Interparticle interactions in composites of nanoparticles of ferrimagnetic (γ-Fe2O3) and antiferromagnetic (CoO,NiO) materials

Article abstract of DOI:10.1103/PhysRevB.70.134416

The magnetic properties of mixtures of ferrimagnetic γ-Fe ₂O₃ (maghemite) and antiferromagnetic NiO or CoO nanoparticles have been studied by use of ⁵⁷Fe Moessbauer spectroscopy, neutron powder diffraction and magnetization measurements. The studies showed that the interaction with antiferromagnetic particles has a significant influence on the magnetic properties of the γ-Fe ₂O₃ nanoparticles. It was found that mixing the γ-Fe₂O₃ nanoparticles with NiO nanoparticles resulted in a faster superparamagnetic relaxation and a reduced coercivity compared to a sample consisting solely of γ-Fe₂O₃ nanoparticles. On the contrary, mixing of γ-Fe₂O₃ nanoparticles with CoO nanoparticles resulted in a suppression of the relaxation and an increase in coercivity. These results suggest that the properties of the ferrimagnetic γ-Fe₂O₃ nanoparticles are influenced by the anisotropy of their neighboring antiferromagnetic particles. The dominating type of magnetic interaction between the particles in the composites seems to be exchange interaction between surface atoms of neighboring particles.

Full text of DOI:10.1103/PhysRevB.70.134416

PHYSICAL REVIEW B 70, 134416 (2004)
Interparticle interactions in composites of nanoparticles of ferrimagnetic „␥-Fe2O3…
and antiferromagnetic „CoO,NiO… materials
C. Frandsen,1 C. W. Ostenfeld,1 M. Xu,1, C. S. Jacobsen,1 L. Keller,2 K. Lefmann,3 and S. Mørup1
*
1Department of Physics, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark
2Laboratory for Neutron Scattering, ETHZ & PSI, CH-5232 Villigen PSI, Switzerland
3Materials Research Department, Risø National Laboratory, DK-4000 Roskilde, Denmark
(Received 8 March 2004; revised manuscript received 1 July 2004; published 26 October 2004)
The magnetic properties of mixtures of ferrimagnetic ␥-Fe2O3 (maghemite) and antiferromagnetic NiO or
CoO nanoparticles have been studied by use of 57Fe Mössbauer spectroscopy, neutron powder diffraction and
magnetization measurements. The studies showed that the interaction with antiferromagnetic particles has a
significant influence on the magnetic properties of the ␥-Fe2O3 nanoparticles. It was found that mixing the
␥-Fe2O3 nanoparticles with NiO nanoparticles resulted in a faster superparamagnetic relaxation and a reduced
coercivity compared to a sample consisting solely of ␥-Fe2O3 nanoparticles. On the contrary, mixing of
␥-Fe2O3 nanoparticles with CoO nanoparticles resulted in a suppression of the relaxation and an increase in
coercivity. These results suggest that the properties of the ferrimagnetic ␥-Fe2O3 nanoparticles are influenced
by the anisotropy of their neighboring antiferromagnetic particles. The dominating type of magnetic interaction
between the particles in the composites seems to be exchange interaction between surface atoms of neighboring
particles.
DOI: 10.1103/PhysRevB.70.134416
PACS number(s): 75.50.Tt, 75.75.ϩa, 76.80.ϩy, 75.60.Ej
NiO.21,22 Most of the studies of exchange interaction across
interfaces have been focused on thin film structures, because
of their use in spin valves, which play an important role in,
for example, read heads in present day computers. It has
recently been shown that the exchange coupling of ferromag-
netic nanoparticles to an antiferromagnetic environment
can lead to a greatly enhanced anisotropy and this may
be utilized to increase the information density in magnetic
recording media.23 Studies of annealed self-assembled
nanoparticles have revealed strong exchange coupling, which
can be used for creating permanent magnets.24 In other stud-
ies, nanoparticles of ferro- or ferrimagnetic materials have
been mixed with antiferromagnetic nanoparticles by
ball-milling.25–27 When using this preparation technique,
nanoparticles of the different materials may be welded to-
gether and therefore come in close contact such that ex-
change coupling between the particles can be significant. In
these studies, it was found that interaction with the antifer-
romagnetic particles can result in both an enhanced coerciv-
ity and a nonzero exchange bias.
We have earlier investigated the influence of interparticle
interactions between antiferromagnetic nanoparticles of dif-
ferent materials, prepared by drying aqueous suspensions of
the particles.9,10 With this preparation technique one might
expect the interparticle interactions to be weak. However,
Mössbauer studies of pure ␣-Fe2O3 (Refs. 7 and 10) or
57Fe-doped NiO (Ref. 11) nanoparticles have shown that the
superparamagnetic relaxation is significantly suppressed in
samples prepared by drying aqueous suspensions. The stud-
ies of samples of nanoparticles of different materials also
gave some unexpected results.9,10 For example, mixing of
␣-Fe2O3 nanoparticles with CoO nanoparticles resulted in a
strong suppression of the superparamagnetic relaxation,
whereas mixing ␣-Fe2O3 nanoparticles with NiO nanopar-
ticles had the opposite effect. Furthermore, surprisingly it
I. INTRODUCTION
Interparticle interactions between magnetic nanoparticles
can have strong influence on the magnetic properties; in par-
ticular, the superparamagnetic relaxation of the particles can
be significantly affected by interactions.1–18 The influence of
dipolar interactions between ferromagnetic or ferrimagnetic
nanoparticles has been studied in, for example, frozen ferro-
fluids with different interparticle distances. Weak dipole in-
teractions can lead to faster relaxation,12 whereas strong in-
teractions may result in slowing down of the relaxation and
eventually to a divergence of the superparamagnetic relax-
ation time at a critical temperature, which increases with
increasing strength of the interaction.1,2,13–16 Below this criti-
cal temperature the samples have many similarities with
spin-glasses.2–5
Dipolar interactions between antiferromagnetic nanopar-
ticles are typically too weak to have any significant influence
on the relaxation behaviour, even if the particles are in close
contact.6,7 Still, Mössbauer spectroscopy studies have shown
that antiferromagnetic nanoparticles in close proximity often
have their superparamagnetic relaxation suppressed com-
pared to noninteracting particles.7,9–11 This suggests that ex-
change interactions between surface atoms of neighboring
particles can be significant.
Exchange interaction across interfaces between ferro- or
ferrimagnetic materials and antiferromagnetic materials has
great technological importance because it can lead to en-
hanced coercivity and shifted hysteresis loops, known as
“exchange bias,” due to the pinning of the magnetization of
the ferromagnet by the antiferromagnet.19,20 Studies of thin
films of Fe3O4 interlayered with either CoO or NiO have
revealed another interesting phenomenon, namely that the
interactions between the two materials can result in a sub-
stantial increase of the Néel temperature of CoO and
1098-0121/2004/70(13)/134416(7)/$22.50
70 134416-1
©2004 The American Physical Society

FRANDSEN et al.
PHYSICAL REVIEW B 70, 134416 (2004)
E = K sin2 ␪ − M · Bint
͑3͒
was found that mixing with NiO particles resulted in a Morin
transition in 9 nm ␣-Fe2O3 nanoparticles, although this mag-
netic phase transition normally is absent in ␣-Fe2O3 particles
with diameters below ϳ20 nm. In this paper, we present the
results of studies of mixtures of 7 nm ferrimagnetic
␥-Fe2O3 nanoparticles with antiferromagnetic NiO or CoO
nanoparticles. By use of Mössbauer spectroscopy, we have
studied the influence of interactions on the relaxation of the
␥-Fe2O3 nanoparticles. The results are qualitatively similar
to those obtained in our previous studies of mixtures with
␣-Fe2O3 nanoparticles.9,10 Magnetization measurements on
the composites show that interparticle interactions affect also
the coercivity of the ␥-Fe2O3 nanoparticles. This result sup-
ports that there is a strong exchange interaction between the
particles in the samples.
,
where Bint is an effective interaction field, which may have
its main contributions from exchange interactions with
neighboring magnetic particles, and M is the (sublattice)
magnetization. If the second term in Eq. (3) is predominant,
there will be only one energy minimum of the magnetic en-
ergy, and the magnetization vector may then fluctuate around
the direction of the effective interaction field. In this case the
average of the magnetic hyperfine field has a finite value, and
the Mössbauer spectra will therefore be magnetically split.
However, even for very fast fluctuations, the spectra will
have broad lines because of the distribution of effective in-
teraction fields in samples of interacting nanoparticles.7,8,10
Only at high temperatures, where the thermal energy be-
comes comparable to or larger than the interparticle interac-
tion energy, there will be a doublet or singlet component in
the spectra.
II. MÖSSBAUER SPECTRA OF NONINTERACTING
AND INTERACTING MAGNETIC NANOPARTICLES
III. EXPERIMENTAL DETAILS
The magnetic anisotropy of an isolated magnetic nanopar-
ticle is often assumed uniaxial with a magnetic anisotropy
energy density given by
␥-Fe2O3 nanoparticles were prepared by oxidation at am-
bient conditions of Fe3O4 nanoparticles, which were made by
co-precipitation of Fe͑II͒ and Fe͑III͒ from an aqueous solu-
tion of 2.0 M Fe͑NO3͒3 and 1.0 M FeSO4 by adding a 1.0 M
aqueous solution of NaOH. The particles were washed
through several steps with H2O and acetone. After washing,
the particles in a part of the sample were coated with oleic
acid and suspended in heptane in order to minimize interpar-
ticle interactions. The particles in the remaining part of the
sample were left uncoated and freeze-dried.
E = K sin2 ␪,
͑1͒
where K is the magnetic anisotropy energy constant, and ␪ is
the angle between the (sublattice) magnetization direction
and an easy direction of magnetization. Very small particles
may perform superparamagnetic relaxation [i.e., thermal
fluctuations of the (sublattice) magnetization between the
two minima at ␪=0° and ␪=180°] with a relaxation time, ␶,
given by the Néel-Brown expression28,29
NiO nanoparticles were prepared by thermal decomposi-
tion of Ni͑OH͒2 at 325°C in air for 3 h, similar to the prepa-
ration described in Ref. 11.
␶ = ␶ exp͑KV/kBT͒,
͑2͒
CoO nanoparticles were prepared by two different meth-
ods. A sample, called CoO-ann, was prepared by heating
cobalt acetate in an argon atmosphere at 300°C for 4 h. This
sample contained a minor impurity of metallic Co. Another
sample of CoO particles, CoO-bm, was prepared by high-
energy ball-milling of a 1:1 molar ratio mixture of Co3O4
and Co in argon for 115 h. This preparation method seems to
be a convenient way to produce pure and fairly small CoO
particles.
Mixtures of nanoparticles of ␥-Fe2O3 with NiO or CoO
(1:1 by weight, unless otherwise indicated) were prepared
by suspending the particles in distilled water, and subse-
quently exposing them to intense ultrasound with the aim to
break apart agglomerates and to obtain a homogeneous mix-
ture. The mixed samples were left to dry in open petri dishes
in air at room temperature for about 2 days or at 200 °C for
2 h. The dried powders were collected from the petri dishes
with a plastic spatula and packed into sample containers. For
Mössbauer spectroscopy and magnetization measurements,
the powders were tightly packed. In particular for magneti-
zation measurements, where magnetic fields of up to 1 T
were applied, the powders were densely compacted to a co-
herent solid, with the aim to avoid rotation of the particles
during measurements. No binding materials such as epoxy
were added to the powders in order not to affect the inter-
particle interactions.
0
where V is the particle volume, kB is Boltzmann’s constant,
and T is the temperature. ␶ is typically in the range
0
10−11–10−9 s. In nanoparticles, for which ␶ is small
0
͑ϳ10−11 s͒ compared to the time scale of Mössbauer spec-
troscopy, ␶M ϳ5·10−9 s, a typical (i.e., not truly monodis-
perse) particle size distribution will result in a very wide
range of relaxation times at temperatures where the average
relaxation time is close to ␶M. This is due to the exponential
dependence of ␶ on V. Therefore, close to the blocking tem-
perature, only a tiny fraction of the particles will have relax-
ation times close to ␶M, which would give rise to broad com-
ponents in the spectra.10,30 Instead, the spectra will mainly
consist of a superposition of a sextet with narrow lines, rep-
resenting those particles, which are below their blocking
temperature ͑␶ӷ␶ ͒, and a sharp central doublet or singlet,
M
representing those particles which exhibit fast superparamag-
netic relaxation ͑␶Ӷ␶ ͒. However, if ␶ is of the order of
M
0
10−10–10−9 s, a large fraction of the particles in a sample has
relaxation times comparable to the time scale of Mössbauer
spectroscopy in a temperature range where KV/kBT is small.
This will result in spectra with broadened lines around the
blocking temperature.
If the particles are in close proximity, Eq. (1) may be
replaced by7,8,10
134416-2

INTERPARTICLE INTERACTIONS IN COMPOSITES OF…
PHYSICAL REVIEW B 70, 134416 (2004)
FIG. 1. 57Fe-Mössbauer spec-
tra obtained at the indicated tem-
peratures of samples of pure 7 nm
␥-Fe2O3 particles. (a) Coated with
oleic acid and suspended in hep-
tane (b) uncoated, freeze-dried, (c)
uncoated,
dried
at
room
temperature.
X-ray diffraction (XRD) was performed using a PW 1390
Philips diffractometer with a Cu K␣ radiation source.
Transmission electron microscopy (TEM) images were
obtained by a Philips EM 430 operated at voltages up to
300 kV.
sample had an average diameter of 16 nm, whereas the par-
ticles of the CoO-bm sample had a diameter of about 10 nm.
The mean diameter of the NiO particles was found to be
about 5 nm. The results of the XRD analysis were confirmed
by electron microscopy studies. TEM studies further showed
that the ␥-Fe2O3 and CoO particles are pseudo-spherical in
shape, while the NiO particles were plate-shaped with a di-
ameter of about 17 nm and a thickness of about 3 nm.
Neutron powder diffraction data of the two CoO samples,
CoO-ann and CoO-bm showed that the Néel temperatures
of the nanoparticles are very close to the bulk value ͑293 K͒.
This was found from following the decrease in integrated
intensity of the antiferromagnetic ͑½½½͒ reflection. Previ-
ous neutron powder diffraction studies of NiO
nanoparticles,31 which are similar to the NiO particles stud-
ied here, have shown that the Néel temperature of these
plate-shaped nanoparticles is about 60 K lower than the bulk
value ͑523 K͒.
By use of neutron diffraction, we have not been able to
resolve a possible increase in the Néel temperature of the
CoO nanoparticles above the bulk value, when in composites
with iron oxides with higher ordering temperatures, such as
reported for CoO/Fe3O4 multilayer structures.22 For both
␥-Fe2O3+CoO and ␥-Fe2O3+NiO composites, overlapping
reflection lines made it impossible to determine the critical
temperatures of CoO or NiO. Complementary, we have
looked at samples of ␣-Fe2O3+CoO, composites in which
significant interparticle interactions have previously been ob-
served by Mössbauer spectroscopy.9,10 In such composites,
the diffraction lines of the two components can clearly be
resolved and it was seen that the antiferromagnetic reflection
of CoO vanished at a temperature close to 300 K. Thus the
Néel temperature of CoO nanoparticles does not seem to
increase due to the interparticle interactions.
Mössbauer spectra were obtained using conventional con-
stant acceleration spectrometers with sources of 57Co in
rhodium. The instruments were calibrated by use of a
12.5 ␮m foil of ␣-Fe. Velocities and isomer shifts are given
relative to the centroid of the calibration spectrum.
Neutron powder diffraction measurements were per-
formed on the DMC diffractometer at the Swiss Spallation
Neutron Source, Paul Scherrer Institute, Villigen, Switzer-
land. The diffractometer uses a multi-detector, which spans a
two-theta angle of 80° with a detector separation of 0.2°. We
used a wavelength of 4.2 Å for the measurements. By using
a two-theta starting angle of 33°, we were able to record, for
CoO, the antiferromagnetic ͑½½½͒ reflection at 1.28 Å−1,
the magnetic ͑½½3/2͒ reflection at 2.45 Å−1, and the struc-
tural ͑111͒ reflection at 2.56 Å−1.
Magnetic hysteresis loops were measured by use of a vi-
brating sample magnetometer with a superconducting coil
magnet. The samples were enclosed in copper cylinders with
typical sample masses of 125 mg. The instrument was cali-
brated using the known magnetic saturation moment of a
cobalt sample. Measurements were performed at tempera-
tures between 5 K and 310 K in applied fields up to 1 T.
Field cooled hysteresis curves were obtained by cooling in a
field of 1 T over less than 30 min and then recording the
loop starting from 1 T.
IV. RESULTS
A. X-ray and neutron powder diffraction
The average particle diameter was estimated for all
samples from the XRD data by use of the Scherrer formula.
In the analysis we neglected the possible influence of strain
on the line broadening. This may result in an underestimate
of the particle size, especially for the ball-milled sample. The
analysis showed that the ␥-Fe2O3 particles had an average
diameter of about 7 nm. The particles of the CoO-ann
B. Mössbauer spectroscopy
Figure 1 shows Mössbauer spectra of samples consisting
solely of 7 nm ␥-Fe2O3 nanoparticles. The spectra shown in
Fig. 1(a) were obtained from a sample of particles coated
with oleic acid. These spectra show an evolution with tem-
134416-3

FRANDSEN et al.
PHYSICAL REVIEW B 70, 134416 (2004)
FIG. 2. 57Fe-Mössbauer spec-
tra obtained at the indicated tem-
peratures of 7 nm ␥-Fe2O3 par-
ticles (a) mixed with NiO
nanoparticles and dried at room
temperature, (b) mixed with NiO
nanoparticles and dried at 200°C,
and (c) pure ␥-Fe2O3 particles
dried at 200°C.
perature, which is typical for noninteracting or weakly inter-
acting superparamagnetic particles, i.e., a coexistence of a
doublet and a sextet, with the relative area of the doublet
increasing with increasing temperature at the expense of the
sextet. In the spectra, where both a doublet and a sextet are
present (e.g., at 150 K), neither of them have narrow lines. In
spectra of noninteracting nanoparticles of ␣-Fe2O3 the lines
of both the sextet and the doublet are considerably
narrower.32,33 The difference between the two iron oxides
aqueous suspension of the particles to ultrasound followed
by drying at room temperature, are shown in Fig. 2(a). Com-
pared to the spectra of the pure ␥-Fe2O3 particles, prepared
in the same way [Fig. 1(c)], the spectra of the mixture indi-
cate a faster relaxation of the ␥-Fe2O3 particles. This is in
particular evident when comparing the spectra obtained at
150 and 200 K of the two samples. Spectra of mixtures of
␥-Fe2O3 and NiO nanoparticles, dried at 200°C [Fig. 2(b)],
show less influence of relaxation, presumably because of a
stronger interparticle interaction induced by the heating. For
comparison, Fig. 2(c) shows spectra of the pure ␥-Fe2O3
nanoparticles, dried at the same temperature. These spectra
also suggest an enhanced interaction compared to the sample
dried at room temperature, but the spectra of samples con-
taining NiO are much more influenced by relaxation than the
samples of pure ␥-Fe2O3 nanoparticles.
can be explained by the different values of the parameter ␶
0
in the two materials as discussed in Sec. II. A small value of
␶ results in a very wide distribution of relaxation times in
0
the temperature range where the Mössbauer spectrum gradu-
ally transforms from a sextet to a doublet such that most of
the particles have relaxation times that are either much
longer or much shorter than ␶M. This is the case for
␣-Fe2O3 nanoparticles where ␶0ϳ10−11 s.32–34 In 7 nm
Figure 3 shows spectra of a mixture of ␥-Fe2O3 and CoO
(CoO-ann) nanoparticles, prepared by exposing an aqueous
␥-Fe2O3 particles the value of ␶ is of the order of
0
5·10−10 s.35 This means that for small values of the param-
eter KV/kBT the particles will have relaxation times of the
order of 10−9–10−8 s. This results in broad lines in the Möss-
bauer spectra.10,30
The spectra in Fig. 1(b) were obtained from a sample of
uncoated particles, which were freeze-dried. These spectra
are quite similar to those shown in Fig. 1(a), indicating that
the interparticle interactions only play a minor role in this
sample.
Figure 1(c) shows spectra of a sample of uncoated par-
ticles, which was prepared by suspending the freeze-dried
particles in water by exposing them to ultrasound and then
allowing them to dry at room temperature. In these spectra
the superparamagnetic relaxation is to a large extent sup-
pressed at intermediate temperatures. For example, at 150 K
only a sextet with broad lines is visible in contrast to Figs.
1(a) and 1(b) in which an intense doublet is also visible at
this temperature. Such a suppression of the superparamag-
netic relaxation is a typical feature of nanoparticles with a
significant interparticle magnetic interaction.7–11,17 A com-
parison of the spectra in Figs. 1(b) and 1(c) shows that the
way in which the samples are prepared plays a crucial role
for the magnetic properties of nanopowders.
FIG. 3. 57Fe-Mössbauer spectra obtained at the indicated tem-
peratures of 7 nm ␥-Fe2O3 particles mixed with CoO-ace nanopar-
ticles at room temperature.
Mössbauer spectra of a sample consisting of a mixture of
␥-Fe2O3 and NiO nanoparticles, prepared by exposing an
134416-4

INTERPARTICLE INTERACTIONS IN COMPOSITES OF…
PHYSICAL REVIEW B 70, 134416 (2004)
FIG. 4. Hysteresis loops of ␥-Fe2O3, ␥-Fe2O3+CoO-bm and
␥-Fe2O3+NiO at 5 K. The magnetization is given per mass unit of
␥-Fe2O3 in the samples.
FIG. 5. The coercivity of pure ␥-Fe2O3 and ␥-Fe2O3+CoO-bm
nanoparticles plotted as a function of temperature. The solid lines
are guides to the eye.
suspension to ultrasound with subsequent drying at room
temperature. A comparison with the spectra in Figs. 1(c) and
2(a) shows that CoO has the opposite effect of NiO, i.e., it
leads to suppression of the superparamagnetic relaxation in
the ␥-Fe2O3 particles. This is most clearly seen in the spectra
obtained at 150 K.
that the sample of ␥-Fe2O3 nanoparticles interacting with
CoO is not as close to saturation at 1 T as the pure sample of
␥-Fe2O3. This is supported by the different slopes of the
magnetization curves of ␥-Fe2O3 and ␥-Fe2O3+CoO-bm
when approaching 1 T. All three loops are smooth and thus
show no sign of having more than one type of contribution,
i.e., the composites seem to behave magnetically as single-
phase materials indicating that there is a strong exchange
interaction between the particles.24 From the loops, we find
that the 7 nm ␥-Fe2O3 particles have a coercivity, ␮0HC
=55͑±2͒ mT, while the composite of ␥-Fe2O3+NiO has a
slightly lower value, about 43͑±3͒ mT. Quite contrary, for
the ␥-Fe2O3+CoO-bm composite, we find that ␮0HC
=185͑±10͒ mT, which is more than three times as large as it
is for the sample consisting purely of ␥-Fe2O3 nanoparticles
from the same batch. Thus, the influence of the interparticle
interactions on the coercivity can be quite significant.
In Fig. 5, it is shown how the coercivity of the samples of
␥-Fe2O3 and ␥-Fe2O3+CoO-bm changes with temperature.
The difference between the two samples observed at low
temperatures diminishes with increasing temperature and at
Tജ100 K, the coercivity is similar and almost negligible for
the two samples. The fact that the coercivity does not be-
come zero seems to be an effect of the magnetometer.
In order to see if the interaction with the antiferromag-
netic particles could be observed directly as a shifted hyster-
esis loop, i.e., as exchange bias, of the ferrimagnetic
␥-Fe2O3 nanoparticles, we recorded the loops of the pure
␥-Fe2O3 and the composites of ␥-Fe2O3+NiO and ␥-Fe2O3
+CoO-bm at 5 K after cooling in a field of 1 T from 310 K
(i.e., from a temperature above the Néel temperature of CoO
and above the blocking temperature of the NiO nanoparticles
as determined by Mössbauer spectroscopy studies of similar
57Fe-doped NiO nanoparticles10,11). For both the pure
␥-Fe2O3 sample and the composites we observed small loop
shifts, which (in relative values) were nearly identical for the
three samples. Therefore, it seems that the loop shifts are not
related to exchange coupling at the interface between ferri-
magnetic and antiferromagnetic particles, but it is rather an
intrinsic property of the ␥-Fe2O3 nanoparticles. Such a be-
havior of pure ␥-Fe2O3 nanoparticles has been reported by
Martínez et al.36
In order to study the influence of the method for prepara-
tion of CoO and the mixing ratio in samples with ␥-Fe2O3
and CoO nanoparticles, we prepared samples with different
ratios of the two oxides. In this series we used the CoO that
was prepared by ball milling (CoO-bm) and we prepared
samples with 75%, 50%, 25%, and 10% ␥-Fe2O3. The spec-
tra of all these samples did not differ much and they were
very similar to those in Fig. 3. Thus, a relatively small
amount of CoO nanoparticles is sufficient to produce the
interaction effect. Moreover, the different preparation tech-
niques and the different average particle size of the two CoO
samples have little influence on the interaction effects.
Drying mixtures of ␥-Fe2O3 and CoO nanoparticles with
different mixing ratios at 200°C did not result in systematic
variations of the relaxation behavior, but XRD studies of the
samples showed that part of the CoO had oxidized to Co3O4,
which is paramagnetic at temperatures above 80 K. There-
fore, the samples containing CoO dried at room temperature
and at 200°C cannot be compared directly.
C. Magnetization measurements
Figure 4 shows hysteresis loops of the pure ␥-Fe2O3
nanoparticles and of the composites of ␥-Fe2O3+NiO and
␥-Fe2O3+CoO-bm nanoparticles at 5 K after cooling in
zero-field. All three samples are those prepared by ultrasound
treatment of aqueous suspensions followed by drying. The
magnetization at 1 T of the sample of pure ␥-Fe2O3 nano-
particles is about 57 A m2/kg. For each sample, the magne-
tization is shown per unit mass ␥-Fe2O3. It can be seen that
the NiO particles have a nonzero moment (of about
25 A m2/kg, presumably due to uncompensated spins of the
plate-shaped NiO particles), whereas a magnetic moment
from CoO particles could not be resolved at the maximum
applied field of 1 T. Rather, the magnetization of the ␥
-Fe2O3+CoO-bm composite is less than that of the pure ␥
-Fe2O3 nanoparticles. An explanation for the latter could be
134416-5

FRANDSEN et al.
PHYSICAL REVIEW B 70, 134416 (2004)
V. DISCUSSION
sate. However, since the drying method influences the
strength of interparticle interaction [Figs. 1(b) and 1(c)], the
results suggest that during drying of aqueous suspensions the
particles are brought in close proximity in such a way that
the magnetic interaction strength is large. Here, van der
Waals or magnetic forces may play a role.
Our studies of differently prepared samples of ␥-Fe2O3
nanoparticles show that the preparation conditions have a
great influence on the magnetic properties. Mössbauer stud-
ies of an uncoated freeze-dried sample of ␥-Fe2O3 nanopar-
ticles and an uncoated sample dried at room temperature
showed significantly different relaxation behavior. Mixing
␥-Fe2O3 nanoparticles with NiO nanoparticles results in
faster relaxation of ␥-Fe2O3 particles. However, mixing
␥-Fe2O3 nanoparticles with CoO nanoparticles had the oppo-
site effect.
It could be argued that the change of relaxation behavior
in the composites might be due to a change in the magnetic
anisotropy of the maghemite nanoparticles induced by
chemisorbed Co2+ or Ni2+ ions. During the ultrasonic treat-
ment in water, such ions may be dissolved from the CoO or
NiO particles. In fact, because Co2+ ions have large single-
ion anisotropy, chemisorbed Co2+ ions might result in an
enhanced anisotropy, which could explain the suppression of
the relaxation. However, we have prepared samples by dry-
ing ␥-Fe2O3 particles from aqueous solutions of Co2+, and
we have found no effect on the relaxation. Moreover, in
agreement with this, previous studies of ␣-Fe2O3 nanopar-
ticles with chemisorbed Ni2+ or Co2+ ions have also shown
that these ions cannot account for the measured effects in
nanocomposites.10
Dipolar interactions in the nanopowders might play a role
for the magnetic properties of the samples, especially since
both the ␥-Fe2O3 and NiO nanoparticles have been found to
have significant magnetic moments. The energy of the dipole
interaction between two adjacent spherical ␥-Fe2O3 particles,
which are 7 nm in diameter and have a magnetization of
57 A m2/kg, is about 100 K, and for two spherical 5 nm
NiO particles with a magnetization of 25 A m2/kg, the di-
pole interaction energy is about 10 K. These interaction en-
ergies might account for some of the observed effects. For
instance, the increased relaxation of ␥-Fe2O3 nanoparticles
in the composite of ␥-Fe2O3+NiO, compared to the pure
␥-Fe2O3 nanoparticles, could be explained by the reduced
interaction energy, when interaction with NiO instead of
␥-Fe2O3 particles. On the other hand, we have seen how
CoO nanoparticles, which have insignificant external mag-
netic moments, had the most profound influence on the prop-
erties of ␥-Fe2O3 nanoparticles. Therefore, exchange interac-
tions between the particles must be prevalent in the samples.
This is also in accordance with the results obtained on anti-
ferromagnetic composites.10
It is remarkable that exchange interactions across the in-
terfaces of nanoparticles in close proximity in a powder are
strong enough to result in a significant change in the relax-
ation and coercivity. If the particles were separated by layers
of, for example, adsorbed water or if there are mismatches
between the directions of sublattice magnetizations, one
would expect the interparticle exchange interaction to be
small (in insulators, one would not expect exchange coupling
if the magnetic materials are separated by a nonmagnetic
spacer, but this can be the case in metallic systems37). Fur-
ther, if a particle interacts with several neighbors, one might
also expect that the interaction fields would partly compen-
It is interesting that mixing ␥-Fe2O3 particles with NiO
and CoO nanoparticles have the opposite effects on the re-
laxation of ␥-Fe2O3 particles. Similar results were found for
mixtures of ␣-Fe2O3 nanoparticles with NiO and CoO
nanoparticles.9,10 We believe that the observed different in-
fluences of NiO and CoO nanoparticles are mainly a conse-
quence of their different anisotropy energies. The NiO par-
ticles have very small volumes and the magnetic anisotropy
constant of NiO is not very large. Therefore, the NiO par-
ticles are expected to have small magnetic anisotropy energy.
It is possible that the effective anisotropy of the NiO par-
ticles is too small compared to that of the ␥-Fe2O3 particles
to induce exchange coupling effects.38 Mössbauer studies of
a similarly prepared sample of NiO, which was doped with
57Fe, showed that the NiO particles are superparamagnetic
with a blocking temperature of about 120 K.11 When the
␥-Fe2O3 particles are separated by NiO particles, their mag-
netic coupling to other ␥-Fe2O3 particles may therefore be
weakened. The CoO particles are larger than the NiO par-
ticles and they have larger magnetic anisotropy energy con-
stant. Therefore, a ␥-Fe2O3 particle, which is coupled to a
CoO particle, cannot relax in the same way as it could when
it was isolated or only in contact with other ␥-Fe2O3 par-
ticles. It may, to a certain extent, have to follow the fluctua-
tions of the magnetization direction of the CoO particles,
which have slow relaxation because of the large anisotropy
energy. The different morphologies of the NiO and CoO par-
ticles also may lead to differences in the physical contact
with the ␥-Fe2O3 particles, and this can also play a role for
the exact strength of interparticle coupling.
The magnetization measurements indicate, in agreement
with Mössbauer spectroscopy studies, that exchange aniso-
tropy influences the properties of the ␥-Fe2O3 nanoparticles.
The coercivity of the ␥-Fe2O3 nanoparticles was tripled at
low temperatures when interacting with CoO nanoparticles,
while it was slightly reduced when interacting with NiO
nanoparticles. This can be explained by the different
anisotropies of the CoO and NiO particles, which the
␥-Fe2O3 particles apparently couple to. At increasing
temperatures we have seen how the coercivity of
␥-Fe2O3+CoO decreases. This is presumably because the
coupled CoO and ␥-Fe2O3 particles perform superparamag-
netic relaxation. The temperature range where it happens in
the magnetization measurements is somewhat lower than the
temperature range, where the Mössbauer spectra change
from a sextet to a doublet, as expected due to the different
time scales of the two techniques. Although exchange cou-
pling seems prevalent between the particles, we were not
able to observe exchange bias in the composite systems. In
studies of thin films it has also been found that exchange
coupling can lead to an enhanced coercivity which is not
related to the size of exchange bias.39 In the present case, the
absence of exchange bias is probably due to rather small
anisotropies of the nanoparticles on an absolute scale. This
134416-6

INTERPARTICLE INTERACTIONS IN COMPOSITES OF…
PHYSICAL REVIEW B 70, 134416 (2004)
implies that the ␥-Fe2O3 particles can drag the sublattice
magnetization of the antiferromagnets around due to the cou-
pling between the particles. This is seen as an effect to the
coercivity of the ferrimagnetic particles, but the anisotropy
of the antiferromagnetic particles is too small to result in
shifted hysteresis loops.
plained by exchange coupling of neighboring particles in
close contact. We suggest that the different influence of the
NiO and CoO nanoparticles on the relaxation of iron oxide
nanoparticles is related to a difference in magnetic aniso-
tropy of the NiO and CoO particles.
ACKNOWLEDGMENTS
VI. CONCLUSIONS
We thank L. Lilleballe for sample preparations, H. K.
Rasmussen for help with Mössbauer measurements, and F.
Grumsen for TEM micrographs. The Danish Technical Re-
search Council (“Nanomagnetism” framework programme)
and the Danish Natural Sciences Research Council (frame-
work programme 51-00-0363 and DANSCATT) have finan-
cially supported the work. The neutron diffraction measure-
ments were performed at the Swiss Spallation Neutron
Source, Paul Scherrer Institute, Villigen, Switzerland.
The present studies have shown that during drying both
suspensions of nanoparticles of the same material and nano-
particles of different materials, a strong magnetic interpar-
ticle interaction can be established. Such interactions play an
important role for the coercivity and the superparamagnetic
relaxation of the nanoparticles. The results also show that the
way in which the nanopowders are dried can have a decisive
influence on the magnetic properties. The effects can be ex-
Present address: School of Materials and Metallurgy, Northeastern
*
21 J. A. Borchers, R. W. Erwin, S. D. Berry, D. M. Lind, J. F.
Ankner, E. Lochner, K. A. Shaw, and D. Hilton, Phys. Rev. B
51, 8276 (1995).
22 P. J. van der Zaag, Y. Ijiri, J. A. Borchers, L. F. Feiner, R. M.
Wolf, J. M. Gaines, R. W. Erwin, and M. A. Verheijen, Phys.
Rev. Lett. 84, 6102 (2000).
University, Shenyang 110004, China.
1 W. Luo, S. R. Nagel, T. F. Rosenbaum, and R. E. Rosensweig,
Phys. Rev. Lett. 67, 2721 (1991).
2 C. Djurberg, P. Svedlindh, P. Nordblad, M. F. Hansen, F. Bødker,
and S. Mørup, Phys. Rev. Lett. 79, 5154 (1997).
3 H. Mamiya, I. Nakatani, and T. Furubayashi, Phys. Rev. Lett. 80,
177 (1998).
23 V. Skumryev, S. Stoyanov, Y. Zhang, G. Hadjipanayis, D. Givord,
and J. Nogués, Nature (London) 423, 850 (2003).
24 H. Zeng, J. Li, J. P. Liu, Z. L. Wang, and S. Sun, Nature (London)
420, 395 (2002).
4 T. Jonsson, P. Svedlindh, and M. F. Hansen, Phys. Rev. Lett. 81,
3976 (1998).
5 Y. Sun, M. B. Salamon, K. Garnier, and R. S. Averback, Phys.
Rev. Lett. 91, 167206 (2003).
25 J. Sort, J. Nogués, X. Amils, S. Surinach, J. S. Munoz, and M. D.
Baro, Appl. Phys. Lett. 75, 3177 (1999).
6 C. J. W. Koch, M. B. Madsen, S. Mørup, G. Christiansen, and L.
Gerward, Clay Miner. 34, 17 (1986) .
26 J. Sort, J. Nogués, X. Amils, S. Surinach, J. S. Munoz, and M. D.
Baro, J. Magn. Magn. Mater. 219, 53 (2000).
7 M. F. Hansen, C. B. Koch, and S. Mørup, Phys. Rev. B 62, 1124
(2000).
27 T. A. Anhøj, C. S. Jacobsen, and S. Mørup, J. Appl. Phys. 95,
3649 (2004).
8 S. Mørup, M. B. Madsen, J. Franck, J. Villadsen, and C. J. W.
Koch, J. Magn. Magn. Mater. 40, 163 (1983).
9 C. W. Ostenfeld and S. Mørup, Hyperfine Interact. 5, 83 (2002).
10 C. Frandsen and S. Mørup, J. Magn. Magn. Mater. 266, 36
(2003).
28 L. Néel, Ann. Geophys. (C.N.R.S.) 5, 99 (1949).
29 W. F. Brown, Jr., Phys. Rev. 130, 1677 (1963).
30 S. Mørup, C. Frandsen, F. Bødker, S. N. Klausen, K. Lefmann,
P.-A. Lindgård, and M. F. Hansen, Hyperfine Interact. 144/145,
347 (2002).
11 F. Bødker, M. F. Hansen, C. B. Koch, and S. Mørup, J. Magn.
Magn. Mater. 221, 32 (2000).
12 S. Mørup and E. Tronc, Phys. Rev. Lett. 72, 3278 (1994).
13 S. Mørup, Europhys. Lett. 28, 3278 (1994).
14 M. F. Hansen and S. Mørup, J. Magn. Magn. Mater. 184, 262
(1998).
31 S. N. Klausen, P.-A. Lindgård, K. Lefmann, F. Bødker, and S.
Mørup, Phys. Status Solidi A 189, 1039 (2002).
32 F. Bødker and S. Mørup, Europhys. Lett. 52, 217 (2000).
33 F. Bødker, M. F. Hansen, C. B. Koch, K. Lefmann, and S. Mørup,
Phys. Rev. B 61, 6826 (2000).
15 F. Luis, F. Petroff, J. M. Torres, L. M. García, J. Bartolomé, J.
Carrey, and A. Vaurès, Phys. Rev. Lett. 88, 217205 (2002).
16 M. F. Hansen and S. Mørup, Phys. Rev. Lett. 90, 059705 (2003).
17 S. Mørup, F. Bødker, P. V. Hendriksen, and S. Linderoth, Phys.
Rev. B 52, 287 (1995).
34 M. F. Hansen, F. Bødker, S. Mørup, K. Lefmann, K. N. Clausen,
and P.-A. Lindgård, Phys. Rev. Lett. 79, 4910 (1997).
35 T. Jonsson, J. Mattsson, P. Nordblad, and P. Svedlindh, J. Magn.
Magn. Mater. 168, 269 (1997).
36 B. Martínez, X. Obradors, Ll. Balcells, A. Rouanet, and C.
Monty, Phys. Rev. Lett. 80, 181 (1998).
18 D. Fiorani, J. L. Dormann, R. Cherkaoui, E. Tronc, F. Lucari, F.
D’Orazio, L. Spinu, M. Nogues, A. Garcia, and A. M. Testa, J.
Magn. Magn. Mater. 196-197, 143 (1999).
37 N. J. Gökemeijer, T. Ambrose, and C. L. Chien, Phys. Rev. Lett.
79, 4270 (1997).
38 M. S. Lund, W. A. A. Macedo, Kai Liu, J. Nogués, Ivan K.
Schuller, and C. Leighton, Phys. Rev. B 66, 054422 (2002).
39 N. J. Gökemeijer, J. W. Cai, and C. L. Chien, Phys. Rev. B 60,
3033 (1999).
19 J. Nogués and I. K. Schuller, J. Magn. Magn. Mater. 192, 203
(1999).
20 A. E. Berkowitz and K. Takano, J. Magn. Magn. Mater. 200, 552
(1999).
134416-7