Large-scale synthesis and magnetic properties of cubic CoO nanoparticles

Article abstract of DOI:10.1016/j.jpcs.2012.01.001

We present the synthesis, microstructural and magnetic characterization of cubic CoO nanoparticles with well-controlled size and shape. The as-synthesized CoO nanoparticles are stable because of the organic coating that occurred in situ. The Néel temperature is 225 and 280 K for the 42 and 74 nm CoO particles, respectively. The CoO nanoparticles exhibit anomalous magnetic properties, such as large moments, coercivities and loop shifts. These results provide evidence for the formation of spin compensated random system in CoO. The structurally distorted and magnetically disordered surface layer ferromagnetic phase played an important role in the magnetic behavior of CoO nanoparticles. The smaller is the particle size, the stronger is the contribution of the ferromagnetic phase and the more is the surface layer helpful to enhance the observed coercivity and the exchange bias.

Full text of DOI:10.1016/j.jpcs.2012.01.001

Journal of Physics and Chemistry of Solids 73 (2012) 646–650
Contents lists available at SciVerse ScienceDirect
Journal of Physics and Chemistry of Solids
journal homepage: www.elsevier.com/locate/jpcs
Large-scale synthesis and magnetic properties of cubic CoO nanoparticles
n
Huigang Shi , Xuemin He
Key Laboratory for Magnetism and Magnetic Materials of the Ministry of Education, Lanzhou University, Lanzhou 730000, People’s Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history:
We present the synthesis, microstructural and magnetic characterization of cubic CoO nanoparticles
with well-controlled size and shape. The as-synthesized CoO nanoparticles are stable because of the
organic coating that occurred in situ. The N e´ el temperature is 225 and 280 K for the 42 and 74 nm CoO
particles, respectively. The CoO nanoparticles exhibit anomalous magnetic properties, such as large
moments, coercivities and loop shifts. These results provide evidence for the formation of spin
compensated random system in CoO. The structurally distorted and magnetically disordered surface
layer ferromagnetic phase played an important role in the magnetic behavior of CoO nanoparticles. The
smaller is the particle size, the stronger is the contribution of the ferromagnetic phase and the more is
the surface layer helpful to enhance the observed coercivity and the exchange bias.
Received 10 June 2011
Received in revised form
8
December 2011
Accepted 2 January 2012
Available online 11 January 2012
Keywords:
A. Oxides
B. Chemical synthesis
C. Electron microscopy
D. Magnetic properties
Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved.
1
. Introduction
In 1961 N e´ el suggested that fine particles of an antiferromag-
material to another. In a random magnetic system, this core–shell
model of the vacancy concentration perhaps can be used to explain
the magnetic behavior of some AFN [9].
netic material should exhibit magnetic properties such as super-
paramagnetism and weak ferromagnetism [1]. N e´ el attributed the
permanent magnetic moment to an uncompensated number of
spins on two sublattices. Naturally, the magnetic behavior of
antiferromagnetic nanoparticles (AFN) was widely studied in the
next half a century [2–5]. Indeed, anomalous magnetic properties
in AFN have been observed; however, the interpretation of the
magnetic source is controversial. Based on the reports of refer-
ences, there exist several mechanisms which determine the
magnetic behavior of AFN: (i) a size effect which substantially
lowers the N e´ el point [6], (ii) the appearance of uncompensated
spins on the particle surface [7,8], (iii) the creation of a spin-glass
structure [9,10], (iv) the formation of the complicated magnetic
structure caused by the canting of the spins and increasing
number of magnetic sublattices [11–13] and (v) the existence of
the random magnetic disorder caused by other factors such as
enhanced grain surface effect [14], and magnetically disordered
shell [15]. Thus, the non-intrinsic property of most AFN makes the
study of magnetic properties very uncertain.
CoO, crystallizing in the rock salt structure, is antiferromag-
netic (T
N
ꢀ298 K) and electrically insulating [19]. Although there
are many reports in the literature on the preparation of CoO
nanoparticles, conflicting magnetic properties have been observed
[20–23]. To get closer to the intrinsic magnetic properties, herein
we present the synthesis, microstructural and magnetic character-
ization of cubic CoO nanoparticles with well-controlled size and
shape. Taking into account the notion of a spin compensated
random system, the anomalous magnetic properties exhibited by
CoO nanoparticles are analyzed in detail. In the present article we
focus on the influence of the particle size on the N e´ el temperature,
magnetization, coercivity and exchange bias.
2
. Experimental
2
.1. Sample preparation
In a typical synthesis process, cobalt(III) acetylacetonate [Co
(
3
acac) ] (0.6 g, 1.68 mmol) and oleylamine (11.1 ml, 33.6 mmol)
Further, the ferromagnetism was also observed on Zn0.85Fe0.15
O
were mixed and stirred in a three-necked flask. The mixture was
heated at 140 1C (oil bath temperature) for 20 min under an argon
atmosphere, resulting in a red solution. Following this dissolution,
diluted magnetic semiconductor [16], inorganic compound [17]
and metal-oxide [18] nanoparticles. Rao and co-workers [17,18]
have demonstrated that the ferromagnetism of the non-magnetic
oxides nanoparticles could arise from surface defects, and the
nature of defects (cation or anion vacancies) could vary from one
ꢁ
1
the red solution was heated up to 215 1C at a rate of 7 1C min
,
and was kept at this temperature for 1 h. Then, the reaction
mixture was cooled down to room temperature and a brown
colloidal solution was obtained. The colloidal nanoparticles were
separated upon the addition of ethanol and hexane, centrifuged,
and washed using a mixture of ethanol and toluene solvent. At the
n
Corresponding author. Tel.: þ86 931 8912688; fax: þ86 931 8914160.
E-mail address: shihuig@lzu.edu.cn (H. Shi).
0022-3697/$ - see front matter Crown Copyright & 2012 Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpcs.2012.01.001

H. Shi, X. He / Journal of Physics and Chemistry of Solids 73 (2012) 646–650
647
end, the brown products (42 nm quasi-cubic CoO nanoparticles)
were dried in an oven at room temperature, overnight. Similarly,
the shape and size of CoO nanoparticles can be controlled by
changing the precursor concentration. A solution of 1:10 M ratio of
42 nm and 74 nm particles, which are not in complete agreement
˚
with the known value for bulk CoO (a¼4.26 A).
Fig. 2 shows one typical FTIR spectrum of CoO nanoparticles
with an average diameter of 74 nm. The peaks at 2852.5 and
ꢁ
1
Co(acac)
3
and oleylamine was employed to prepare 74 nm sphe-
2922.9 cm
are due to the symmetric and asymmetric CH
2
rical CoO nanoparticles. Using FTIR we found that a layer of
oleylamine remained on the particles as a protective layer.
2.2. Characterization
Powder X-ray diffraction (XRD) patterns were obtained with a
Philips X’pert diffractometer equipped with a rotating anode and
Cu K radiation (
FTIR) spectra were recorded from 400 to 4000 cm
a
l¼0.15418 nm). Fourier transform infrared
ꢁ1
(
on Nexus
6
70 spectrophotometer in KBr pellets. The particle size and
morphology were examined by a Hitachi S-4800 scanning elec-
tron microscope (SEM) with an accelerating voltage of 5 kV and a
Tecnai G2 F30 transmission electron microscope (TEM). Further,
high-resolution TEM (HRTEM), selected-area electron diffraction
(SAED) and energy-dispersive X-ray (EDX) analysis were also
conducted with a Tecnai G2 F30 operating at 300 kV. The
magnetic measurements were performed on a MPMS-XL super-
conducting quantum interference device (SQUID) magnetometer.
3
. Results and discussion
Fig. 2. FTIR spectrum of spherical CoO nanoparticles with an average diameter of
4 nm.
7
The XRD patterns of the CoO samples, provided in Fig. 1,
clearly show pure nanocrystalline cubic phase. All five diffraction
peaks can be assigned undisputably to (111), (200), (220), (311)
and (222) lattice planes, which are in good agreement with those
of the bulk CoO (JCPDS No. 74-2392). There is no evidence for
parasitic phases like Co
CoO was calculated from the major diffraction peak of the base of
200) using the Scherrer formula:
3 4
O or metallic Co. The particle size D of
(
D ¼ 0:89 cos
where is the X-ray wavelength,
maximum (FWHM) of the peak,
from Eq. (1), the average sizes of the CoO particles obtained under
/20 and 1/10 precursor concentration are 42 and 74 nm, respec-
l=b
y
ð1Þ
l
b
is the full width at half
y
is the Bragg angle. Evaluated
1
tively. This result is in agreement with the SEM observations which
show average sizes of 41.4 and 73.1 nm (Fig. 3). Further, the lattice
˚
parameter a was found to be 4.2512(4) and 4.2640(9) A for the
Fig. 1. XRD patterns of (a) 42 nm quasi-cubic and (b) 74 nm spherical CoO
Fig. 3. SEM images of the as-synthesized CoO particles obtained under (a) 1:20
nanoparticles.
and (b) 1:10 precursor concentrations.

6
48
H. Shi, X. He / Journal of Physics and Chemistry of Solids 73 (2012) 646–650
ꢁ
1
˚
stretching modes, the peak at 3005.8 cm
is due to the n(C–H)
shown in Fig. 4c. The lattice spacing of 2.45 A in the image
mode of the C–H bond adjacent to the C¼C bond, and the small
corresponds to the interplanar separation between (111) lattice
planes. As can be seen in the SAED pattern for a collection of
particles in the inset of Fig. 4c, pure cubic symmetry was clearly
identified for the CoO nanoparticles. In the EDX spectrum
(Fig. 4d), apart from the copper and carbon signals from the
carbon-coated copper grid, the average composition was found to
be Co: 78.71 wt% (50.10 at%) and O: 21.28 wt% (49.89 at%), which
is close to the stoichiometric composition of CoO. This indicates
that the obtained nanoparticles are chemically CoO.
Fig. 5 shows 300 Oe field-cooled (FC) and zero-field-cooled (ZFC)
curves of the CoO nanoparticles with various sizes. At temperatures
in the 40–320 K range, the ZFC magnetization has a maximum.
In the context of magnetic nanoparticles, this could correspond to a
ꢁ
1
peak at 1643.2 cm
is due to the n(C¼C) stretching mode.
ꢁ1
In addition, the peak at 1550.0 cm is due to the NH
2
scissoring
mode. The results clearly reveal that the nanoparticles are
coated with oleylamine, which prevents them from being further
oxidized in air.
Fig. 3 shows the representative SEM images of the CoO
3
particles. As the molar ratio of Co(acac) and oleylamine reduced
from 1:10 to 1:20, the configuration of CoO particles changes
from spherical shape (Fig. 3b) to quasi-cubic shape (Fig. 3a). The
average sizes of the two CoO samples were found to be about 41.4
and 73.1 nm. Simultaneously, with the increase of the particle
size, the uniformity of the size distribution deteriorates. It is
reasonable to deduce that the different concentration of oleyla-
mine will cause a different density of organic functional groups on
the particle surface, which may be the direct reason for easy
control over the particle size and size distribution during the
growth process.
N e´ el temperature T
N B
or to a blocking temperature T . The former is
associated to the intraparticle antiferromagnetic ordering. The latter
is associated to the temperature below which the magnetic
moment of the nanoparticles is not able to across the anisotropy
energy barrier. Based on the powder neutron-diffraction measure-
ments, Silva et al. [15] have suggested that the peaks of the ZFC
Typical TEM images of the CoO nanoparticles are shown in
Fig. 4a–b. When a solution of 1:20 M ratio of Co(acac)
oleylamine was employed, 42 nm quasi-cubic CoO nanoparticles
Fig. 4a) were formed. However, 74 nm spherical CoO nanoparti-
cles (Fig. 4b) were obtained when a solution of 1:10 M ratio was
used. One typical HRTEM image of an isolated CoO nanoparticle is
3
and
magnetization are associated with T
for the 42 and 74 nm CoO particles, respectively. Obviously, the size
dependence of T demonstrates that in the CoO nanoparticles, a size
effect is relevant and contributes to the reduction in T . According
to the random magnetic system [6,14,24], we expect that T may
N N
. Thus, the T is 225 and 280 K
(
N
N
N
Fig. 4. TEM images of the CoO nanoparticles formed at (a) 1:20 and (b) 1:10 precursor concentrations. (c) HRTEM lattice fringe image and (d) EDX spectrum of CoO
nanoparticles. Inset in (c): SAED pattern of a collection of CoO nanoparticles.

H. Shi, X. He / Journal of Physics and Chemistry of Solids 73 (2012) 646–650
649
caused by the distortion of the structure and the canting of the
spins. The loop (a) shows more curvature than loop (b), indicating
that there is a stronger ferromagnetic interaction in 42 nm CoO
particles than that in 74 nm CoO particles [8]. Clearly, the existence
of a structurally distorted and magnetically disordered surface
layer opens the question about the creation of this random
magnetic system [15]. One magnetic source is the symmetry break
of the exchange oaths of surface atoms, which gives rise to a
progressively canting of the moments [25]. Another magnetic
source for the generation of a pinned moment was discussed for
NiO [12,13]. Where the reduced coordination of surface spins leads
to different sublattice configurations, this causes changes in the
magnetic order throughout the particle. This work of Kodama et al.
[12,13] is compatible with the present experimental results.
The enlargement of the hysteresis loop is shown in the inset of
Fig. 6. It is evident that the loops are strongly displaced from the
origin. Apart from the coercivity H , there also exists the exchange
bias H . As generally accepted [10], the H
C
E
C
and H
E
are defined as
ꢀ
ꢀ
ꢀ
ꢀ
H
C
¼ ðjHCꢁjþ HCþ Þ=2,
H
E
¼ ꢁðHCꢁ þHCþ Þ=2
ð2Þ
where HCꢁ and HCþ are the negative and the positive coercive
fields, respectively. Based on Eq. (2), the estimated values of H
C
and
H
E
are 139 and 24 Oe for the 42 nm CoO particles; 70 and 10 Oe for
Fig. 5. FC–ZFC magnetization curves of the CoO particles with the sizes of (a) 42
the 74 nm CoO particles, respectively. Herein the exchange bias
needs two phases, one spin compensated and one with a ferro-
magnetic component. It is reasonable to deduce that the contribu-
tion of the surface layer ferromagnetic phase for 42 nm CoO
particles is large, and for 74 nm CoO particles is small. The
relatively weak coupling between the two magnetic phases allows
a variety of reversal paths for the spins upon cycling the applied
field, resulting in large coercivities and loop shifts. Therefore, the
smaller is the particle size, the stronger is the contribution of the
ferromagnetic phase and the more is the surface layer helpful to
enhance the observed coercivity and the exchange bias.
and (b) 74 nm.
4. Conclusions
Cubic CoO nanoparticles have been prepared by the thermal
decomposition of Co(acac)
CoO particles, their anomalous magnetic behavior includes
i) irreversibility of ZFC and FC magnetizations below the bifurca-
tion temperature; (ii) shift of antiferromagnetic transition
towards lower temperature (typical T values are 225 and
80 K); (iii) creation of hysteresis at 5 K (specific H values are
39 and 70 Oe) and (iv) appearance of loop shift (obvious H
3
in oleylamine. For the 42 and 74 nm
(
N
2
1
C
Fig. 6. Hysteresis loops of the CoO nanoparticles with the sizes of (a) 42 and
E
(b) 74 nm at 5 K. Inset: an enlargement of the region at low magnetic field.
values are 24 and 10 Oe). Further, the present results provide
evidence for the existence of random magnetic system in CoO
particles. The structurally distorted and magnetically disordered
surface layer ferromagnetic phase played an important role in the
anomalous magnetic properties of CoO nanoparticles.
also be affected by the magnetically disordered component. Simul-
taneously, the magnetization begins to increase greatly at low
temperatures; this might arise from the increase in surface ferro-
magnetic phase owing to the reduction of the particle size. Further,
a very distinct splitting of the ZEC and FC magnetization curves is
Acknowledgments
N
observed as the temperature is below the T . This behavior is a
generic feature of disordered system. Considering our samples,
which is the spin compensated random system, the splitting can
be thought of as due to slow relaxation and competing interaction
among these different magnetic components.
The authors thank Kuo Qi for help with the TEM measure-
ments. This work was supported by the National Natural Science
Foundation of China (Grant no. 50801033).
Fig. 6 shows typical variation of the magnetization (M) with the
magnetic field (H) at 5 K for 42 and 74 nm CoO particles. Both
samples show hysteresis behavior at 5 K and show no tendency to
saturate even at 10 kOe, indicating the presence of strong compet-
ing ferromagnetic and antiferromagnetic interactions in the system
References
[
[
1] L. N e´ el, Low Temperature Physics, in: C. Dewitt, B. Dreyfus, P.D. de Gennes
(Eds.), Gordon and Beach, New York, 1962, p. 413.
2] A. L o´ pez-Ortega, D. Tobia, E. Winkler, I.V. Golosovsky, G. Salazar-Alvarez,
S. Estrad e´ , M. Estrader, J. Sort, M.A. Gonz a´ lez, S. Suri n˜ ach, J. Arbiol, F. Peir o´ ,
R.D. Zysler, M.D. Bar o´ , J. Nogu e´ s, J. Am. Chem. Soc. 132 (2010) 9398–9407.
[
16]. More specifically, the linear part of loop is mainly due to the
antiferromagnetic volume phase; the curved part is attributed to
the formation of the surface layer weak ferromagnetic phase
[3] Y.L. Hou, Z.C. Xu, S.H. Sun, Angew. Chem. 119 (2007) 6445–6448.

6
50
H. Shi, X. He / Journal of Physics and Chemistry of Solids 73 (2012) 646–650
[
4] D.P. Dutta, G. Sharma, P.K. Manna, A.K. Tyagi, S.M. Yusuf, Nanotechnology 19
2008) 245609.
[15] N.J.O. Silva, A. Mill a´ n, F. Palacio, M. Martins, T. Trindade, I. Puente-Orench,
(
J. Campo, Phys. Rev. B 82 (2010) 094433.
[
[
5] M. Feygenson, A. Kou, L.E. Kreno, A.L. Tiano, J.M. Patete, F. Zhang, M.S. Kim,
V. Solovyov, S.S. Wong, M.C. Aronson, Phys. Rev. B 81 (2010) 014420.
6] V. Markovich, I. Fita, A. Wisniewski, D. Mogilyansky, R. Puzniak, L. Titelman,
C. Martin, G. Gorodetsky, Phys. Rev. B 81 (2010) 094428.
7] L.Y. Zhang, D.S. Xue, C.X. Gao, J. Magn. Magn. Mater. 267 (2003) 111–114.
8] H.T. Zhang, X.H. Chen, Nanotechnology 16 (2005) 2288–2294.
[16] S.K. Mandal, T.K. Nath, A. Das, R.K. Kremer, Appl. Phys. Lett. 89 (2006)
162502.
[17] A. Sundaresan, C.N.R. Rao, Solid State Commun. 149 (2009) 1197–1200.
[18] L.S. Panchakarla, Y. Sundarayya, S. Manjunatha, A. Sundaresan, C.N.R. Rao,
ChemPhysChem 11 (2010) 1673–1679.
[
[
[
[19] C.N.R. Rao, B. Raveau, Transition Metal Oxides, 2nd ed., Wiley-VCH, Wein-
heim, 1995.
9] Suman Mandal, S. Banerjee, Krishnakumar S.R. Menon, Phys. Rev. B 80 (2009)
2
14420.
[20] M. Gruyters, Phys. Rev. Lett. 95 (2005) 077204.
[
[
[
10] J. Nogu e´ s, J. Sort, V. Langlais, V. Skumryev, S. Suri n˜ ach, J.S. Mu n˜ oz, M.D. Bar o´ ,
Phys. Rep. 422 (2005) 65.
11] R. Zysler, D. Fiorani, J.L. Dormann, A.M. Testa, J. Magn. Magn. Mater. 133
[21] M. Lawecka, A.L. Waniewska, K. Racka, M. Leonowicz, I. Dzhardimalieva,
A.S. Rozenberg, A.D. Pomogailo, J. Alloys Compd. 369 (2004) 244–246.
[22] C.F.J. Flipse, C.B. Rouwelaar, F.M.F. de Groot, Eur. Phys. J. D 9 (1999) 479–481.
[23] Moumita Ghosh, E.V. Sampathkumaran, C.N.R. Rao, Chem. Mater. 17 (2005)
2348–2352.
[24] Ch. Zock, L. Haupt, K. B a¨ rner, B.M. Todris, K. Asadov, E.A. Zavadskii, T. Gron, J.
Magn. Magn. Mater. 150 (1995) 253–262.
[25] L. N e´ el, J. Phys. Soc. Jpn. 17 (1962) 676. Suppl. B.
(
1994) 71.
12] R.H. Kodama, S.A. Makhlouf, A.E. Berkowitz, Phys. Rev. Lett. 79 (1997)
393–1396.
1
[
[
13] R.H. Kodama, A.E. Berkowitz, Phys. Rev. B 59 (1999) 6321–6336.
14] P. Dey, T.K. Nath, A. Banerjee, Appl. Phys. Lett. 91 (2007) 012504.