Hydrogen reduction of Cu-Mn-Fe-O spinel solid solutions

Article abstract of DOI:10.1007/s10789-005-0122-0

The crystal-chemical transformations accompanying the hydrogen reduction of Cu_xMn₁ - xFe₂O₄ (x = 0.2, 0.5, 0.8) solid solutions are investigated, and general equations describing these processes over the entire

Full text of DOI:10.1007/s10789-005-0122-0

Inorganic Materials, Vol. 41, No. 3, 2005, pp. 272–278. Translated from Neorganicheskie Materialy, Vol. 41, No. 3, 2005, pp. 332–338.
Original Russian Text Copyright © 2005 by E. Zinovik, M. Zinovik.
Hydrogen Reduction of Cu–Mn–Fe–O Spinel Solid Solutions
E. V. Zinovik and M. A. Zinovik
Kirovograd National Technical University, Universitetskii pr. 8, Kirovograd, 25006 Ukraine
e-mail: ikc@im.net.ua
Received November 13, 2003; in final form, June 30, 2004
Abstract—The crystal-chemical transformations accompanying the hydrogen reduction of Cu Mn Fe O
2 4
x
1 – x
(
x = 0.2, 0.5, 0.8) solid solutions are investigated, and general equations describing these processes over the
entire composition range (0 ≤ x ≤ 1) are derived. The results on the character of the processes involved differ
radically from earlier findings. The equilibrium oxygen pressure is evaluated for the low-oxygen phase bound-
ary
MnFe O ) (Cu Fe O ) (CuFe O )
4
1 – x – x
1 2
of
the
spinel
solid
solutions
throughout
their
stability
region.
For
(
solid solutions, which are of practical interest, the
2
4
x
0.5 2.5
4
x
2
1
2
log pO2 (1/T) data are well represented by the equation log(pO2 /Pa) ± 0.4 = 12.8 – 3.7x – 2.8x – 10280/T,
1
2
which is of importance in controlled synthesis of single-phase materials in this system.
INTRODUCTION
Phase transitions and changes in the cation compo-
sition of solid solutions influence their p (η) and a(η)
curves. The reduction process can be divided into six
steps.
O
2
The presence of metals of variable valence in Cu–
Mn–Fe–O spinel solid solutions and the Jahn–Teller
2
+
3+
nature of the Cu and Mn ions [1] may lead, under
certain conditions, to complex redox processes and
structural transformations in these materials, resulting
in structure-sensitive properties of practical impor-
tance, in particular, square hysteresis loops [2]. Since
these solid solutions are attractive electronic materials
During the first step, pO2 decreases, and a rises
sp
(
Fig. 1).According to x-ray diffraction (XRD) data, this
step involves the formation of the rhombohedral phase
CuFeO , which coexists with the spinel phase. Based
2
on earlier results [8, 9], we conclude that these changes
[
3], the study of their formation and decomposition is
are associated with the reduction of CuFe O and an
2
4
of considerable interest. Such information is also
important in the processing of polymetallic ores [4].
These issues, however, have not yet been studied in suf-
ficient detail [5].
increase in the content of Cu Fe O , which has larger
0
.5
2.5
4
a and lower p .
O
2
Therefore, the crystal-chemical transformations
involved in the reduction of Cu Mn Fe O solid solu-
For example, only the initial steps of reduction pro-
x
1 – x
2
4
cesses have been investigated. Some of the reported tions (0 < x < 1) in this step can be represented by the
data are in conflict with well-known laws of thermody- scheme
namics. No equations have been reported for the com-
Cu Mn Fe O + mH
plete reduction of the solid solutions in question. Data
on the synthesis and cooling conditions for single-
phase samples are also missing.
x
1 – x
2
4
2
=
(1 – m)[(CuFe O )
x – 2m/(1 – m)
2
4
(
1)
×
(Cu Fe O )
(MnFe O )
]
1 – x/(1 – m)
This paper addresses the points raised above.
0.5
2.5
4
m/(1 – m)
2
4
+
1.5mCuFeO + mH O.
2 2
RESULTS AND DISCUSSION
At the end of this step, CuFe O is fully reduced (all
2
4
2
+
+
The synthesis procedure and characterization tech- of the Cu is reduced to Cu ). It follows from (1) that,
niques we used were described in detail elsewhere [6, 7]. for x = 1, 0.8, 0.5, and 0.2, this occurs at m = 0.5, 0.4,
0
.25, and 0.1, respectively, or at η = 12.5, 10, 6.25, and
.5% (m = η/25). The corresponding compositions of
Figure 1 shows the equilibrium oxygen pressure
and the lattice parameters of the spinel and wüstite
2
p
O2
the spinel phase are represented in Fig. 2 by points B
(
B
Cu Fe O ),
B1
)
((Cu Fe O ) (MnFe O ) ),
phases as functions of the degree of reduction η for
0.5
2.5
4
0.5 2.5 4 2/3 2 4 1/3
Cu Mn Fe O spinel solid solutions (x = 0.8, 0.5, 0.2)
2
((Cu0.5Fe2.5
O
4
1/3(MnFe
2
O
4
)
2/3),
and
associated
changes in the composition of the spinel phase and
B
3
x
1 – x
2
4
at 1270 K (η = 100% corresponds to the complete ((Cu0.5Fe2.5
O
)
4
0.11(MnFe
O
)0.89).
The
2
4
removal of oxygen from the solid solution).
0
020-1685/05/4103-0272 © 2005 Pleiades Publishing, Inc.

HYDROGEN REDUCTION OF Cu–Mn–Fe–O SPINEL SOLID SOLUTIONS
273
log(p /Pa)
aw, Å a , Å
sp
O
2
4
2
0
2
4
6
8
(
a)
4
.45
8.52
× [8]
4
.43
1
'
2'
8
.50
.48
.46
.44
.42
1
2
–
–
–
–
4.41
4.39
4.37
4.35
4.33
4.31
8
8
8
8
3
'
1
'
2
'
2
1
'
1
'
2'
3
1
'
1
–
–
10
12
1
'
2
8.40
0
10
20
30
40
50
80
90
100
log(p /Pa)
aw, Å a , Å
O
sp
2
4
2
0
2
4
6
8
(
b)
4.45
4.43
4.41
4.39
4.37
4.35
8.47
×
[8]
3
1
8
.45
2
–
–
–
–
8.43
1
8
.41
2
2
1
–
10
8.39
0
10
20
30
40
50
80
90
η, %
Fig. 1. (1, 1') Equilibrium oxygen pressure and lattice parameters of the wüstite (2, 2') and spinel (3, 3') phases vs. degree of reduc-
tion for Cu Mn Fe O solid solutions with x = (a) (1–3) 0.8, (1'–3') 0.2, and (b) 0.5 at 1270 K.
x
1 – x
2 4
INORGANIC MATERIALS Vol. 41 No. 3 2005

2
74
E.V. ZINOVIK, M.A. ZINOVIK
Ä'
Ä'1
4
[8]
2
Ä'2
0
Ä'3
–2
log(p /Pa)
O
2
B'1
B'
B'2
2
0
2
4
6
8
2
–4
–6
–8
–10
–12
–14
–16
B'3
0
C'3
D'3
C'2
–
–
–
–
–2
–4
D'2
C'1
D'1
C'
–
1.8
E'3
E'2
–
–
–
–
10
12
14
16
E'1
–18
Ä
E'
Ä1
–8
CuFe O
2
4
0
.2
Ä2 0.4
0
.8
Ä3
0.6
–
18
0
.8
B3
B2
0.6
B1
D1
B
D3
MnFe O
2
4
D2
0
.4
Cu Fe O
0.5 2.5 4
E3
C2
C3
C1
E2
0
.2
C
E1
Cu0.15Fe2.85O4
E
Fe O
3
4
Fig. 2. Equilibrium oxygen pressure vs. composition of the spinel phase for Cu Mn
Fe O solid solutions reduced at 1270 K.
2 4
x
1 – x
log pO2 are represented in Fig. 2 by theA–B, A –B , A – A –B cuts coincide with experimental data to within
1
1
2
3
3
1
the measurement accuracy, which lends support to the
proposed mechanism, represented by scheme (1).
B , A –B , A'–B', Ä' –B' , Ä' –B' , and Ä' –B' lines
2
3
3
1
1
2
2
3
3
in the (CuFe O ) (Cu Fe O ) (MnFe O )
2
4
C
0.5 2.5
4
C
2
4
1 – C – C
1
2
1
2
In the second step of reduction, the phase composi-
composition triangle. The data for x = 1 are taken from
Zalazinskii et al. [8]. According to Shchepetkin et al.
10], the composition dependence of the lattice param-
tion remains unchanged, but the dependences of a on η,
pO2 on η, and pO2 on composition change sharply
[
eter for the solid solutions in question follows the addi- because of the qualitative changes in the composition of
tivity rule,
the spinel phase: magnetite (Fe O ) replaces CuFe O
3 4 2 4
2
+
2+
(
Fe instead of Cu ). The composition of the spinel
phase
varies
Cu Fe O ) (Fe O ) (MnFe O )
4
1 – C – C
2 3
within
the
a(Å) = 8.389C + 8.414C + 8.511(1 – C – C ), (2)
1
2
1
2
(
compo-
0
.5
2.5
4
C
3
4
C
2
2
3
where C and C are the mole fractions of CuFe O and
1
2
2
4
1
Cu Fe O , respectively.
Given that m = η/25 and that, according to (1), the composition of
the solid solution depends on m at a given value of x, the compo-
sition dependence of a can be converted easily to a(η) (Fig. 1)
and vice versa.
0
.5
2.5
4
The composition dependences of the lattice param-
eter calculated by Eq. (2) for the A –B , A –B , and
1
1
2
2
INORGANIC MATERIALS Vol. 41 No. 3 2005

HYDROGEN REDUCTION OF Cu–Mn–Fe–O SPINEL SOLID SOLUTIONS
275
sition triangle. The content of Cu Fe O decreases, Their compositions vary along an isobar in the ranges
0
.5
2.5
4
2
and those of Fe O and CuFeO increase, as represented
C
1
–D
For example, it follows from scheme (3) that, at x =
.5, the amount of CuFeO at the end of the second step
1 2 2 3 3
(x = 0.8), C –D (x = 0.5), and C –D (x = 0.2).
3
4
2
by the scheme
0
2
Cu Mn Fe O + mH
x
1 – x
2
4
2
is 0.456 mol. The complete reduction of this phase
according to scheme (5) yields 0.16Cu0.15Fe2.85O . Dis-
4
=
(1 – m)[(Cu Fe O )
solution of the latter phase in the spinel of composition
2
0
.5
2.5
4
2x – 3m/(1 – m)
(3)
C leads to the formation of a spinel phase of composi-
×
(MnFe O )
(Fe O )
]
2m – x/(1 – m)
2
4
1 – x/(1 – m)
3
4
tion (Cu Fe O ) (MnFe O ) (Fe O ) (D2)
0
.5
2.5 4 0.159
2
4 0.584
3
4 0.257
at the end of the third step (∆m = 0.271, η = 7.6%, η =
2
+
1.5mCuFeO + mH O.
2 2
14.3%). At x = 0.8 and 0.2, η at the end of this step is
2
3 and 5.75%, respectively. The corresponding compo-
At the end of the second step, pO2 ceases to decrease sitions are (Cu Fe O ) (MnFe O ) (Fe O )
0.5
2.5 4 0.284
2
4 0.260
3
4 0.456
(
(
D ) and (Cu Fe O ) (MnFe O ) (Fe O )
D ) (Fig. 2).
3
(
Fig. 1). At x = 1, 0.8, 0.5, and 0.2, this step terminates
1
0.5
2.5 4 0.062
2
4 0.849
3
4 0.089
at η = 15.7, 12.2, 7.6, and 3%, respectively. According
to scheme (3), the corresponding compositions are
Cu0.15Fe O (C), (Cu Fe O ) (MnFe O )
(
(
(
The lattice parameter of the spinel phase decreases
·
·
·
2.85
4
0.5
2.5 4 0.276
2
4 0.389
in the third step of reduction (Fig. 1) owing to the
decrease in the content of MnFe O , the component
with the largest a. The a values calculated for the third
step agree well with experimental data.
Fe O )
(C ), (Cu Fe O ) (MnFe O )
(C ), and (Cu Fe O ) (MnFe O )
(C ) (Fig. 2).
3
4 0.335 1 0.5 2.5 4 0.126 2 4 0.719
2
4
Fe O )
3
4 0.155
2
0.5
2.5 4 0.045
2
4 0.91
Fe O )
3
4 0.045
3
This mechanism of crystal-chemical transforma-
Note that the reduction of CuFeO to Cu and Fe O
tions in the second step is supported by the fact that the assumed by Shchepetkin et al. [5] is thermodynami-
a(η) curves displayed in Fig. 1 coincide to within the cally implausible since at 273 K CuFeO2 has
2 3 4
experimental accuracy with those calculated using
scheme (3) and the equation [10]
log(p /Pa) = –(1.6–1.8) (see Fig. 1 and [12]), while
O
2
Fe O has log(p /Pa) = –7.8 (see Fig. 2 and [7, 13]).
3
4
O
2
Moreover, according to Shchepetkin et al. [5] the equi-
librium phase at the beginning of the reduction of
a(Å) = 8.414C + 8.395C + 8.511(1 – C – C ), (4)
2
3
2
3
CuFeO at constant p is (Cu Fe O ) (MnFe O )
where C and C are the mole fractions of Cu Fe O
4
2
O2
0.5 2.5 4 z
2
4 1 – z
2
3
0.5
2.5
and Fe O , respectively.
with z from 0.11 to 0.67. Actually, the pO2 of these
3
4
In the third step, the CuFeO compound is reduced solid solutions depends on composition and differs
2
to copper metal and a spinel phase of composition from that of CuFeO
by two to three orders of magni-
tude (Fig. 2, lines B'–B' , C'–C' ). According to
2
Cu0.15Fe O [11] according to the scheme
2.85
4
3
3
Shchepetkin et al. [5], during the reduction of CuFeO2
ACuFeO + ∆mH = (A – 1.68∆m)CuFeO
2
2
2
these solid solutions become richer in Fe O , which has
3
4
(5)
+
0.59∆mCu0.15Fe2.85O + 1.59∆mCu + ∆mH O,
an even lower p . This would lead to further strong
4
2
O
2
changes in equilibrium pO2 (Fig. 2), in conflict with the
thermodynamics of the reduction of phases of constant
end of the third step, all of the CuFeO is reduced, i.e., composition ( pO2 = const).
where A is the amount of CuFeO at the end of the sec-
ond step, which can be found using scheme (3). At the
2
2
A – 1.68∆m = 0 and ∆m = A/1.68. Given that ∆η =
In the fourth step of reduction, copper coexists with
a spinel phase, and both pO2 and a decrease (Fig. 1)
2
5∆m, we obtain that at the end of the third step η =
η + 25∆m, where η is the degree of reduction at the
2
2
owing to the reduction of the residual Cu Fe O :
end of the second step. Since pO2 is constant, the com-
0.5
2.5
4
position of the spinel phase may only vary during
reduction along a line of constant pO2 (isobar), equal to
Cu Mn Fe O + mH
2
x
1 – x
2
4
the
p
for CuFeO2 and Cu0.15Fe2.85O , i.e.,
= (1 – 0.25m)[(Cu Fe O )
2x – 1.5m/(1 – 0.25m)
O2
4
0.5
2.5
4
(
6)
log(p /Pa) = –1.7 (an average of the data reported
in [12] and displayed in Fig. 1). In Fig. 2, this line is
labeled C–C . It passes through the representative
points C, C , C , and C of the solid solutions corre-
sponding to the beginning of the third step, which dis-
solve Cu0.15Fe2.85O during the reduction of CuFeO .
O
× (MnFe O )
(Fe O )
]
1.25m – x/(1 – 0.25m)
2
2
4
1 – x/(1 – 0.25m)
3
4
3
+ (x – 0.75m)Cu + mH O.
2
1
2
3
2
At x = 1, the composition of the spinel phase remains unchanged
(C = Cu Fe O ), but its content increases.
4
2
0.15 2.85 4
INORGANIC MATERIALS Vol. 41 No. 3 2005

2
76
E.V. ZINOVIK, M.A. ZINOVIK
a, Å
lattice parameters compared to Fe O and FeO. Near
3
4
the end of this step, the percentage of the spinel phase
[
13]
decreases, and pO2 and a approach those of Fe O .
sp
3
4
4
3
4
4
.42
.38
.34
.30
Finally, the spinel phase disappears, and the percentage
of the wüstite phase reaches a maximum. The crystal-
chemical transformations involved in the reduction of
Cu Mn Fe O solid solutions (0 < x < 1) can be rep-
x
1 – x
2
4
resented by the scheme
Cu Mn Fe O + mH
2
x
1 – x
2
4
=
(1 + x – m)[(MnFe O )
(1 – x) – (3m – 4x)y/(1 + x – m)
2
4
(
7)
0
0.2
0.4
0.6
0.8
1.0
y
× (Fe O )
4
1 – [(1 – x) – (3m – 4x)y/(1 + x – m)]
3
]
+
(3m – 4x)Mn Fe O + xCu + mH O.
Fig. 3. Composition dependence of the lattice parameter for
Mn Fe O solid solutions in equilibrium with a spinel
y
1 – y
2
y
1 – y
phase at 1270 K.
At the end of this step, 1 + x – m = 0, i.e., m = 1 + x.
This occurs at m = 2 (η = 50%), 1.8 (η = 45%), 1.5
The compositions of the spinel phases with x = 1.0, (η = 37.5%), 1.2 (η = 30%), and 1 (η = 25%) for x = 1,
.8, 0.5, and 0.2 vary along the C–E, D –E , D –E , and 0.8, 0.5, 0.2, and 0, respectively.
0
1
1
2
2
D –E lines, respectively, and log p varies along the
3
3
O2
As seen from scheme (7), the composition of the
C'–E', D' –E' , D' –E' , and D' –E' lines (Fig. 2). spinel phase depends on that of the wüstite phase
1
1
2
2
3
3
Mn Fe1 – yO. The latter can be evaluated from the lattice
At the end of this step, Cu Fe O is fully reduced.
y
0
.5
2.5
4
It follows from scheme (6) that this occurs at parameter a (Fig. 1) using the plot of a versus y in
w
w
η = 33.3% for x = 1.0, 26.7% for x = 0.8, 16.7% for Fig. 3, which is well represented by the best fit equation
x = 0.5, and 6.7% for x = 0.2. The corresponding com-
positions are Fe O (E), (MnFe O )
(Fe O )
3 4 0.727
3
4
2
4 0.273
aw = 4.3 + 0.143y.
(8)
(
E1),
(MnFe O ) (Fe O )
(E2),
and
2
4 0.6
3
4 0.4
(
MnFe O ) (Fe O )
(E3).
2
4 0.857
3
4 0.143
Using the plot in Fig. 1 to evaluate a for a given
w
The lattice parameters calculated using Eqs. (4) and
5) agree with experimental data to within the measure-
value of η (within the step in question), one can then
determine y (the composition of the wüstite phase)
from Eq. (8). Next, substituting y in (7), one can find the
composition of the spinel phase corresponding to the
preset η. Comparison of the experimentally determined
lattice parameter (Fig. 1) with the a calculated by
(
ment accuracy.
In the fifth step, copper is in equilibrium with spinel
and wüstite phases of variable composition. In the
course of reduction, pO2 decreases steadily, as do asp
and a , attesting to a decrease in the concentrations of Eq. (4) allows one to assess the accuracy of this proce-
w
MnFe O and MnO, respectively, which have smaller dure. The data for the x = 0.5 solid solution (Table 1)
2
4
Table 1. Composition and lattice parameter of the spinel phase and composition of the wüstite phase in Cu Mn Fe O dur-
0
.5
0.5
2
4
ing the fifth step of reduction (η = 16.7–37.5%)
meas
sp
calc
sp
η, %
Wüstite phase
Spinel phase
a
, Å
a
, Å [Eq. (4)]
16.7
20.0
27.5
35.0
37.5
–
(MnFe O ) (Fe O )
4 0.4
8.463
8.460
8.438
8.403
–
8.465
8.459
8.436
8.405
–
2
4 0.6
3
Mn0.289Fe0.711
O
O
O
(MnFe O )
(Fe O )
2
4 0.549
3
4 0.451
Mn0.266Fe0.734
Mn0.224Fe0.776
Mn Fe O
(MnFe O )
(Fe O )
4 0.612
2
4 0.388
3
(MnFe O ) (Fe O )
4 0.92
2
4 0.08
3
–
0
.2 0.8
INORGANIC MATERIALS Vol. 41 No. 3 2005

HYDROGEN REDUCTION OF Cu–Mn–Fe–O SPINEL SOLID SOLUTIONS
277
validate this approach and lend support to the mecha-
log(p /Pa)
O
2
5
nism represented by scheme (7).3
In the sixth step, a wüstite phase of variable compo-
sition is in equilibrium with Cu and Fe, pO2 decreases,
1
2
3
and a rises. The changes in p and a are related to
w
O2
w
3
4
5
1
the decrease in the content of FeO, which has a higher
equilibrium oxygen pressure and smaller lattice param-
eter in comparison with MnO. In this step, the FeO in
Mn Fe O is reduced to Fe metal according to
–
–
–
1
3
5
y
1 – y
4
the scheme
Cu Mn Fe O + mH
2
x
1 – x
2
4
=
(4 – m)Mn(1 – x)/(4 – m)Fe1 – (1 – x)/(4 – m)
(m – x – 1)Fe + xCu + mH O.
O
(9)
–
–
7
9
6
+
2
4
–1
7
8
9
10 /T, K
This scheme is supported by the fact that the exper-
Fig. 4. Plots of log pO2 vs. inverse temperature for solid solu-
tions in equilibrium with (1–5) CuFeO and (6) Mn Fe1 – yO
imentally determined values of a (Fig. 1) agree with
w
the a calculated by Eq. (8) for the wüstite solid solu-
w
2
y
[
13]: (1) CuFe O , (2) Cu0.8Mn Fe O , (3) Cu0.5Mn Fe O ,
tions with the compositions obtained from (9) for preset
η values.
2 4 0.2 2 4 0.5 2 4
(4)
(CuFe O ) (Cu0.5Fe2.5O ) (MnFe O )
,
2
4 0.3
4 0.3
2
4 0.4
(
5) Cu0.2Mn0.8Fe O , (6) MnFe O .
2
4
2 4
At the end of the sixth step, all of the FeO is reduced
to Fe, i.e., m – x – 1 = 2. This occurs at m = 4 (η =
1
00%), 3.8 (η = 95%), 3.5 (η = 87.5%), 3.2 (η = 80%), under consideration. Using the data in Fig. 4 and the
and 3 (η = 75%) for x = 1, 0.8, 0.5, 0.2, and 0, respec- well-known formula [13]
tively. Hydrogen does not reduce MnO. For this reason,
after the sixth step, Fe and Cu are in equilibrium with
∆
H (kJ/g-at)
0
MnO. The solid solution with x = 1 contains no MnO,
and that with x = 0 contains no Cu, which also follows
from scheme (9).
(
10)
–
3
1/2
=
19.154 × 10 ∂log p /∂(1/T),
O
2
Thus, the proposed reaction schemes make it possi-
ble to quantify the crystal-chemical transformations
involved in the reduction of CuFe O , MnFe O , and
we calculated the partial solution enthalpy of oxygen
Table 2). The results demonstrate that the ∆H of the
(
0
2
4
2
4
Cu-containing solutions (A) with x ≤ 0.8 varies little
their solid solutions.
1
It follows from the data in Fig. 2 that, to obtain sin-
gle-phase CuFe O –MnFe O –Fe O solid solutions,
one must know pO2 as a function of temperature for
2
4
2
4
3
4
Table 2. Composition dependence of ∆H0
each composition. Such data are presented in Fig. 4 for
the low-oxygen phase boundary of some
Composition
–∆H , kJ/g-at
0
(
MnFe O ) (Cu Fe O ) (Fe O ) solid solu-
4
1 – x – x
1 2
CuFe2O4
Cu Mn Fe O
90.0
97.1
2
4
x
0.5 2.5
4
x
3
1
2
tions (A) of practical importance [3, 14].
0
.8
0.2
2
4
4
Cu Mn Fe O
104.7
104.3
106.3
308.2
It is of interest to derive a general expression for
0.5
0.5
2
p (T) applicable to any solid solution in the system (CuFe O ) (Cu Fe O ) (MnFe O )
4 0.4
O
2
4 0.3
0.5 2.5 4 0.3
2
2
Cu Mn Fe O
0
.2
0.8
2
4
3
4
The same refers to the other solid solutions.
MnFe O
2
4
As in the reaction schemes above, the possible variations in oxy-
gen stoichiometry are here left out of consideration because they
are relatively small and cannot influence the character of the crys-
tal-chemical transformations under discussion.
Note: ∆H0 averaged over the Cu-containing compositions is
−98.4 ± 8.4 kJ/g-at.
INORGANIC MATERIALS Vol. 41 No. 3 2005

2
78
E.V. ZINOVIK, M.A. ZINOVIK
with composition. In view of this, log p (1/T) can be pure spinel solid solutions can only be prepared at con-
O
2
trolled p .
O
2
expressed through the average of ∆H , equal to
0
−98.4 kJ/g-at:
REFERENCES
log pO2 = B – 10280/T,
(11)
1
. Jahn, H.A. and Teller, E., Stability of Polyatomic Mole-
cules in Degenerate Electronic States: 1. Orbital Degen-
eracy, Proc. R. Soc. London, A, 1937, vol. 161,
pp. 220−235.
. Zinovik, M.A., Structural Inhomogeneities in Ferrites
and Their Effect on the B–H Loop Shape, Elektron.
Tekh., Ser. 6: Mater., 1974, no. 3, pp. 16–22.
where B depends on composition. It follows from the
data in Fig. 4 that, for solutions A with x ≤ 0.8, log pO
varies almost linearly with composition. The 1270-K
data are well represented by
1
2
2
3
4
. Zinovik, M.A., Shchepetkin, A.A., and Chufarov, G.I.,
USSR Inventor’s Certificate no. 427401, Byull. Izobret.,
log(p /Pa) ± 0.3 = x + 1.9x + 4.7(1 – x – x ). (12)
O
1
2
1
2
2
1
974, no. 17.
. Chufarov, G.I., Men’, A.N., Balakirev, V.F., et al., Ter-
modinamika protsessov vosstanovleniya okislov metal-
lov (Thermodynamics of Reduction of Metal Oxides),
Moscow: Metallurgiya, 1970.
Substituting this relation in (11), we obtain
B = 12.8 – 3.7x – 2.8x .
(13)
1
2
Combining Eqs. (11) and (13), we find log pO2 as a
5. Shchepetkin, A.A., Zinovik, M.A., and Chufarov, G.I.,
Synthesis and Decomposition of Cu Mn Fe O Solid
c
1 – c
2 4
function of composition and temperature for solutions A
Solutions under Equilibrium Conditions, Zh. Neorg.
Khim., 1970, vol. 15, no. 10, pp. 2633–2636.
with x ≤ 0.8:
1
6
. Zinovik, M.A., Shchepetkin, A.A., and Chufarov, G.I.,
Crystal Chemistry of Cu–Mn–Fe–O Spinel Oxides,
Dokl. Akad. Nauk SSSR, 1969, vol. 187, no. 6,
pp. 1304−1307.
log(p /Pa) ± 0.4
O
2
(14)
=
12.8 – 3.7x – 2.8x – 10280/T.
1 2
Relation (14) is of theoretical and practical
interest because it allows one to find, using
simple calculations, the equilibrium oxygen pres-
sure at the low-oxygen phase boundary of
7. Zinovik, M.A., Zalazinskii, A.G., Dubrovina, I.N., et al.,
Crystal-Chemical Transformations during Equilibrium
Reduction of Copper Magnesium Spinel Ferrites, Zh.
Fiz. Khim., 1984, vol. 58, no. 8, pp. 1930–1933.
8
. Zalazinskii, A.G., Balakirev, V.F., Chebotaev, N.M., and
Chufarov, G.I., Crystal-Chemical Transformations dur-
ing Hydrogen Reduction of CuFe O , Izv. Vyssh.
(
MnFe O ) (Cu Fe O ) (CuFe O )
4
1 – x – x
1
solid
2
4
x
0.5
2.5
4
x
2
1
2
2
solutions in broad composition and temperature ranges,
which is vital for controlled synthesis of single-phase
materials in this system.
2
4
Uchebn. Zaved., Tsvetn. Metall., 1970, no. 5, pp. 22–24.
9
. Sapozhnikov, E.Ya., Dovidovich, A.G., Zinovik, M.A.,
et al., Oxidation States of Cations in CuFe O –
2
4
Cu Fe O –Fe O Solid Solutions, Zh. Neorg. Khim.,
0
.5 2.5
4
3
4
CONCLUSIONS
1
981, vol. 26, no. 7, pp. 1751–1754.
Hydrogen reduction of Cu Mn Fe O solid solu- 10. Shchepetkin, A.A., Zinovik, M.A., and Chufarov, G.I.,
x
1 – x
2
4
Property–Composition Relationships in Multicompo-
nent Solid Solutions of Metal Oxides, Dokl. Akad. Nauk
SSSR, 1970, vol. 195, no. 5, pp. 1155–1157.
tions occurs in six steps and involves the following con-
secutive changes in the oxidation states of Cu and Fe:
2
+
+
3+
2+
+
0
Cu
Fe
Cu , Fe
Fe , Cu
Cu , and
2+
0
1
1
1. Zalazinskii, A.G., Balakirev, V.F., Chebotaev, N.M., and
Chufarov, G.I., Equilibrium Composition of Phases
Resulting from Reduction of Cu Fe O , Izv. Akad.
Fe . The final reduction products are Cu, Fe,
and MnO. The oxide with x = 1 contains no MnO, and
that with x = 0 contains no Cu. The observed crystal-
chemical transformations are well described by the pro-
posed reactions schemes.
0
.5 2.5
4
Nauk SSSR, Neorg. Mater., 1970, vol. 6, no. 1,
pp. 162−163.
2. Zalazinskii, A.G., Balakirev, V.F., and Chufarov, G.I.,
Oxygen Pressure–Composition Phase Diagram of the
Cu–Fe–O System at 1000°C, Zh. Fiz. Khim., 1969,
vol. 43, no. 6, pp. 1636–1637.
The reduction of CuFeO2 yields Cu and
Cu0.15Fe2.85O , rather than Cu and Fe O as was
4
3
4
reported earlier. As a result, the Cu content of the spinel
phase increases again, and oxygen is released at con-
1
1
3. Tret’yakov, Yu.D., Termodinamika ferritov (Thermody-
stant p .
namics of Ferrites), Leningrad: Khimiya, 1967, p. 304.
O
2
4. Zinovik, M.A., Effect of Heat Treatment on the Proper-
ties of Copper Manganese Ferrites, Poroshk. Metall.
(Kiev), 1976, no. 3, pp. 69–72.
The equilibrium oxygen pressure varies widely with
composition and temperature. In view of this, phase-
INORGANIC MATERIALS Vol. 41 No. 3 2005