Structural and magnetic properties of iron doped ZrO2

Article abstract of DOI:10.1016/j.jallcom.2016.04.170

Zr_1-xFe_xO₂ samples were synthesized by a freeze-drying process, varying the iron concentration from x = 0 to x = 0.40. The solid solutions prepared were structurally and magnetically characterized. The results showed that the samples crystallized with the tetragonal structure of zirconia for low iron concentrations, with the respective cubic structure for intermediate iron concentrations, and that hematite is formed secondarily at the highest doping levels. It was also revealed that the lattice parameter of the zirconia solid solutions decreases almost linearly with increasing dopant concentration. All the monophasic samples are paramagnetic at room and low temperatures, except for the x = 0.25 sample, which revealed an incipient magnetization at 13 K. The fluctuations are antiferromagnetic throughout the temperature range and the exchange interaction was attributed to two mechanisms occurring simultaneously: a direct exchange interaction between nearest neighbors magnetic moments, dominant at the lowest temperatures and an indirect exchange interaction, induced by charge carriers, more effective at the highest temperatures. Both mechanisms are more active for higher iron concentrations.

Full text of DOI:10.1016/j.jallcom.2016.04.170

Journal of Alloys and Compounds 680 (2016) 701e710
Contents lists available at ScienceDirect
Journal of Alloys and Compounds
journal homepage: http://www.elsevier.com/locate/jalcom
Structural and magnetic properties of iron doped ZrO2
a
a
a
a, *
^
ꢀ
Antonio O. de Souza , Flavio F. Ivashita , Valdecir Biondo , Andrea Paesano Jr.
,
Dante H. Mosca b
a
ꢀ
ꢀ
Departamento de Física, Universidade Estadual de Maringa, Av. Colombo, 5790, Zona 7, 87.020-900 Maringa, PR, Brazil
Departamento de Física, Universidade Federal do Parana, Brazil
b
ꢀ
a r t i c l e i n f o
a b s t r a c t
Article history:
Zr1-xFexO2 samples were synthesized by a freeze-drying process, varying the iron concentration from
x ¼ 0 to x ¼ 0.40. The solid solutions prepared were structurally and magnetically characterized. The
results showed that the samples crystallized with the tetragonal structure of zirconia for low iron
concentrations, with the respective cubic structure for intermediate iron concentrations, and that he-
matite is formed secondarily at the highest doping levels. It was also revealed that the lattice parameter
of the zirconia solid solutions decreases almost linearly with increasing dopant concentration. All the
monophasic samples are paramagnetic at room and low temperatures, except for the x ¼ 0.25 sample,
which revealed an incipient magnetization at 13 K. The fluctuations are antiferromagnetic throughout
the temperature range and the exchange interaction was attributed to two mechanisms occurring
simultaneously: a direct exchange interaction between nearest neighbors magnetic moments, dominant
at the lowest temperatures and an indirect exchange interaction, induced by charge carriers, more
effective at the highest temperatures. Both mechanisms are more active for higher iron concentrations.
© 2016 Elsevier B.V. All rights reserved.
Received 14 January 2016
Received in revised form
22 March 2016
Accepted 18 April 2016
Available online 21 April 2016
Keywords:
DMS
DMO
Iron doped ZrO2
Rietveld refinement
Magnetic properties
€
Mossbauer spectroscopy
1. Introduction
are commonly prepared using different chemical routes [10,16,17].
These undesirable phases may not be detected using ordinary X-ray
Diluted magnetic semiconductors (DMS) and diluted magnetic
oxides (DMO) have been intensively investigated within the last
decade [1e6]. II-VI, III-V and IV-VI family semiconductors, as well as
M4þO2-type oxides (e.g., TiO2, ZrO2 and SnO2) [7e10] have been
doped with magnetic transition metals (e.g., Co, Fe and Mn) in the
search for ferromagnetic (FM) order. The driving force of this
research field is the potential applications of such semiconductors
in spintronics [11e13]; a device made with a diluted (ferro)mag-
netic semiconductor could, plausibly, transport a spin polarized
current while still exhibiting all of the basic functions of an undo-
ped matrix. Models and theories have been reported to foresee or
even to explain the FM behavior of such systems [2,14,15]. However,
there is still no general agreement on any of these models and
theories.
diffraction methods. Despite the small size of crystallites and the
low relative amount, minor phases can yield a magnetic response
that is strong enough to result in incorrect interpretations about the
true properties of the doped semiconductor. This situation explains
why many reports in the literature pertaining to FM order for DMS
and DMO must be interpreted with caution. Close attention must
accordingly be paid to the phase identification of the doped sam-
ples before assertively classifying them as DMS or DMO.
We recently developed a novel routine for doping oxide sys-
tems, which was shown to be a very efficient procedure for the
preparation of single phases [18]. Several types of monoxides, di-
oxides and sesquioxides were prepared using different dopants.
Given the determined concentration limits, the samples were un-
doubtedly monophasic after low-temperature thermal annealing.
Here, we report structural and magnetic results of zirconium
dioxide (ZrO2) doped with iron atoms prepared via the routine
noted above. We focus on examining the presence or absence of
magnetism in real solid solutions.
Scientists in this area of research are faced with a complicated
experimental challenge regarding phase formation in samples.
Since researchers are handling diluted systems, small fractions of
(magnetic) secondary phases can be present in the samples, which
Zirconium oxide, or simply zirconia, is a wide-band gap metal
oxide that may stabilize in three crystalline structures depending
on temperature: monoclinic (m-ZrO2), from room temperature (RT)
up to 1170 ꢀC, tetragonal (t-ZrO2), between 1170 ꢀC and 2370 ꢀC, and
* Corresponding author.
E-mail address: paesano@wnet.com.br (A. Paesano).
http://dx.doi.org/10.1016/j.jallcom.2016.04.170
0925-8388/© 2016 Elsevier B.V. All rights reserved.

702
A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
cubic (c-ZrO2), above 2370 ꢀC [19]. In the cubic structure (shown in
Fig. 1a), each cation is at the center of a perfect cube with eight
oxygen atoms at the corners. In the tetrahedral structure (shown in
Fig. 1b), the oxygen cube is distorted in such a way that the first-
and second-nearest neighbors form two non-regular tetrahedra.
For the cubic/tetrahedral structure, the zirconium sub lattice is
formed by regular/non-regular octahedra (i.e., the main axis is
stressed along the c lattice axis) connected by edges, and the basal
plane lies on the x-y plane (i.e., the main axis is parallel to the c
lattice axis). The high-temperature phases, t-ZrO2 and c-ZrO2, can
be stabilized at RT by doping ZrO2 with suitable metal cations
[20,21].
depending on the specific doping process, the dopant concentra-
tion and the temperature of the thermal treatment.
Ferromagnetic order may appear depending on the specific
conditions of the (Zr.Fe)O2 sample preparation. Therefore, it is still
unclear if the system is or is not a DMO. If it is a DMO, it remains to
be understood how this property is connected with the stable
structures of zirconia and their methodology of production.
€
It is also worthwhile noting that Mossbauer spectroscopy (MS)
was used as a characterization technique in some studies in which
iron was used as a dopant [25,26,29]. Indeed, MS is able to identify
iron-containing phases, their magnetic state and their relative
amount. In addition, valence information about the iron chemical
environment (coordination symmetry) in the zirconia matrix may
be extracted from MS data [25,29]. This spectroscopy technique is
very selective to the specific nuclear probe (in this case, 57Fe) and
can detect nanostructured iron precipitates.
Zirconia is an n-type semiconductor with a wide band gap
(5.0e5.5 eV) for all phases [22e24] and was characterized
regarding the structural and magnetic properties after doping with
various transition metals [1,3e5,7,8].
Other authors have carried out similar studies. For example,
Okabayashi et al. characterized Zr1-x(Co,Fe)xO2 and Zr1-xFexO2
prepared via a sol-gel method [25] and observed that dilute Fe and
Co co-doping exhibited FM behavior at RT.
Sahoo et al. reported the presence of RT FM in Zr1-xFexO2 syn-
thesized via combustion (via microwaves) [26].
Garcia et al. focused on a Zr0.9Fe0.1O2 solid solution prepared by a
nitrate/urea combustion route [27]. These authors showed that
using an appropriate amount of urea made it possible to obtain a
completely stabilized Zr0.9Fe0.1O1.95 solid solution without free iron
oxide species.
Besides bulk and nanocrystalline samples, Zippel et al. [28] also
analyzed Mn-doped and undoped ZrO2 thin films grown by pulsed
laser deposition. These authors detected FM related to defects in
these systems.
Here, we extensively used MS for the crystallographic and
magnetic characterizations; we made use of X-ray diffraction (XRD)
and magnetometry techniques for these investigations, respec-
tively. We applied these techniques to very pure bulk samples
prepared by significantly varying the doping concentration. How-
ever, we retained very good control over the phase(s) present in the
samples. All of our results were effectively interconnected, which
makes this study unique.
Our findings allow us to describe the Zr1-xFexO2 system on the
atomic level and demonstrate the absence of any RT magnetic order
in monophasic samples. We estimated the solubility limit for iron
doping in zirconia for samples in an equilibrium state.
2. Experimental details
More recently, Kuryliszyn-Kudelska et al. studied iron-doped
ZrO2 nanocrystals synthesized via the hydrothermal method [29]
and observed that ZrO2 samples doped with 2.8 wt% of Fe were
paramagnetic.
Jiang et al. applied high-energy ball milling of ZrO2 e Fe2O3 for
preparing and studying iron-doped zirconia [30]. These authors
showed that the maximum solubility of iron in c(or t)-ZrO2 was
approximately 18.5 mol%.
Guo et al. found FM order in Fe-doped nanotubes prepared via
anodization of a Zr-Fe alloy [31].
Based on these and other investigations, it is clear that doped
zirconia can stabilize in monoclinic, tetragonal or cubic structures
2.1. Sample preparation
First, commercial salts containing zirconium acetate,
Zr(CH3COO)4 (solution in dilute acetic acid, 15e16% Zr) and iron (II)
acetate, Fe(CH3COO)2, were homogeneously diluted in 80 ml of
distilled-deionised water, in prescribed concentrations. Thereafter
this aqueous mixture was slowly frozen inside a glass flask, which
rotated in contact with a refrigerant fluid (liquid nitrogen). After
that, the flask was connected to a freeze-dryer, which consists of a
vacuum pump and a water trap. During freeze-drying, the frozen
sample was sublimed under low pressure (~250 mm Hg) and tem-
perature (ꢁ58 ꢀC). The process for complete drying of the samples
Fig. 1. Cubic (a) and tetragonal (b) ZrO2 crystalline structures. Blue balls ¼ Oxygen; Green balls ¼ Zirconium. (For interpretation of the references to color in this figure legend, the
reader is referred to the web version of this article.)

A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
703
took approximately 24 h. Finally, all powders were heat treated in
air, first increasing the temperature at a rate of 10 ꢀC/min up to
600 ꢀC, where the samples remained for 3 h. Then, the furnace was
switched off and the samples cooled naturally to RT. Following this
routine, Zr1-xFexO2 powder samples, with x ¼ 0, 0.05, 0.07, 0.09,
0.15, 0.20, 0.25, 0.30, 0.35 and 0.40, were prepared and immediately
characterized.
2.2. Characterization techniques
2.2.1. Scanning electron microscopy
The morphological aspects of doped and undoped materials
were analyzed through the image formed by secondary electrons in
a scanning electron microscopy.
2.2.2. X-ray diffraction
The XRD measurements were taken with an ordinary diffrac-
tometer, operating in the conventional
Brentano). The X-ray radiation used came from a cooper tube (i.e.,
Ka emission lines with
¼ 1.5406 Å) with 40 kV voltage and 30 mA
q-2q geometry (Bragg-
l
filament current. Diffraction patterns were taken in the range of
20ꢀꢂ 2
q
ꢂ 90ꢀ, with 0.02ꢀ angular step and 3 s acquisition time.
Herein, the experimental diffractograms are presented with the
vertical axis expressing the counts in logarithmic scale, because it
shows the smallest peaks and better reveals the minor phases. The
FULLPROF program [32] was applied to refine the c-ZrO2 and t-ZrO2
crystalline structures by the Rietveld method considering the Fm-
3m and P42/nmc space groups, respectively, according to the sym-
metries usually attributed to these phases [33,34]. Pseudo-Voigt
peak-shape functions were used to fit the experimental data. The
data refined were the lattice parameter, the peak shape and ther-
mal parameters.
Fig. 2. SEM images of the microstructure for the x ¼ 0 (a) and x ¼ 0.20 (b) samples.
2.2.3. Magnetization techniques
Specific magnetization (emu/g) versus temperature curves were
measured in the 10 Ke300 K temperature range, with the appli-
cation of a 500 Oe magnetic field. For such measurements, the ZFC
(Zero Field Cooling) and FC (Field Cooling) ordinary procedures
were adopted. Magnetization versus applied magnetic field (kOe)
cycles were also measured at 300 K and 10 K. All the magnetization
measurements were performed using the vibrating sample
magnetometer option in a Physical Property Measurement System
(PPMS e Quantum Design e model Evercool II).
[35e37], in which ZrO2 was obtained in the form of nanowires,
spherical agglomerates or rounded particles (polyhedral). The
thickness of the flakes remains practically the same with the iron
doping, and their aspect ratio can be estimated to be ~0.01.
3.2. X-ray diffraction
Fig. 3 shows diffractograms for some representative samples.
For the x ¼ 0 sample (Fig. 3a) all the peaks belong to the t-ZrO2
phase. Diffractograms for x ¼ 0.05, 0.07 (not shown) reproduce
qualitatively this result. However, a closer inspection of the x ¼ 0.15
diffractogram (Fig. 3b) reveals that the peaks became narrower.
This suggests the appearance of the c-ZrO2 structure, because in the
diffractometric profile of a cubic phase several peaks come from the
collapse of pairs of neighboring peaks existing in the respective
tetragonal phase. Further Rietveld refinements showed by a com-
parison of errors that this is correct.
Thus, from x ¼ 0.15 (Fig. 3c) up to x ¼ 0.25 (Fig. 3d) the cubic
phase was assumed to be present, at least as the very major con-
stituent. Obviously, this means that increasing the iron concentra-
tion up to these levels stabilizes the c-(Zr1-xFex)O2 phase, as
previously pointed out [26].
€
2.2.4. Mossbauer spectroscopy
57
€
Fe Mossbauer transmission spectra were obtained at room and
low temperatures, in a spectrometer operating with a 57Co(Rh)
source moved with constant acceleration. Zr1-xFexO2 powder
samples with about 20 mg cmꢁ2 were used as absorbers. The
equipment was calibrated from the spectrum of an alpha-Fe thin
foil, also measured at RT. The low temperature measurements were
run in a closed cycle He cryostat.
The numerical fits were made considering a Lorentzian line
shape and applying the minimum chi-square method. Each rele-
vant and specific site was represented by an individual sub-
spectrum and a hyperfine magnetic field distribution was used for
the low temperature spectra.
3. Results and discussion
The XRD patterns of the x ꢃ 0.30 samples reveal peaks of a
second phase, initially of low intensity (Fig. 3e), but becoming more
intense as the iron concentration reaches x ¼ 0.40 (Fig. 3f). They
belong to a pattern that can be unequivocally attributed to hema-
tite, thus establishing a solubility limit for iron in c-ZrO2, prepared
by the synthesis method above described. This large limit for the
formation of a second phase in a ZrO2 matrix can be explained by
3.1. Scanning electron microscopy
Fig. 2 shows micrographs of some as-prepared samples and
their morphology of flakes, structured in the sub micrometer scale.
This result is quite different from those reported in literature

704
A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
refinement).
Fig. 5 shows the volume of primitive unitary cell (VC) as a
function of the iron concentration (x). The cell volume decreases
almost linearly with the iron concentration, at a rate of dVC/
dx ~ ꢁ0.06 Å3 and ꢁ0.01 Å3, for the tetragonal and cubic structures,
respectively.
This decrease may be explained based on the Fe3þ cation radius,
smaller than that of the Zr4þ cation, and confirms that iron enters
substitutional to zirconium in either the t-ZrO2 matrix, or the c-
ZrO2 matrix, forming Zr1-xFexO2 solid solutions. Vacancies gener-
ated by the doping (as shown below) may contribute to the volume
decrease as well.
Compared to the data reported in Ref. [39], the present values
for the volume of the tetragonal lattice differ only about 1%, but
present the same rate of change under the dopant concentration
variation. On the other hand, the results obtained by Ref. [40] differ
from the ours not only in magnitude (~3%) but also regarding the
range of stability of the c-Zr1-xFexO2 (mono)phase.
3.3. Magnetization characterization
Fig. 6 shows the curves of magnetization and the respective
reciprocal susceptibility as a function of temperature, for the
x ¼ 0.05, 0.07 and 0.09 samples.
Fig. 3. X-ray experimental diffractograms for the x ¼ 0 (a), 0.09 (b), 0.15 (c) 0.25 (d),
0.30 (e) and 0.40 (f) samples. Miller indexes in (a) and (c) are respective to the
tetragonal and cubic zirconia structures, respectively. The red arrows in (f) indicate the
peaks of the Fe2O3 phase. (For interpretation of the references to color in this figure
legend, the reader is referred to the web version of this article.)
For these concentrations, the magnetization grows mono-
tonically with decreasing temperature. FC and ZFC curves overlap
without evidence of magnetic irreversibility (i.e., a temperature,
Tirr, which indicates the separation of ZFC and FC curves). Therefore,
these samples show a feature typical of paramagnetic (PM) sys-
tems. However, the temperature dependence of reciprocal sus-
ceptibility curves reveals the possible existence of two Curie-Weiss
(C-W) regimes, one at higher temperatures and another at lower
temperatures.
the fact that the ionic radius of Fe3þ (the valence will be given by
€
the Mossbauer results presented later in this paper) (0.92 Å) is
smaller than that of Zr4þ (0.98 Å) in an eight-coordinated neigh-
borhood [38]. Hematite was pointed out as the phase formed from
the exceeding iron, also in samples prepared by other methods
[26,29].
In this sense, the cꢁ1 vs. T curves were fitted, using the
c
¼
c0 þ (T ꢁ
q)/C equation (i.e., a modified Langevin function),
considering different temperature ranges. The temperature-
independent signal c0 was added to the Langevin fitting pro-
cedures in order to allow for magnetic contributions other than the
specific Zr1-xFexO2 solid solution phase, composing the specimen to
be characterized (e.g., the sample holder). The parameters fitted
specifically taking into account separately the T ꢃ 100 K and
T ꢂ 100 K ranges are listed in Table 2 (which also includes data for
the x ¼ 015 and 0.25 samples). For the lowest temperature range, all
Fig. 4 shows refined diffractograms for the x ¼ 0.05 (Fig. 4a) and
x ¼ 0.20 (Fig. 4b) samples. All the refinements were performed
considering only one phase (i.e., tetragonal for 0 ꢂ x ꢂ 0.09 and
cubic for 0.15 ꢂ x ꢂ 0.25) and resulted in lattice parameters (see
Table 1) which are in good agreement with those previously ob-
tained, either for undoped ZrO2 or for iron doped zirconia [39e41].
Small differences observed between these and the present values
may be attributed to different methods of refining the X-ray profiles
(i.e., refinement programs used and parameters fixed initially in the
the C-W coefficients (q
0s) are small and negative, whereas for the
highest temperature range they totalize some tens of negative
Kelvin degrees. All these coefficients reflect AFM fluctuations,
although a change in the magnetic regime is effectively revealed in
going from RT to 10 K, for the three concentrations. This may be
attributed to a variation with temperature of the molecular field
constant,
g
(¼
q/r C) [42], which would weaken with decreasing
temperature, until an “almost ideal” paramagnet. However, in
every case the interactions between the magnetic moments are
weak and are overcame by thermal fluctuations, even at low
temperatures.
On the other hand, c0 presents typical values for frustrated
oxide systems [43e45], increasing slightly with x and the T range,
whereas the specific constant, C, does not show any noteworthy
trend, as the concentration or the T range vary.
In principle, the AFM fluctuations should involve only iron
magnetic moments, stabilized in trivalent state, as revealed by
€
Mossbauer spectroscopy. Considering the concentrations used and
a good dilution of the dopant in the ZrO2 matrix, the nearest ferric
cations are apart by two or three lattice parameters. Therefore, not
close enough for an effective exchange interaction. In other words,
a percolation threshold was not reached by the cations, somehow
Fig. 4. Refined X-ray diffractograms, for the x ¼ 0.05 (a) and x ¼ 0.20 (b) samples.

A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
705
Table 1
Crystalline structure data for the Zr1-xFexO2 phases, obtained from the Rietveld refinements.
Sample (x)
Phase
Lattice
parameter (Å)
Edge length of the oxygen
tetrahedra/cubes (Å)
Edge lengths (el1, el2)
Cation e O (co) bond
length (Å)
Residues (%)
Rwp/Rexp
of the cation octahedraa (Å)
Rwp
Rexp
S
0
t-ZrO2
a ¼ 3.592
c ¼ 5.158
3.3018 (x4)
3.5920 (x2)
3.5920 (x2)
3.9679 (x4)
3.2978 (x4)
3.5900 (x2)
3.5900 (x2)
3.9620 (x4)
3.2900 (x4)
3.5890 (x2)
3.5890 (x2)
3.9488 (x4)
2.5355 (x8)
2.5305 (x8)
2.5280 (x8)
3.5920 (x4)
3.6197 (x8)
2.0828 (x4)
2.3556 (x4)
15.5
6.01
4.45
5.15
2.56
0.05
0.09
t-Zr0.95Fe0.05O2
a ¼ 3.590
c ¼ 5.147
3.5900 (x4)
3.6148 (x8)
2.0808 (x4)
2.3527 (x4)
7.56
1.70
2.00
t-Zr0.91Fe0.09O2
a ¼ 3.589
c ¼ 5.119
3.5890 (x4)
3.6044 (x8)
2.0775
(x4)
2.3470 (x4)
10.3
0.15
0.20
0.25
c-Zr0.85Fe0.15O2
c-Zr0.80Fe0.20O2
c-Zr0.75Fe0.25O2
a ¼ 5.071
a ¼ 5.061
a ¼ 5.056
3.5857 (x12)
3.5787 (x12)
3.5751 (x12)
2.1958 (x8)
2.1915 (x8)
2.1893 (x8)
10.9
12.6
13.7
5.17
6.19
6.17
2.11
2.03
2.22
a
el1 ¼ el2 ¼ el, for the cubic phase.
arranged in the lattice of this p-doped wide gap semiconductor.
Possibly, the Zr3þ cations - having a 3 d1 electronic configuration
- formed in consequence of the presence of “original” oxygen va-
cancies in the undoped matrix, could contribute to the magnetism
of the system [39,46]. However, they are not numerically signifi-
cant, considering the present iron doping levels.
done before for the x ¼ 0.05, 0.07 and 0.09 samples. The curve of
the x ¼ 0.20 sample was not fitted because its behavior is excep-
tional and we do not have yet any explanation yet for that. In
addition, the part of the x ¼ 0.25 curve, respective to the
10 Ke100 K range, was also disregarded since its performance is not
typical for a PM system. Some kind of magnetic order seems to take
place at the lowest temperature range, according to the respective
ZFC curve.
Fitting the x ¼ 0.15 curve, again resulted in a small C-W coeffi-
cient for the lowest temperature range. However, the coefficient
obtained for the highest temperature range is quite large, though of
the same order of magnitude as those reported for some common
oxides [42]. An even higher value for this parameter was found for
the x ¼ 0.25 sample, indicating its dependence on the iron
concentration.
In our opinion, excluding a crystallographic transition at low
temperatures, the change in the magnetic behavior of the studied
samples, under temperature and concentration variations, may be
interpreted based on the simultaneous occurrence of two types of
exchange interactions. One is the superexchange interaction
involving neighboring iron moments. This is more significant for
the x > 0.09 samples, a concentration beyond which the percolation
limit is reached. The other is the indirect exchange interaction,
mediated by the charge carriers in the impurity (conduction) band
and/or holes in the valence band of the Zr1-xFexO2 system. Here, it is
plausible to consider that the acceptor dopant creates a impurity
band because of the relatively high concentration of Fe3þ cations in
the ZrO2 matrix. Indeed, it has been stated that the RKKY mecha-
nism takes place in highly doped DMS and DMO systems
[15,47e49].
The magnetization and respective reciprocal susceptibility as a
function of temperature curves for the x ¼ 0.15, 0.20 and 0.25
samples are shown in Fig. 7. The magnetization still grows up to-
wards the lower temperatures. However, for the x ¼ 0.15 sample
the ZFC and FC curves separate around 200 K. The presence of a Tirr
for this iron concentration indicates a turning point in the magnetic
behavior of the material. Furthermore, for the x ¼ 0.20 and 0.25
samples, which show Tirr ¼ 50 K and 100 K, respectively, it is
noticeable that the ZFC curves saturate and, even, decrease at lower
temperatures. Such a magnetic behavior cannot be found in a
simple paramagnet, but should appear in a material with antifer-
romagnetic interactions or presenting magnetic frustration. In fact,
it is expected that, beyond some doping concentration (say,
x ( 0.15), clusters of iron cations become so numerous that they
may, eventually, behave as spin-glass at low temperatures.
Fits with a modified Curie-Weiss law were also applied to the
reciprocal susceptibility curves of the x ¼ 0.15 and 0.25 samples, as
Hypothetically, this indirect exchange could induce “stronger”
AFM correlations at the highest temperatures, where the charge
carrier concentration is higher, for every doping concentration.
However, thermal fluctuations prevent AFM ordering and all the
systems are non-ordered at RT. Under decreasing temperature, the
concentration of charge carriers in the conduction band would
decrease, weakening the carrier capability of mediating the inter-
action between the diluted (or not so much) iron moments.
Throughout this (low) temperature range, the RKKY interaction
significantly weakens and no significant interaction exists in the
x ꢂ 0.09 samples e which remain close to a PM state (i.e., according
Fig. 5. Primitive cell volume for the Zr1-xFexO2 phases, as a function of the iron con-
centration: t-Zr1-xFexO2 (blue balls)/c-Zr1-xFexO2 (red balls). Empty symbols are values
reported in Refs. [37] and [38] for t and c phases, respectively. The solid lines are the
best linear fits of experimental points. (For interpretation of the references to color in
this figure legend, the reader is referred to the web version of this article.)
to the small q
0s). On the other hand, the superexchange interactions
persist at lower temperatures for the 0.15 ꢂ x ꢂ 0.25 samples,
maintaining the systems with more coupled AFM fluctuations. The
existence of regions with many iron atoms close to each other e i.e.,

706
A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
Fig. 6. Temperature dependence of magnetization, M (left), and reciprocal susceptibility (taken from FC),
c
ꢁ1(T) (right), for the Zr1-xFexO2 samples with 0.05 ꢂ x ꢂ 0.09. Yellow/
green plots are the C-W fits for cꢁ1 vs. T curves, in the T ꢂ 100 K/T ꢃ 100 K temperature range. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
Table 2
Parameters obtained by fitting the
c
(T)ꢁ1 curves with a modified Langevin function.
Sample (x)
0.05
Range (K)
q
co ꢄ 10ꢁ5 (emu/g Oe)
C ꢄ 10ꢁ4 (K emu/g Oe)
T ꢃ 100 K
T ꢂ 100 K
T ꢃ 100 K
T ꢂ 100 K
T ꢃ 100 K
T ꢂ 100 K
T ꢃ 100 K
T ꢂ 100 K
T ꢃ 100 K
T ꢂ 100 K
ꢁ56.1
0.8
1.2
1.9
2.1
2.3
18
8
14
10
29
ꢁ2.8
ꢁ20.2
ꢁ2.6
0.07
0.09
0.15
0.25
ꢁ86.5
ꢁ1.8
2.9
10
4,200
15
23,500
e
ꢁ1,656
ꢁ9.9
12.8
ꢁ54.0
e
ꢁ4.7
ꢁ2,095
e
forming clusters e is predictable. Of course, the magnetic behavior
changes progressively with doping concentration and the coexis-
tence of a paramagnetic fraction with the spin-glass behavior of the
clusters may be expected nearby and above x z 0.15.
overlapped orbitals. Presumably, these are shallow orbitals, placed
right above the top of the valence band. Analogously, the oxygen
vacancies may generate shallow orbitals right below the bottom of
the conduction band. The number of donor and acceptor orbitals is
the same, since for each pair of dopants a vacancy is created. Thus,
magnetic cations and vacancies, sharing the same region in the
zirconia matrix, form a “two-bands” construction inside the gap.
The donors electrons will not stay at the more energetic levels but
will occupy the less energetic (acceptor) levels, in a configuration
with null spin polarization at 0 K (i.e., two antiparallel spins in each
orbital). In this condition, half of the levels is fully occupied and half
is completely empty. This explain why the charge carriers have a
negligible effect on the cation moments orientation, at low tem-
peratures. When the temperature increases, more energetic levels
in the acceptor impurity band become occupied by electrons,
whereas holes may also form in the valence band. Both charge
carriers provide the conductivity for an indirect interaction be-
tween the localized moments. Thus, the BMP model seems to be
It is attractive to analyze the present results under the model of
Bound Magnetic Polarons (BMP’s), as proposed by Coey et al. [15].
However, the situation here seems to be more complex than that
considered by those authors, since the presence of magnetic dop-
ants comes together with (numerous) vacancies, as pointed out
above and established ahead. Like acceptors dopants (i.e., not just
magnetic isovalent dopants, as in Ref. [15]), the iron cations may
trap holes in hydrogenic orbits [50], whereas the vacancies have
donor electrons available to trap. Each iron cation/oxygen vacancy
exerts a proper central force under which the carrier charge (i.e, the
hole/the electron) is bound. Thus, when they are “close” to each
other, say, apart by some lattice parameters, the mutual interaction
between the moment carriers is not mediated by a vacancy (as
assumed in Ref. [15]), but by the holes in the band formed by

A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
707
Fig. 7. Temperature dependence curves of magnetization, M (left), and reciprocal susceptibility (taken from FC),
c
ꢁ1(T) (right), for the Zr1-xFexO2 samples (0.15 ꢂ x ꢂ 0.25). Yellow/
green plots are the C-W fits for cꢁ1 vs. T curves, in the T ꢂ 100 K/T ꢃ 100 K temperature range. (For interpretation of the references to color in this figure legend, the reader is
referred to the web version of this article.)
not applicable to the present situation.
response is accompanied by the AFM fluctuations resulting from
The isothermal magnetization (M) versus applied magnetic field
(H) curves for the Zr1-xFexO2 samples is shown in Fig. 8. As seen in
the measurements performed at 300 K (Fig. 8a), all curves reveal a
paramagnetic-like behavior and no significant magnetic hysteresis.
Thus, it is clearly indicated that the AFM couplings are not of the
long-range type and fluctuations of the magnetic moments pre-
dominate. Qualitatively similar results were previously reported in
the literature by some authors [26,29,40]. The low temperature
M ꢄ H curves (Fig. 8b) reveal a stronger PM moment than at RT, but
it still was not possible to observe any magnetic hysteresis.
Attempts to fit M vs. H curves using a Langevin function, L(a), a
behavior which would be expected for an ideal paramagnet (i.e.,
composed by individual magnetic moments, non-interactive be-
tween themselves), resulted in a poor quality for all the 300 K
measurements, particularly for the higher iron concentration
samples (Fig. 9a). This may be attributed to the “strong” (AFM)
interactions between moments induced by the conduction elec-
trons, numerous enough to indirectly connect magnetic cations. On
the other hand, fits for the 10 K curves are very reasonable, mainly
for the lower iron concentration samples (Fig. 9b). The relevant
magnetic parameters, resultant from the Langevin fits, are listed in
Table 3.
the Fe3þ pairs that are somehow magnetically coupled.
€
3.4. Mossbauer spectroscopy
€
RT Mossbauer spectra of some selected samples are shown in
Fig. 10. For the x ꢂ 0.25 samples the spectra were fitted with one
doublet and for x ꢃ 0.30 with a doublet and a sextet. The hyperfine
parameters for all samples are listed in Table 4.
Despite different theoretical fitting procedures, the present
experimental spectra for iron diluted in zirconia are, in general,
similar to those found for cubic and tetragonal phases [26,40].
The doublet may be associated either to a t-(Zr1-xFex)O2 phase
(for x ꢂ 0.09) or to a c-(Zr1-xFe)O2 phase (for 0.15 ꢂ x < 0.30),
consistent with the X-ray diffraction results. Based on the values of
the isomeric shifts (
(high-spin) and that the quadrupole splittings (
distorted site for the ferric iron, in both ZrO2 matrices. The values
found here for EQ are comparable to the mean values found in
Refs. [25,26,29], which authors considered two doublets for fitting
the spectra of (Zr,Fe)O2 phases.
d) it follows that iron is in the trivalent state
DEQ) show a highly
D
The large quadrupole splitting and, therefore, the non-cubicity
of the iron site may be associated with a certain amount of lattice
disorder caused by replacement of zirconium by iron. This is
particularly striking for the cubic phase, where the iron site should
As general trends, it is observed that MS increases with the iron
concentration, whereas
m shrinks, revealing an increasing degree of
AFM coupling between (neighboring) moments, plausibly two.
Indeed, this suggests a progressive pairing between couples of iron
atoms in the zirconia matrix as long as their number rises, while the
average distance separating them diminishes. In contrary direction,
the major presence of isolated Fe3þ ions in high-spin states (as it
be cubic in a perfect lattice and, hence, DEQ y 0. However, because
of the difference between the valence states of iron (3þ) and zir-
conium (4þ) ions, it follows that the cation replacement generates
vacancies of oxygen atoms in the zirconia matrix. It is plausible to
suppose that for each pair of substitutional iron atoms an oxygen
vacancy is generated to preserve the local electroneutrality in the
lattice. In fact, we believe that substitutional solute atom-vacancy
€
will be seen in the Mossbauer analyzes) for low x contributes to
increasing the individual magnetic moment. Their paramagnetic

708
A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
Fig. 8. Temperature dependence curves of magnetization (M) vs. applied magnetic
field (H) measured at RT (a) and 10 K (b), for the Zr1-xFexO2 samples.
Fig. 9. Langevin fits of the magnetization vs. applied magnetic field curves (1st.
quadrant), at RT (a) and 10 K (b), for the x ¼ 0.15 and 0.09 samples, respectively.
complex formation e formed by two iron atoms (as suggested
above) plus an oxygen vacancy, all of them separated by one bond
length - is a predominant configuration resulting from the iron
doping. Actually, Li et al. previously depicted a similar picture, for a
vacancy-undersized dopant association [51].
Fig. 11 shows the configuration of iron atoms separated by one
edge length (el), with an oxygen vacancy equidistant from them,
the three point defects constituting a triangle. A close oxygen (O1) e
also equidistant from the iron atoms - may be the mediator of the
superexchange interaction, responsible for the AFM fluctuations in
the doped systems. Other configurations forming a similar clus-
tered defect are also plausible considering, e.g., the substitutional
iron atoms in the octahedron basal plane (i.e., also separated by el)
plus an oxygen vacancy placed not symmetrically from them. Very
probably, this clustered defect has more energy relative to that of
Fig. 11, but may exist as fluctuations. This means that fittings
considering a distribution for the quadrupole interaction would
indirect (mediated by carriers) interactions are not able to freeze
the iron moments, at least in time intervals large enough to be
€
“seen” by the Mossbauer spectroscopy. This result reinforces the
earlier assumption of a PM state for the Zr1-xFexO2 system, in spite
of observable AFM correlations.
The x ¼ 0.09 and 0.25 samples were measured also at 13 K and
their spectra are shown in Fig. 12. The spectrum of the former was
fitted with a doublet and the one of the later with a doublet and a
Bhf distribution.
These spectra are also consistent with the magnetometry re-
sults, which showed that at low temperatures the x ¼ 0.09 sample
is PM and the x ¼ 0.25 sample exhibits AFM fluctuations, derived
from the direct exchange interaction (and not because of an indirect
interaction, mediated by charge carriers). Therefore, the hyperfine
Table 3
€
also be plausible for the Mossbauer spectra. Actually, the relatively
Magnetic parameters obtained from Langevin fits of the M vs. H curves, for some of
large linewidth found in the present spectra reveals the existence of
some distribution in the quadrupole splitting. However, using only
two subspectra, as often reported in the literature, is rather arbi-
trary. In summary, whatever the fit procedure, the picture is justi-
fied by the existence of a quadrupole interaction for iron in a site
that otherwise should be cubic, or slightly distorted, in c-Zr1-xFexO2.
It is also worthwhile to note that Zr1-xFexO2 solid solutions did
not show any magnetic hyperfine pattern at RT, whatever the iron
concentration. In other words, possible direct (super-exchange) or
the Zr1-xFexO2 samples measured at 10 K.
Sample (x)
Ms (emu/g)
m (mB)
0.05
0.07
0.09
0.15
0.20
0.25
5.71
7.16
6.72
7.14
6.80
11.42
2.49
2.22
1.61
1.01
0.71
0.94
*M ¼ MS L(a) ¼ MS [coth(a) ꢁ (1/a)]; a ¼
mH/kT.

A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
709
€
Fig. 10. RT Mossbauer spectra for the x ¼ 0.09 (a), 0.25 (b) and 0.30 (c) samples.
€
Fig. 12. 13 K Mossbauer spectra for the x ¼ 0.09 (a) and 0.25 (b) samples.
Table 4
Hyperfine parameters for the Zr1-xFexO2 samples.
nearest neighbor magnetic moments at very low temperatures. In
this sense, Archer et al. [52] calculated theoretically the energy of
magnetic coupling between nearest-neighbors iron atoms (appar-
ently without considering any vacancy presence) and concluded for
a ferromagnetic alignment. Nonetheless, at this time, we do not
have elements to depict assertively the situation and more exper-
iments are needed to make this clear.
Sample (x)
Site
d
(mm/s)
DEQ (mm/s)
Bhf (T)
G (mm/s)
RT
0.05
0.07
0.09
0.15
0.20
0.25
0.30
Doublet
Doublet
Doublet
Doublet
Doublet
Doublet
Doublet
Sextet
0.34
0.34
0.34
0.35
0.35
0.35
0.34
0.36
0.33
0.38
1.03
1.05
1.06
1.11
1.14
1.12
1.12
e
0.67
0.66
0.65
0.77
0.72
0.70
0.69
0.60
0.73
0.42
e
e
e
e
e
e
ꢁ0.21
51.3
e
4. Conclusions
0.40
Doublet
Sextet
1.13
ꢁ0.21
51.7
Zr1-xFexO2 solid solutions were successfully prepared with
flaked-nanostructures, by a chemical routine involving freeze-
drying and thermal treatment.
The x ꢂ 0.0.9 samples crystallized with the tetragonal structure
of zirconia and for 0.15 ꢂ x ꢂ 0.30 with the respective cubic
structure. Throughout all the studied iron concentration range, the
samples were found to be virtually monophasic, and the lattice
parameter of the solid solutions decreases almost linearly with
increasing dopant concentration.
Iron is very well dissolved but constitutes clustered defects
composed by a pair of substitutional iron atoms, separated by an
edge distance, plus an oxygen vacancy one bond length far from
both.
13 K
0.09
0.25
Doublet
Doublet
Bhf Dist.
0.43
0.45
0.46
1.11
1.07
ꢁ0.21
e
0.70
1.34
0.28b
e
32.8a
a
Average value of Bhf Dist.
b
Value fixed in the fitting process.
magnetic field distribution observed for the latter is due to an
insipient spin freezing of a pair of moments, aligning antiferro-
magnetically in a direction determined by the local crystalline field.
This is opposed to the possibility of FM alignment between two
The solubility limit for iron in c-Zr1-xFexO2 for samples prepared
by the synthesis method above pointed out is x z 0.30, beyond
which hematite is formed as a second phase. The hematite fraction
increases with the iron concentration, although up to x ¼ 0.40 the
major phase is always the c-Zr1-xFexO2 solid solution.
Monophasic Fe-doped zirconia with tetragonal and cubic
structures exhibits an overall paramagnetic behavior from 10 K to
300 K, with AFM fluctuations of the magnetic moments throughout
the entire temperature range.
Two mechanisms are responsible for the AFM fluctuations: first,
a superexchange interaction, coupling two close iron cations,
mediated by a common neighbor oxygen (and/or a vacancy). Sec-
ond, a RKKY interaction involving charge carriers in the impurity
conduction and valence bands. The former mechanism is more
significant at lower temperatures and high iron concentrations. The
latter may exist near RT, but is negligible at lower temperatures and
for low-doped samples, due to the small population of charge
carriers.
Fig. 11. Cationic octahedra in the cubic structure, with a pair of iron atoms (red balls)
substituting two nearest neighbors zirconium atoms (green balls), and an oxygen va-
cancy (white ball) placed symmetrically to them. Other oxygens (blue balls) remain
nearby. (For interpretation of the references to colour in this figure legend, the reader
is referred to the web version of this article.)
The concept of a clustered defect based on the hypothesis of
local electronic neutrality may be extended to other doped oxide
systems and is currently under intense investigation in our research

710
A.O. de Souza et al. / Journal of Alloys and Compounds 680 (2016) 701e710
(1999) 2751e2756.
group.
[25] J. Okabayashi, S. Kono, Y. Yamada, K. Nomura, Fabrication and magnetic
properties of Fe and Co co-doped ZrO2, AIP Adv. 1 (2011), 042138e042138-7.
[26] T.R. Sahoo, S.S. Manoharan, S. Kurian, N.S. Gajbhiye, Mossbauer spectroscopic
Acknowledgements
€
study of iron-doped zirconia synthesized by microwave route, Hyperfine
Interact. 188 (2009) 43e49.
[27] F.L. Garcia, V.G. Resende, E. De Grave, A. Peigney, A. Barnabe, C. Laurent, Iron-
stabilized nanocrystalline ZrO2 solid solutions: synthesis by combustion and
thermal stability, Mater. Res. Bull. 44 (2009) 1301e1311.
The authors would like to thank the Brazilian agencies, CAPES/
~
ꢀ
CNPq/Fundaçao Araucaria (Grant PRONEX 17386 #118/2010), and
SisNANO/UFPR for financial support.
[28] J. Zippel, M. Lorenz, A. Setzer, G. Wagner, N. Sobolev, P. Esquinazi,
M. Grundmann, Defect-induced ferromagnetism in undoped and Mn-doped
zirconia thin films, Phys. Rev. B 82 (2010), 125209-125209-5.
[29] I. Kuryliszyn-Kudelska, M. Arciszewska, A. Małolepszy, M. Mazurkiewicz,
L. Stobinski, A. Grabias, M. Kopcewicz, W. Paszkowicz, R. Minikaev,
V. Domukhovski, N. Nedelko, W. Dobrowolski, Influence of Fe doping on
magnetic properties of ZrO2 nanocrystals, J. Alloy. Compnd. 632 (2015)
609e616.
[30] J.Z. Jiang, F.W. Poulsen, S. Mørup, Structure and thermal stability of nano-
structured iron-doped zirconia prepared by high-energy ball milling, J. Mater.
Res. 14 (4) (1999) 1343e1352.
[31] M. Guo, J. Zhao, X. Xu, W. Yu, X. Wang, Preparation of Fe-doped ZrO2 nanotube
arrays by anodization of ZreFe alloy and their magnetic properties, Mater.
Lett. 111 (2013) 93e96.
References
[1] T. Dietl, A ten-year perspective on dilute magnetic semiconductor and oxides,
Nat. Mater. 9 (2010) 965e974.
[2] T. Dietl, H. Ohno, F. Matsukura, J. Cibert, D. Ferrand, Zener model description
of ferromagnetism in zinc-blende magnetic semiconductors, Science 287
(2000) 1019e1022.
ꢀ
[3] A.X. Gray, J. Minar, S. Ueda, P.R. Stone, Y. Yamashita, J. Fujii, J. Braun,
L. Plucinski, C.M. Schneider, G. Panaccione, H. Ebert, O.D. Dubon, K. Kobayashi,
C.S. Fadley, Bulk electronic structure of the dilute magnetic semiconductor
Ga1ꢁxMnxAs through hard X-ray angle-resolved photoemission, Nat. Mater.
11 (2012) 957e962.
[4] K.R. Kittilstved, W.K. Liu, D.R. Gamelin, Electronic structure origins of polarity-
dependent high-TC ferromagnetism in oxide-diluted magnetic semi-
conductors, Nat. Mater. 5 (2006) 291e297.
[5] Y.C. Cao, Impurities enhance semiconductor nanocrystal performance, Science
332 (2011) 48e49.
[6] T. Dietl, H. Ohno, Engineering magnetism in semiconductors, Mater. Today 9
(2006) 18e26.
[7] Y. Matsumoto, M. Murakami, T. Shono, T. Hasegawa, T. Fukumura,
M. Kawasaki, P. Ahmet, T. Chikyow, S.-Y. Koshihara, H. Koinuma, Room-
temperature ferromagnetism in transparent transition metaledoped titanium
dioxide, Science 291 (2001) 854e856.
[8] C.B. Fitzgerald, M. Venkatesan, A.P. Douvalis, S. Huber, J.M.D. Coey, T. Bakas,
SnO2 doped with Mn, Fe or Co: room temperature dilute magnetic semi-
conductors, J. Appl. Phys. 95 (2004) 7390e7392.
[9] Y. Yamada, K. Ueno, T. Fukumura, H.T. Yuan, H. Shimotani, Y. Iwasa, L. Gu,
S. Tsukimoto, Y. Ikuhara, M. Kawasaki, Electrically induced ferromagnetism at
room temperature in cobalt-doped titanium dioxide, Science 332 (2011)
1065e1067.
[10] A. Pucci, G. Clavel, M.-G. Willinger, D. Zitoun, N. Pinna, Transition metal-doped
ZrO2 and HfO2 nanocrystals, J. Phys. Chem. C 113 (2009) 12048e12058.
[11] F. Pan, C. Song, X.J. Liu, Y.C. Yang, F. Zeng, Ferromagnetism and possible
application in spintronics of transition-metal-doped ZnO films, Mater. Sci.
Eng. R. 62 (2008) 1e35.
[32] Rodriguez-Carvajal, J., Short Reference Guide of the Program FULLPROF,
http://www-llb.cea.fr/fullweb/powder.htm.
[33] P. Ghigna, G. Spinolo, U. Anselmi-Tamburini, F. Maglia, M. Dapiaggi, G. Spina,
L. Cianchi, Fe-doped zirconium oxide produced by self-sustained high-tem-
perature synthesis: evidence for an Fe-Zr direct bond, J. Am. Chem. Soc. 121
(1999) 301e307.
[34] W. Daning, G. Yongquan, L. Kaiming, T. Kun, Crystal structure of zirconia by
Rietveld refinement, Sci. China Ser. A 42 (1999) 80e86.
[35] W.-S. Dong, F.-Q. Lin, C.-L. Liu, M.-Y. Li, J. Colloid Interface Sci. 333 (2009)
734e740.
[36] E.S. Elshazly, O.A.A. Abdelal, Nickel stabilized zirconia for SOFCs: synthesis and
characterization, Int. J. Metall. Eng. 1 (6) (2012) 130e134.
ꢀ
ꢀ
[37] A. Gomez, R. Villanueva, D. Viea, S. Murcia-Mascaros, E. Martínez, A. Beltran,
~
ꢀ
F. Sapina, M. Vicent, E. Sanchez, Large scale synthesis of nanostructured zir-
coniabased compounds from freeze-dried precursors, J. Solid State Chem. 197
(2013) 120e127.
[38] R.D. Shannon, Revised effective ionic radii and systematic studies of inter-
atomie distances in halides and chaleogenides, Acta Cryst. A32 (1976)
751e767.
[39] S. Kumar, S. Bhunia, J. Singh, A.K. Ojha, Absence of room temperature ferro-
magnetism in Fe stabilized ZrO2 nanostructures and effect of Fe doping on its
structural, optical and luminescence properties, J. Alloy. Compnd. 649 (2015)
348e356.
[40] G. Stefanic, B. Grzeta, K. Nomura, R. Trojko, S. Music, The influence of thermal
treatment on phase development in ZrO2eFe2O3 and HfO2eFe2O3 systems,
J. Alloy. Compnd. 327 (2001) 151e160.
[41] J. Yu, L.B. Duan, Y.C. Wang, G.H. Rao, Absence of ferromagnetism in Mn- and
Fe-stabilized zirconia nanoparticles, Phys. B 403 (2008) 4264e4268.
[42] B.D. Cullity, C.D. Graham, Introduction to Magnetic Materials, second ed., John
Wiley & Sons, New Jersey, 2009.
[12] H. Ohno,
A
window on the future of spintronics, Nat. Mater.
9
(2010)
952e954.
ꢀ
[13] S.A. Wolf, D.D. Awschalom, R.A. Buhrman, J.M. Daughton, S. von Molnar,
M.L. Roukes, A.Y. Chtchelkanova, D.M. Treger, Spintronics:
a spin-based
electronics vision for the future, Science 294 (2001) 1488e1495.
[14] J.M.D. Coey, K. Wongsaprom, J. Alaria, M. Venkatesan, Charge-transfer ferro-
magnetism in oxide nanoparticles, J. Phys. D. Appl. Phys. 41 (2008),
134012e134012-6.
[15] J.M.D. Coey, M. Venkatesan, C.B. Fitzgerald, Donor impurity band zirconia in
dilute ferromagnetic oxides, Nat. Mater. 4 (2005) 173e179.
[43] A.V. Korolev, G.V. Bazuev, Paramagnetic and spin-glass properties of
pyrochlore-like oxides Ln2Mn2/3Mo4/3O7 (Ln ¼ Sm, Gd, Tb, or Y), Phys. Solid
State 46 (2004) 294e302.
[16] S.-I. Park, G.Y. Ahn, C.S. Kim, Interpretation of ferromagnetic Fe doped ZnO by
[44] N.P. Raju, E. Gmelin, R.K. Kremer, Magnetic-susceptibility and specific-heat
studies of spin-glass-like ordering in the pyrochlore compounds R2Mo207
(R ¼Y, Sm, or Gd), Phys. Ver. B 46 (1992) 5405e5411.
€
Mossbauer spectroscopy, J. Appl. Phys. 101 (2007), 09H113e09H113-3.
[17] F.H. Alhassan, U. Rashid, M.S. Al-Qubaisi, A. Rasedee, Y.H. Taufiq-Yap, The
effect of sulfate contents on the surface properties of irone manganese doped
sulfated zirconia catalysts, Powder Technol. 253 (2014) 809e813.
[18] Unpublished results.
[45] T.F. Qi, O.B. Korneta, X. Wan, L.E. De Long, P. Schlottmann, G. Cao, Strong
magnetic instability in correlated metallic Bi2Ir2O7, J. Phys. Condens. Matter 24
(2012) 345601e345606.
ꢀ
[19] J.A. Rodríguez, M. Fernandez-García, Synthesis, Properties, and Applications of
[46] G. Bouzerar, T. Ziman, Model for vacancy-induced d0 ferromagnetism in oxide
compounds, Phys. Rev. Lett. 96 (2006), 207602e207602-4.
[47] A.M. Werpachowska, Z. Wilamowski, The RKKY coupling in diluted magnetic
semiconductors, Mater. Pol. 24 (3) (2006) 1e6.
[48] C. Xia, C. Hu, Y. Tian, P. Chen, B. Wan, J. Xu, Room- temperature ferromagnetic
properties of Fe-doped ZnO rod arrays, Solid State Sci. 13 (2011) 388e393.
[49] F. Matsukura, H. Ohno, A. Shen, Y. Sugawara, Transport properties and origin
of ferromagnetism in GaMnAs, Phys. Rev. B (1998) R2037eR2040.
[50] N.W. Aschrift, N.D. Mermin, Solid State Physics, Harcourt College Publishers,
New York, 1976.
[51] P. Li, I.-W. Chen, J.E. Penner-Hahn, Effect of dopants on zirconia stabilization-
an X-ray absorption study: I, trivalent dopants, J. Am. Ceram. Soc. 77 (1994)
118e128.
[52] T. Archer, C. Das Pemmaraju, S. Sanvito, Magnetic properties of ZrO2-diluted
magnetic semiconductors, J. Magn. Magn. Mater. 316 (2007) 188e190.
Oxide Nanomaterials, John Wiley & Sons, Hoboken, 2007.
[20] V.G. Myagkov, L.E. Bykova, O.A. Bayukov, V.S. Zhigalov, I.A. Tambasov,
S.M. Zharkov, A.A. Matsynin, G.N. Bondarenko, Solid state synthesis and
characterization of FeeZrO2 ferromagnetic nanocomposite thin films, J. Alloy.
Compd. 636 (2015) 223e228.
[21] R.E. Hummel, Electronic Properties of Materials, fourth ed., Springer, New
York, 2010.
[22] S.G. Botta, J.A. Navío, M.C. Hidalgo, G.M. Restrepo, M.I. Litter, Photocatalytic
properties of ZrO2 and Fe/ZrO2 semiconductors prepared by a solegel tech-
nique, J. Photoch. Photobio. A Chem. 129 (1999) 89e99.
[23] V. Milman, A. Perlov, K. Refson, S.J. Clark, J.G.B. Winkler, Structural, electronic
and vibrational properties of tetragonal zirconia under pressure: a density
functional theory study, J. Phys. Condens. Matter 21 (2009), 485404e485404-
12.
[24] A.P. Bechepeche, O. Treu Jr., E. Longo, C.O. Paiva-Santos, J.A. Varela, Experi-
mental and theoretical aspects of the stabilization of zirconia, J. Mater. Sci. 34