Phase stability study of the pseudobinary system Gd2O 3-Nd2O3 (T1350°C)

Article abstract of DOI:10.1007/s10973-007-8744-x

The phase stability of the system Gd₂O₃-Nd ₂O₃ up to 1350°C was studied. Mixed oxides were synthesized by coprecipitation of mixed oxalates. Depending on the thermal treatment, three possible structures of the oxides, namely the A (hexagonal, P6₃/mmm), B (monoclinic, C2/m) and C (cubic, Ia3) forms, were obtained. Below 1300°C the system Gd₂O₃-Nd ₂O₃ shows three monophasic (A, B and C phase) and three biphasic regions (A+B, B+C and A+C phase). Below 550°C the C phase is stable over the whole composition range. A validation procedure for the existence of the C phase in pure Nd₂O₃ was also set up.

Full text of DOI:10.1007/s10973-007-8744-x

Journal of Thermal Analysis and Calorimetry, Vol. 92 (2008) 1, 101–104
PHASE STABILITY STUDY OF THE PSEUDOBINARY SYSTEM
Gd O –Nd O (T£1350°C)
2
3
2
3
1
*
1
2
3
4
G. A. Costa , C. Artini , A. Ubaldini , M. M. Carnasciali , P. Mele and R. Masini
5
1
DCCI & LAMIA-INFM-CNR, Via Dodecaneso 31, 16146 Genova, Italy
NIMS, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan
2
3
DCCI & INSTM, Via Dodecaneso 31, 16146 Genova, Italy
Kyushu Institute of Technology, Department of Materials Science, Kitakyushu 804-8550, Japan
CNR-IMEM, Via Dodecaneso 33, 16146 Genova, Italy
4
5
The phase stability of the system Gd
mixed oxalates. Depending on the thermal treatment, three possible structures of the oxides, namely the A (hexagonal, P6
B (monoclinic, C2/m) and C (cubic, Ia3) forms, were obtained. Below 1300°C the system Gd –Nd shows three monophasic
A, B and C phase) and three biphasic regions (A+B, B+C and A+C phase). Below 550°C the C phase is stable over the whole com-
2 3
O
–Nd
2
O
3
up to 1350°C was studied. Mixed oxides were synthesized by coprecipitation of
3
/mmm),
2
O
3
2 3
O
(
2 3
position range. A validation procedure for the existence of the C phase in pure Nd O was also set up.
Keywords: phase stability, rare earth sesquioxides, X-ray diffraction
Introduction
the B structure at room temperature because their
transformation rates are low. The sesquioxides of the
largest ions from La O to Nd O are usually reported
to crystallise in the A-type structure [8]. The question
whether they can exist in the C form, if the synthesis
conditions are appropriate [10], is still open in litera-
ture [8, 11–14].
Due to their physical and chemical properties, rare
earth (RE) sesquioxides are applied in many techno-
logical fields, such as catalysts for a large number of
organic reactions [1, 2], host materials for powerful
lasers [3] or improved phosphors [4]. They are also
used as precursors of high critical temperature super-
2
3
2
3
Dealing with mixed rare earth oxides
¢
(RE1–x RE ) O , if the two parent oxides have the
conductors [5] like REBa
2
Cu
3
O
7
, generally indicated
x
2
3
as the most promising materials for power applica-
tions of superconductivity. Particularly, some super-
conducting cuprates containing solid solutions of two
or three RE, exhibit high critical temperature and cur-
rent density [6].
same structure, a complete solid solution which fol-
lows the Vegard’s law forms. Otherwise, two terminal
solid solutions develop with the structure of the par-
ent oxides and a polyphasic system exists for interme-
diate values of x. If the difference between the RE
ionic radii is large enough to satisfy the Goldschmidt
tolerance factor [15], a perovskitic compound forms
for the equimolar composition.
An extensive literature exists on the rare earth
mixed oxides synthesis, generally consisting of a di-
rect solid state reaction between the pure oxides
[16, 17], a combustion synthesis [18] or a copre-
cipitation of mixed hydroxides [16]. Considering the
high thermal stability of these materials and the low
diffusion coefficients, high temperatures, usually
greater than 1400°C, are required in order to attain
equilibrium. For this reason the low temperature re-
gion (T<1300°C) of the phase diagrams is generally
not considered.
It is well known that rare earth sesquioxides
RE O ) exist in five distinct crystalline types de-
2 3
(
pending on their RE ionic radius and on temperature.
These polymorphic modifications are usually labelled
with the letters A and H (hexagonal), B (monoclinic),
C and X (cubic). The X and H forms are stable above
2000°C, while below this temperature A, B and C are
commonly observed [7, 8].
The C form (space group Ia3) is the most stable
for lanthanides from Tb to Lu, while the A form
(
space group P6 /mmm) is typical for lanthanides
3
from La to Nd; the intermediate rare earth oxides exist
both in the B- and C-type polymorphs, the B form
(
1
space group C2/m) being stable above 875, 1000 and
250°C for Sm , Eu and Gd , respec-
tively; [9] however, these oxides can also be found in
O
O
O
This work focuses on the (Gd1–xNd ) O system
3
2
3
2
3
2
3
x 2
(0£x£1). Since these mixed oxides are often used as
*
Author for correspondence: costa@chimica.unige.it
1
388–6150/$20.00
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands
©
2008 Akadémiai Kiadó, Budapest

COSTA et al.
precursors for new materials [5, 19], the existence of a
Results and discussion
complete solid solution at room temperature is very
important and it is necessary to define the existence
fields of the monophasic regions as a function of tem-
perature and composition.
Gd O –Nd O phase stability
2
3
2
3
At 600°C the decomposition of oxalates is complete
for every x value [20]. XRD analyses performed on
samples treated at this temperature show that the
C form is stable over the whole range. As shown in
Fig. 1, the cell parameters follow the Vegard’s law
increasing with x and a complete solid solution forms.
As Nd can exist in the C and the A form,
while Gd O exhibits the C- and B-type structures at
3
O
2
3
2
room temperature and above 1250°C, respectively,
the phase diagram of the mixed oxides system is
rather complex below 1350°C. This leads to large
monophasic and polyphasic regions, depending on
the Gd/Nd ratio. This work reports on the results of
the Gd O –Nd O phase diagram up to 1350°C. The
2 3 2 3
mixed oxides were prepared by thermal decomposi-
tion of the corresponding mixed oxalates.
Tests on the existence of C-type Nd
lized by H O or CO were also performed by thermal
O not stabi-
2 3
2
2
decomposition of Nd(OH) .
3
Experimental
Nd/%
x 2 3
Fig. 1 Vegard’s law for the C-type (Gd1–xNd ) O over the
All (Gd Nd ) [C O ] ·nH O samples (0£x£1) were
2
1–x
x 2
2
4 3
whole composition range
prepared by a coprecipitation method, starting from
commercial Gd and Nd powders (Aldrich; 4N).
Due to the Nd O hygroscopicity, the starting powders
O
O
2 3
2
3
Increasing temperature above 650°C, the C phase
field narrows, so that at 1200°C the C phase extent is
limited to the range 0£x£0.05. Correspondingly, the A
phase field, typical of samples with high Nd content,
2
3
were calcined at 600°C overnight. A solution was
prepared using a slight excess of HCl (10% m/V),
afterwards the precipitation of mixed oxalates was
achieved by adding an excess of a solution of oxalic
acid. The precipitate was filtered and washed with
deionized water till neutrality and successively dried in
air at 80°C for 24 h. The chemical composition of the
so prepared powders was checked by SEM-EDAX.
The precursor powders were quickly heated up
to a defined temperature in a Carbolite STF 15/450
tube furnace (uniform length: ±1°C in 4 cm) in the
range 500°C£T£1350°C and quickly cooled. In order
to attain the final equilibrium, different calcination
times were scheduled depending on the temperature:
widens towards the Gd
200°C. The B field is positioned between the A and
the C field and it becomes larger towards the Gd O
3
2 3
O side, reaching x=0.97 at
1
2
side as the temperature increases, so that over 1250°C
it extends from x=0 to 0.9. Therefore, in the range
650–1250°C three monophasic, as well as two biphasic
regions (C+B and B+A) between them are present;
above 1250 up to 1350°C, two monophasic (B and A)
and one biphasic regions are observed.
In the monophasic areas the Vegard’s law is fol-
lowed, while in the biphasic regions, as expected, the
cell parameters do not change. The positions of the
phase boundaries have been determined by X-ray dif-
fraction, calculating the percentage amount of each
phase from the ratio between the areas of the most in-
tense peaks, according to the following equation:
2
4 and 72 h for T>1000°C and for 800£T£1000°C
range, respectively. Below 800°C a one week calcina-
tion was performed.
C-type Nd
starting from Nd(OH) powders heated in static air at
O samples were also synthesized
2 3
3
max
5
50°C and then in flowing O (99.999%) at the same
2
Aa
max
Aa +Ab
a %
=
×1 00
max
temperature to remove the residual water. Effusion
technique was used in order to attain a complete
decomposition.
max
where a is the a phase percentage amount, A and
%
a
max
The boundaries between the phase fields were
determined studying the structures of all the samples
by X-ray powder diffraction, using a Philips PW1830
Ab are the most intense peak areas of the a and
b phase, respectively. In Table 1 the data concerning
the percentage amounts of B and C phase for some T
and x values are reported.
On the basis of the XRD patterns of all the con-
sidered samples, it is possible to draw a representation
diffractometer (CuK
5°<2q<78°.
a
radiation) in the range
1
102
J. Therm. Anal. Cal., 92, 2008

2 3 2 3
PHASE STABILITY STUDY OF THE PSEUDOBINARY SYSTEM Gd O –Nd O
Table 1 Percentage amounts of B and C phase for some T
and x values
can be explained by the different features of the three
crystal structures. Since density and coordination
number increase moving from C to B to A, the C form
T/°C
x
B phase/%
C phase/%
3
+
could easily accept a partial substitution of Gd by a
bigger cation, while this is more difficult for the
A form. Similarly, the boundary between the B and
the B+A field has a steeper slope than the one
detected between the B and the B+C field.
9
9
9
8
8
7
7
50
50
00
50
00
50
50
0.30
0.40
0.40
0.60
0.70
0.78
0.80
42
75
13
4
58
25
87
96
82
40
30
The phase stability in rare earth sesquioxide sys-
tems is driven by three main factors: temperature,
pressure and cationic size, since a pressure decrease
causes the stabilization of the B form [9, 21], the sub-
18
60
70
3
+
3+
stitution of Nd by a smaller cation as Gd , acting
like a decrease of an external applied pressure, pro-
motes the B type arrangement of the atoms. On the
of the Gd O –Nd O phase diagram in the range
2
2
3
3
5
50–1350°C, as shown in Fig. 2. The dashed line in
3
+
the Nd rich part at low temperature is referred to a not
yet well defined region where many phase boundaries
may be present.
other side of the diagram, the substitution of Gd by
3
+
Nd behaves in the opposite way widening the
C field.
As can be observed from the phase diagram, in-
It is noteworthy that for x=0.5, where the Gd/Nd
average cationic radius is as large as that correspond-
3+
ing to Sm ion (1.100 and 1.098 , respectively), the
3
+
creasing temperature the Nd solubility in C-type
3
Gd O decreases, while the Gd solubility in A-type
+
2
3
Nd
2
O
3
slightly increases; moreover, at any tempera-
transition temperature between the C and C+B fields,
around 850°C, is very close to that reported for the
transition C/B for Sm O (875°C) [9], indicating that
the average ionic radius plays an essential role in the
phase stability of these systems.
ture above 600°C, the transition C®B®A occurs in-
creasing the Nd content. This evidence can not be ex-
plained considering the crystallographic properties of
the involved structures. Phase transitions associated
to a coordination number and density increase, as well
as to a contemporary symmetry decrease, as observed
moving from C to B to A, are in fact expected to be fa-
voured by a temperature decrease or a mean cationic
radius reduction. However, the same phenomenon is
observed over the whole series of the rare earth ses-
quioxides, as the transition C®B®A is well estab-
lished at any temperature below 2000°C moving from
the biggest to the smallest rare earth ions [7].
From the analysis of the phase diagram it can
also be noticed that the boundary separating the A
from the A+B field is characterized by a very steep
slope, in contrast to what is observed between the C
and the B+C field, meaning that the solubility of Gd
in the A structure is very low in comparison with the
+
solubility of Nd in the C structure. This evidence
2
3
C-type Nd
O
2 3
The thermodynamic stability of C-type Nd O is a
2
3
controversial point in literature: even though the tran-
sition C®B or C®A occurs at gradually lower tem-
peratures moving from heavier to lighter rare earths
[22], the C field is reported to extend from Nd to Lu
oxide [11], and in fact several authors report on the
successful synthesis of C-Nd O [23–26]. For exam-
2
3
ple, recently Tong and Eyring [13] reported a TEM
study of the in situ thermal decomposition of
Nd(OH) into C-type Nd O at about 450°C. Never-
3
2
3
3
+
theless, some works question Nd O thermodynamic
2 3
stability, stating that it could be metastable [8, 27] or
stabilized by impurities, such as CO or H O [28]. In
3
2
2
order to determine its thermodynamic stability in the
range 550–600°C and whether it is stabilized by the
presence of H
performed.
O or CO , the following test has been
2 2
Nd(OH)
quartz tube at 550°C for three days. The obtained sam-
ple consists of Nd ·0.3H O, as calculated by the ob-
3
powders have been heated in air in a
O
2
3
2
served mass loss, and the X-ray analysis shows that it
belongs to the C form. To investigate whether the
C phase is stabilized by the presence of water, the sam-
ple has been reheated at 550°C in flowing O
to ensure
2
the removal of water. A sample of anhydrous Nd O
2
3
has been obtained and its diffraction pattern, reported
Fig. 2 Pseudobinary phase diagram of the system
2 3 2 3
Gd O –Nd O . Lines are a guide for the eyes
J. Therm. Anal. Cal., 92, 2008
103

COSTA et al.
4
5
E. Antic-Fidancev, J. Hölsa and M. Lastusaari, J. Alloys
Compd., 341 (2002) 82.
P. Mele, C. Artini, R. Masini, G. A. Costa, A. Hu,
N. Chikumoto and M. Murakami, Physica C,
3
91 (2003) 49.
A. Ubaldini, F. Giovannelli and I. Monot-Laffez, Phys. C,
83 (2003) 107.
6
7
3
L. Eyring, Handbook of Physics and Chemistry of Rare
Earths, K. A. Gschneidner and L. Eyring, Eds, North
Holland 1979, Vol. 3, p. 337.
8
9
G. Adachi and N. Imanaka, Chem. Rev., 98 (1998) 1479.
H. R. Hoekstra, Inorg. Chem., 5 (1966) 754.
10 N. Imanaka, T. Masui and Y. Kato, J. Solid State Chem.,
(
11 J. P. Traverse and M. FoÁx, Rev. Int. Hautes Temp.
2005) 395.
2q/degree
Refract., 3 (1966) 429.
2 N. Hirosaki, S. Ogata and C. Kocer, J. Alloys Compd.,
2 3
Fig. 3 Nd O experimental and theoretical C-type XRD
1
patterns. * – two A-phase main peaks
351 (2003) 31.
1
1
1
3 J. Tong and L. Eyring, J. Alloys Compd., 225 (1995) 139.
4 M. Zinkevich, Prog. Mater. Sci., 52 (2007) 597.
5 V. M. Goldschmidt, T. Barth, G. Lunde and
W. Zachariasen, Skr. Norske Vidensk.-Akad. Oslo I. Mat.
Naturvidensk. K1., 2 (1926).
in Fig. 3 shows the C-type peaks, as well as the two
A phase main peaks as a minor impurity.
The same thermal treatment performed on
Nd
firming that Nd O is stable at 550°C in the C form
[C O ] ·10H O resulted in a C-type Nd O , con-
2 2 4 3 2 2 3
1
6 L. Eyring, Handbook of Physics and Chemistry of Rare
Earths, K. A. Gschneidner and L. Eyring, Eds, North
Holland 1979, Vol. 3., 401E I. Warshaw and R. Roy,
J. Phys. Chem., 65 (1961) 2048.
2
3
without retaining H O or CO .
2
2
1
1
7 J. P. Coutures, R. Verges and M. FoÁx, Rev. Int. Hautes
Temp. Refract., 12 (1975) 181.
Conclusions
8 J. C. Panitz, J. C. Mayor, B. Grob and W. Durisch,
J. Alloys Compd., 303–304 (2000) 340.
The pseudobinary phase diagram of the system
Gd –Nd has been studied in the range
50–1350°C by X-ray diffraction. The samples have
O
O
19 P. Mele, C. Artini, R. Masini, G. A. Costa, A. Hu,
N. Chikumoto and M. Murakami, Phys. C,
412–414 (2004) 196.
2
3
2
3
5
been prepared by thermal decomposition of mixed
oxalates at different temperatures and for different
times, in order to ensure the reaction completion.
In the range 600–1250°C three monophasic and
two biphasic regions can be observed, while above
2
2
2
2
0 A. Ubaldini, C. Artini, G. A. Costa, M. M. Carnasciali and
R. Masini, J. Therm. Anal. Cal., in press.
1 J. O. Sawyer, B. R. Hyde and L. Eyring, Inorg. Chem.,
3
(1965) 426.
2 G. Brauer and E. Mohr-Rosenbaum, Z. Anorg. Allg.
Chem., 394 (1972) 301.
3 K. Lohberg, Z. Phys. Chem., [B] 28 (1935) 402.
1
250°C only the A, B and A+B fields are present.
The existence of Nd in the C form has been
proved: by the analysis of the experimental results it
is possible to conclude that the Nd C form is not
O or CO
O
3
2
24 H. Bommer, Z. Anorg. Allg. Chem., 241 (1939) 273.
25 A. Iandelli, Gazz. Chim. Ital., 77 (1947) 312.
26 C. Boulesteix, Handbook of Physics and Chemistry of
Rare Earths, K. A. Gschneidner and L. Eyring, Eds,
North Holland 1982, Vol. 5, p. 321.
O
3
2
stabilized by the presence of H
2
.
2
2
2
7 J. Felsche, Naturwissenschaften, 4 (1969) 212.
8 P. P. Fedorov, M. V. Nazarkin and R.M. Zakalyukin,
Crystallography Reports, 47 (2002) 316.
References
1
2
3
A. G. Dedov and A. S. Loktev, Appl. Catal. A: Gen.,
45 (2003) 209.
2
R.-Q. Liu, Y.-H. Xie, J.-D. Wang, Z.-J. Li and
B.-H. Wang, Solid State Ion., 177 (2006) 73.
R. Reisfeld and K. Jorgensen, Lasers and excited states of
rare earths, Springer, New York 1978.
DOI: 10.1007/s10973-007-8744-x
104
J. Therm. Anal. Cal., 92, 2008