0025-5408(98)00076-2

New phases of the type M_1-xBi_xO_2-x/2 with a defect fluorite structure have been prepared where M is Zr or Hf. These apparently metastable phases were prepared by precipitation from solution followed by calcination at 600°C. For the Hf_1-xBi_xO_2-x/2 system, x ranges from x = 0.40 to x = 0.75. For the Zr_1-xBi_xO_2-x/2 system, x ranges from x = 0.50 to x = 0.75. Heating the defect fluorite Hf_1-xBi_xO_2-x/2 phases to temperatures of 700 to 950°C leads to the formation of Bi₂Hf₂O₇ with the Bi₂Sn₂O₇ structure. Thus, Bi₂Hf₂O₇ is pseudocubic, but apparently face centered tetragonal with a = 21.66 A and c = 21.84 A. When Bi₂Hf₂O₇ is heated to 1000°C, it decomposes into HfO₂ and Bi_1.94Hf_0.06O_3.03 with the β-Bi₂O₃ structure. Heating the defect fluorite Zr_1-xBi_xO_2-x/2 phases to higher temperatures never produced a Bi₂Zr₂O₇ phase with the Bi₂Sn₂O₇ structure. Instead, the compositional range of the defect fluorite structure steadily decreases with increasing temperature, and this phase had completely decomposed at 750°C into ZrO₂ and Bi_1.84Zr_0.16O_3.08 With the β-Bi₂O₃ Structure.

Full text of DOI:10.1016/S0025-5408(98)00076-2

Materials Research Bulletin, Vol. 33, No. 7, pp. 1077–1081, 1998
Copyright © 1998 Elsevier Science Ltd
Printed in the USA. All rights reserved
0025-5408/98 $19.00 ϩ .00
PII S0025-5408(98)00076-2
NEW PHASES IN THE ZrO2–Bi2O3 AND HfO2–Bi2O3 SYSTEMS
S.L. Sorokina and A.W. Sleight*
Department of Chemistry and Center for Advanced Materials Research, Oregon State
University, Corvallis, OR 97331-4003 USA
(Refereed)
(Received November 21, 1997; Accepted December 13, 1997)
ABSTRACT
New phases of the type M1ϪxBixO2Ϫx/2 with a defect fluorite structure have
been prepared where M is Zr or Hf. These apparently metastable phases were
prepared by precipitation from solution followed by calcination at 600°C. For
the Hf1ϪxBixO2Ϫx/2 system, x ranges from x ϭ 0.40 to x ϭ 0.75. For the
Zr1ϪxBixO2Ϫx/2 system, x ranges from x ϭ 0.50 to x ϭ 0.75. Heating the
defect fluorite Hf1ϪxBixO2Ϫx/2 phases to temperatures of 700 to 950°C leads
to the formation of Bi2Hf2O7 with the Bi2Sn2O7 structure. Thus, Bi2Hf2O7 is
pseudocubic, but apparently face centered tetragonal with a ϭ 21.66 Å and
c ϭ 21.84 Å. When Bi2Hf2O7 is heated to 1000°C, it decomposes into HfO2
and Bi1.94Hf0.06O3.03 with the ␤-Bi2O3 structure. Heating the defect fluorite
Zr1ϪxBixO2Ϫx/2 phases to higher temperatures never produced a Bi2Zr2O7
phase with the Bi2Sn2O7 structure. Instead, the compositional range of the
defect fluorite structure steadily decreases with increasing temperature, and
this phase had completely decomposed at 750°C into ZrO2 and
Bi1.84Zr0.16O3.08 with the ␤-Bi2O3 structure. © 1998 Elsevier Science Ltd
KEYWORDS: A. oxides, C. X-ray diffraction, D. ionic conductivity
INTRODUCTION
No significant literature appears to exist on the Bi2O3–HfO2 system. The reported phase
diagram [1] for the Bi2O3–ZrO2 system indicates a small solubility of ZrO2 in the various
forms of Bi2O3 but no compound formation and undetectable solubility of Bi2O3 in ZrO2.
*To whom correspondence should be addressed.
1077

1078
S.L. SOROKINA and A.W. SLEIGHT
Vol. 33, No. 7
Hund [2] reported a tetragonal phase (a ϳ 10.97 Å and c ϳ 5.58 Å) for Zr1ϪxBixO2Ϫx/2
compositions prepared at 800°C where x ranged from 0.3 to 1.0. This would basically be a
very large solubility of ZrO2 in ␤-Bi2O3. This claim of Hund is not supported by our work
or the work of others [1].
EXPERIMENTAL
Reactants were A.C.S. Reagent Grade Bi(NO3)3⅐5H2O (Aldrich) and zirconium and hafnium
oxynitrates (99.9%, Wah Chang). These reactants were first dissolved separately in water
with the pH of the bismuth nitrate solution brought to 2.0 by the addition of nitric acid. These
0.1 N solutions were then mixed in various ratios depending on the desired Bi:Hf or Bi:Zr
ratio. The mixed solutions were heated to about 55°C. The solutions were then neutralized
by the addition of ammonium hydroxide (10% NH3) with constant stirring. A precipitate
forms immediately, but stirring was continued for 2 h. The precipitate was washed with
distilled water and dried at about 95°C for 10 h. This product was ground in an agate mortar
and pestle, pressed into a pellet, and heated at various temperatures from 300 to 1000°C for
10 h in air.
X-ray powder patterns were obtained using a Siemens D-5000 diffractometer with Cu K␣
radiation. Cell dimensions were refined by least squares. Several samples that appeared single
phase by X-ray diffraction were examined by electron microprobe analysis. This confirmed
the homogeneity of the samples and also that the compositions were as expected from the
ratio of reactants used. Microprobe analysis was used to estimate the amounts of ZrO2 and
HfO2 that dissolved in Bi2O3. Microprobe analysis could not detect any solubility of Bi2O3
in either ZrO2 or HfO2.
RESULTS
X-ray diffraction on products that had been heated only to 95°C showed these products to be
amorphous. Heating to 600°C or higher led to crystalline products in both the ZrO2–Bi2O3
and HfO2–Bi2O3 systems.
Heating to 600°C in the hafnium system gave a Hf1ϪxBixO2Ϫx/2 defect fluorite phase
from x of about 0.40 to about 0.75. The unit cell edge was 5.266 Å for the x ϭ 0.40
sample and increased monotonically to 5.459 Å for the x ϭ 0.75 sample. Samples with
lower x contain ␣-Bi2O3 as a second phase, and samples with higher x contain mono-
clinic HfO2 as a second phase. With increasing temperature, the compositional range of
the Hf1ϪxBixO2Ϫx/2 defect fluorite phase decreases somewhat. At about 700°C a
Bi2Hf2O7 compound with the Bi2Sn2O7 structure is formed. This phase is stable to
1000°C, where it decomposes into HfO2 and Bi1.94Hf0.06O3.03 with the ␤-Bi2O3 structure.
The X-ray diffraction powder pattern of Bi2Hf2O7 is given in Table 1. The pseudocubic
cell edge of this phase is twice that expected for the normal pyrochlore structure and four
times that of the related fluorite structure. Because of this relationship to the face-
centered cubic fluorite structure, the indices of the strongest reflections obey the face-
centered rules when divided by four, i.e., the 4 4 4, 8 0 0, and 8 8 0 reflections.
Reflections with odd indices are superstructure reflections relative to the pyrochlore
structure. The strongest of the superstructure reflections are the 5 3 1, 7 3 1, and 9 3 1;
these are also the strongest superstructure reflections for Bi2Sn2O7 [3]. The pattern is

Vol. 33, No. 7
BISMUTH ZIRCONATE HAFNATE
1079
TABLE 1
X-ray Diffraction Powder
Pattern for Bi2Hf2O7
h k l
d
I
2 2 2
6.27
W
4 0 4
4 4 0
3.845
3.829
M
M
3 1 5
5 1 3
5 3 1
3.683
3.669
3.662
2 2 6
6 2 2
3.288
3.268
M
4 4 4
3.135
VS
3 1 7
7 1 3
7 3 1
2.839
2.823
2.820
M
0 0 8
8 0 0
2.730
2.707
S
S
6 2 6
6 6 2
2.494
2.486
M
M
4 0 8
8 0 4
8 4 0
2.438
2.426
2.422
3 1 9
9 1 3
9 3 1
2.287
2.272
2.271
M
2 2 10
6 6 6
10 2 2
2.100
2.090
2.085
M
S
8 0 8
8 8 0
1.922
1.914
pseudocubic with only the 8 0 0 reflection showing well-resolved splitting. However, the
shapes of the other peaks are consistent with a tetragonal distortion. Because Bi2Hf2O7
cannot be heated higher than 950° without decomposition, high-temperature annealing
cannot be used to obtain a sharper X-ray diffraction pattern. The unit dimensions of
Bi2Sn2O7 at room temperature are a ϭ 21.328 Å and c ϭ 21.545 Å [3]. The cell

1080
S.L. SOROKINA and A.W. SLEIGHT
Vol. 33, No. 7
dimensions of Bi2Hf2O7 are estimated to be a ϭ 21.66 Å and c ϭ 21.84 Å. Due to the
unresolved peak splitting, the unit-cell volume based on this cell is more accurate than
the cell edges. The conventional body centered tetragonal cell for Bi2Hf2O7 is a ϭ 15.32
and c ϭ 21.84 Å. A second harmonic signal was observed for Bi2Sn2O7, which
disappeared on heating to 680°C [3]. We also observed a significant a second harmonic
signal for Bi2Hf2O7 consistent with Bi2Hf2O7 and Bi2Sn2O7 being isostructural at room
temperature.
After heating at 600°C in the zirconium system, a Zr1ϪxBixO2Ϫx/2 defect fluorite phase is
obtained from x about 0.50 to about 0.75. The unit-cell edge for the x ϭ 0.50 sample was 5.39
Å, and this increased monotonically up to 5.473 Å for x ϭ 0.75. Increasing the temperature
of calcination did not give a Bi2Zr2O7 phase with a Bi2Sn2O7-type structure. The range of
composition of the Zr1ϪxBixO2Ϫx/2 defect fluorite phase decreased with increasing temper-
ature. Due to this narrowing of the phase field, samples calcined at 700°C showed only a
fluorite-type cell edge of 5.439 Å indicative of a Zr1ϪxBixO2Ϫx/2 sample with an x value of
about 0.60. Samples calcined at 750°C or higher showed no Zr1ϪxBixO2Ϫx/2 defect fluorite
phase; only the equilibrium phases ZrO2 and Bi1.84Zr0.16O3.08 were present.
DISCUSSION
The new compound Bi2Zr2O7 is the first example of a compound isostructural with Bi2Sn2O7.
The precise structure of Bi2Sn2O7 is unknown, but it is assumed to be pyrochlore related
[3,4]. Relative to the normal pyrochlore structure, Bi2Sn2O7 has a superstructure requiring
doubling the unit-cell edges and the symmetry has dropped from cubic to tetragonal. At about
90°C, Bi2Sn2O7 becomes cubic, but the superstructure remains. At about 680°C, the super-
structure disappears. The structure of Bi2Sn2O7 has recently been refined based on neutron
diffraction powder data obtained at 723°C [4]. That study confirmed that Bi2Sn2O7 does have
the pyrochlore structure, at least above 680°C. Finding Bi2Hf2O7 may be helpful in solving
the elusive room-temperature structure of Bi2Sn2O7. In view of the ease of formation of
Bi2Hf2O7 with the Bi2Sn2O7 structure, the lack of occurrence of a Bi2Zr2O7 compound with
the Bi2Sn2O7 structure is surprising.
Defect fluorite phases of the type M1ϪxRxO2Ϫx/2 are very well known where M is Zr or Hf
and R is a rare earth. Such phases have not been reported where R is Bi, except by Hund [2],
and these findings have not been confirmed by others [1,5]. Thus, we believe that we have
prepared the first M1Ϫ-xRxO2Ϫx/2 defect fluorite phases where R is Bi. The M1ϪxRxO2Ϫx/2
phases where R is a rare earth are notable for their high oxygen conductivity. The lack of
thermal stability of our samples prevents us from obtaining well-sintered dense pellets.
Preliminary measurements [6] on low-density pellets of our Zr1ϪxBixO2Ϫx/2 phases do not
suggest high oxygen conductivity, despite the fact that Bi3ϩ frequently seems to promote
high oxygen conductivity in many mixed oxide systems.
ACKNOWLEDGMENT
We thank Dr. Shiv Halasyamani and Dr. Pam Thomas for assistance with the SHG mea-
surement. This work was supported through NSF grant DMR-9308530.

Vol. 33, No. 7
BISMUTH ZIRCONATE HAFNATE
1081
REFERENCES
1. E.M. Levin, C.R. Robbins, and H.F. McMurdie, in Phase Diagrams for Ceramists, ed. M.K.
Reser, p. 128, American Ceramic Society, Westerville, OH (1964).
2. V.F. Hund, Z. Anorg. Allg. Chem. 333, 248 (1964).
3. R.D. Shannon, J.D. Bierlein, J.L. Gillson, G.A. Jones, and A.W. Sleight, J. Phys. Chem. Solids
41, 117 (1980).
4. V. Kahlenberg and Th. Zeiske, Z. Kristallogr. 212, 297 (1997).
5. X.L. Chen and W. Eysel, J. Solid State Chem. 127, 128 (1996).
6. M.A. Subramanian, E. I. Dupont Co., private communication.