Properties of Co/Ni codoped ZnO based nanocrystalline DMS

Article abstract of DOI:10.1016/j.jmmm.2011.06.069

Nanoparticles of Co and Ni codoped zinc oxide, Zn_0.9Co _0.1-xNi_xO (x=0.0, 0.03, 0.06 and 0.09), diluted magnetic semiconductors (DMSs) are synthesized by the solgel method at annealing temperature of 500 °C. X-ray diffraction (XRD) patterns confirm the single phase character of the samples with x=0.0 and 0.03. However, minor NiO secondary phase is detected in the samples with x=0.06 and 0.09. All of them possess the hexagonal wurtzite structure. There is no significant change in the lattice parameters due to variation of doping concentration. The average particle size is found to be 19.3125.71 nm. FTIR and UVvis spectroscopic results confirm the incorporation of the dopants into the ZnO lattice structure. Magnetization data reveal the presence of room temperature ferromagnetism (RTFM). The XRD patterns rule out the formation of secondary phase of either metallic Co cluster or CoO in the samples. Nevertheless, the secondary phases are a concern in any DMS system as a source of spurious magnetic signals. Therefore, we carried out the XPS studies from which the oxidation states of Co and Ni are found to be Co ² and Ni², respectively. Moreover, XPS O 1s spectra show evidence of the presence of the oxygen vacancy in the ZnO matrix.

Full text of DOI:10.1016/j.jmmm.2011.06.069

Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
Contents lists available at ScienceDirect
Journal of Magnetism and Magnetic Materials
journal homepage: www.elsevier.com/locate/jmmm
Properties of Co/Ni codoped ZnO based nanocrystalline DMS
Rezq Naji Aljawfi, S. Mollah n
Department of Physics, Aligarh Muslim University, Aligarh 202002, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Nanoparticles of Co and Ni codoped zinc oxide, Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and 0.09), diluted
magnetic semiconductors (DMSs) are synthesized by the sol–gel method at annealing temperature of
500 1C. X-ray diffraction (XRD) patterns confirm the single phase character of the samples with x¼0.0
and 0.03. However, minor NiO secondary phase is detected in the samples with x¼0.06 and 0.09. All of
them possess the hexagonal wurtzite structure. There is no significant change in the lattice parameters
due to variation of doping concentration. The average particle size is found to be 19.31–25.71 nm. FTIR
and UV–vis spectroscopic results confirm the incorporation of the dopants into the ZnO lattice
structure. Magnetization data reveal the presence of room temperature ferromagnetism (RTFM). The
XRD patterns rule out the formation of secondary phase of either metallic Co cluster or CoO in the
samples. Nevertheless, the secondary phases are a concern in any DMS system as a source of spurious
magnetic signals. Therefore, we carried out the XPS studies from which the oxidation states of Co and
Ni are found to be Co2þ and Ni2þ, respectively. Moreover, XPS O 1s spectra show evidence of the
presence of the oxygen vacancy in the ZnO matrix.
Received 18 April 2011
Received in revised form
23 June 2011
Available online 7 July 2011
Keywords:
DMS
XRD
Optical property
Magnetic property
& 2011 Elsevier B.V. All rights reserved.
1. Introduction
codoped ZnO, the origin of RTFM in DMS materials is not under-
stood completely. There are still some reports showing no
Diluted magnetic semiconductor (DMS) is referred to a semi-
conductor in which a small fraction of host cations are replaced
by transition metal ions (TMI) as magnetic elements [1]. In the
recent years, large efforts have been devoted to the investigation
of DMS due to their potential application in spintronic devices [2]
for which the basic requirements are room temperature ferro-
magnetism (RTFM) and good optical transmittance. Zinc oxide
(ZnO) is an attractive semiconductor for numerous applications
and a building block material for many advanced device technol-
ogies because of its large band gap (3.35 eV) at room temperature.
It has large excitonic binding energy (60 meV), optical transpar-
ency, chemical stability, hardness and piezo-electric properties
[5,6]. Doping of ZnO with magnetic ions such as Fe, Co, Ni, etc.
introduces magnetic properties forming DMS [7]. Dietl et al. [3]
have predicted that p-type ZnO-based DMS may exhibit RTFM.
Later, it is suggested theoretically that the RTFM may also exist
in n-type doped ZnO systems [1,4]. These findings prompt the
intense investigations on ZnO-based DMS. Ferromagnetism in
TMI doped ZnO is proved by theoretical calculations and experi-
mentally, RTFM has been reported in V-, Co-, Ni-, Fe-, Cu- and
Mn-doped ZnO systems [8]. Though there are many reports on
structural, optical and magnetic characteristics of TMI-doped or
indication of RTFM or suggesting the presence of secondary
phases as the origin of RTFM in DMS [9]. A number of studies
indicate that the RTFM may originate from the precipitation of
magnetic clusters or from the secondary magnetic phases [10].
Some of these are valuable contributions to the magnetization
mechanism in the DMS system. For example Cr/Co doped ZnO
thin film has been studied by Yan et al. [11], who showed that
Zn0.92Co0.08O film consists of substitutional Co ions and Co metal-
lic clusters. But in the case of codoped of Co and Cr, the Co
metallic clusters and all the Co ions located at Zn sites disappear.
Therefore, the codoped Cr plays an important role in mediating
the atomic distribution of Co ions in ZnO lattice. Lathiotakis et al.
[12] have demonstrated that in codoped Zn(Co–Cu)O, the Cuþ
ions act as superexchange mediators by causing a remote delo-
calization through the hybridization of the Cu d3z2ꢀr2 spin
majority state with the O-p states.
Pure ZnO in nanoscale exhibits ferromagnetic properties due
to the presence of oxygen vacancy on the surface of nanoparticles
[13]. This underscores the fact that the magnetization is not
necessarily resulted from Co metallic culster in DMS. Since the
intrinsic defects form donor states in ZnO, it is concluded that
(Zn,Co)O and (Zn,Ni)O are promising candidates for high transi-
tion temperature (Tc) ferromagnetic materials [14]. Stirred from
the above mentioned facts, it would be quite interesting to
synthesize and investigate the properties exhibited by Co and Ni
codoped ZnO. The present article reports the effect of Co/Ni
n Corresponding author. Tel./fax: þ91 571 2701001.
E-mail address: smollah@rediffmail.com (S. Mollah).
0304-8853/$ - see front matter & 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2011.06.069

R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
3127
codoped on structural, optical and magnetic properties of ZnO
based DMS in the series of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and
0.09) which are synthesized by the sol–gel method at annealing
temperature of 500 1C. The total percentage of dopants (Co/Ni) is
kept constant at 10%; only Ni doping gives rise to secondary
phases. The samples show RTFM. Ni and Co have been chosen as
doping elements since these are important transition metals
with ionic radii close to that of Zn2þ,which leads to an easy
incorporation of Ni2þ and Co2þ ions into the ZnO lattice.
Zn0.9Co0.1-XNixO
∗ NiO
∗
∗
x=0.09
x=0.06
2. Experimental procedure
All Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and 0.09) samples were
synthesized by the sol–gel method [15]. Analytical reagents
(AR) grade zinc nitrate [Zn (NO3) ꢁ 6H2O], nickel nitrate
[Ni (NO3)2.6 ꢁ H2O], cobalt nitrate [Co (NO3)2.6 ꢁ H2O] and citric
acid [C6H8O7] were used as the raw materials without any special
treatment. These were taken at the desired stoichiometric ratio.
Nitrates were dissolved in distilled water to get homogeneous
solution by using magnetic stirrer at 70 1C. Citric acid had been
dissolved separately in distilled water for 30 min and then added
to the nitrate solution. Slowly, the color of nitrate solution
became brighter when the molarity ratio of nitrate to citric acid
was 1:1. The solution was continuously stirred at 70 1C until it
was converted into a gel form. Then it was heated at 130 1C for
12 h. The product was ground for 30 min and finally annealed at
500 1C in air for 12 h. X-ray diffraction (XRD) patterns of the
samples were obtained at room temperature, with a step of 0.021,
x=0.03
x=0.00
20
30
40
50
60
70
80
90
100
2θ (degree)
Fig. 1. X-ray diffraction (XRD) pattern of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and
0.09) samples.
Table 1
Lattice parameters, unit cell volume and particle size of the Zn0.9Co0.1ꢀxNixO
(x¼0.0, 0.03, 0.06 and 0.09) samples.
Samples
c/a
Unit cell
Size of
˚
˚
a¼b (A)
c (A)
3
particle (nm)
˚
volume (A )
using Bruker D8 ADVANCE X-ray diffractometer with Cu K
˚
a
x¼0.0
3.2495
3.2490
3.2493
3.2490
5.2030
5.2050
5.2055
5.2050
1.6011
1.6020
1.6020
1.6020
47.5792
47.5828
47.5874
47.5828
23.6
radiation (
l¼1.54178 A) in the range of 01r2yr1001 at 40 kV.
x¼0.03
x¼0.06
x¼0.09
25.71
22.27
19.31
Fourier transform infrared (FTIR) spectra of the samples were
recorded at room temperature using a Perkin–Elmer BX2 FTIR
spectrometer using KBr pellets as medium. The optical absor-
bance spectra were taken by using a UV–vis double beam Perkin–
Elmer LAMDA 35 spectrophotometer at room temperature in the
wavelength range of 250–800 nm. Magnetization of the samples
has been studied by utilizing superconducting quantum inter-
ference device (SQUID) system (MPMS-XL) of QUANTUM DESIGN
with a sample of 0.2 g in the temperature range of 10–300 K.
Valence states of the elements have been studied from X-ray
photoelectron spectroscopy (XPS) obtained by using the instru-
ment VSW (UK made). The resolution was 0.9 eV. The used X-ray
PowderX diffraction refinement program. No significant change in
the values of lattice parameters ‘a’ and ‘c’ is observed with the
variation of (Co/Ni) doping as the total concentration of (Co/Ni) is
kept at 10%. Furthermore, the values of lattice parameters for all
the samples are close to the standard values of ZnO lattice
parameters. The result is consistent with the fact that the ionic
radii of Co2þ (0.72 A) and Ni
(0.69 A) are close to that of Zn
2þ
2þ
˚
˚
(0.74 A). The XRD results indicate that the Co and Ni2þ ions are
incorporated into the ZnO lattice and replace Zn2þ sites without
changing the wurtzite structure. The size of nanoparticles has
been calculated using Debye–Scherer’s formula
2þ
˚
was Al K and base vacuum was maintained at 10–9 Torr.
a
3. Results and discussion
D ¼ 0:9
l
=ßcos
y
ð1Þ
where D is the diameter of the nanoparticles,
l
is the wavelength
3.1. X-ray diffraction
˚
of X-ray (1.54178 A),
y is the Bragg angle and ß is the line broad-
ening at the full width and half maxima (FWHM) on the highest
peak of plane (1 0 1), which is atꢂ2 ¼36.271. The average of
Fig. 1 shows the XRD pattern of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03,
0.06 and 0.09) samples. The patterns are found to be in good
agreement with the standard peak positions of ZnO (JCPDS Card
No. 36-1451). These peaks reveal that all the investigated samples
are nanocrysalline powder of hexagonal wurtzite structure (space
group P63mc). The diffraction peak at 43.1251 (2 0 0) is attributed
to NiO (JCPDS Card No. 78-0643). It begins to appear in sample
with x¼0.06 (Co¼4% and Ni¼6%) and grows in intensity with the
increasing x. Therefore, the formation of NiO may be from un-
reacted Ni2þ ions present in the solution at high doping concen-
tration. This gives the limit of solid solubility of Ni in the ZnO
crystal lattice. It has already been found that Ni can dissolve less
than 5 at% into ZnO matrix [16]. The relatively higher annealing
temperature (500 1C) also enhances the formation of NiO.
y
nanocrystalline size is found to be 19.31–25.71 nm. The cluster
and secondary phases are a concern in any DMS system as a
source of magnetism. The NiO can be considered as an extrinsic
magnetic source, but it is AFM and no Co metallic cluster or CoO is
detected in all the samples within the sensitivity of the present
XRD technique.
3.2. Optical absorption spectra
Ultraviolet–visible (UV–vis) optical absorption spectra have
been recorded at room temperature in the wavelength range of
250–800 nm as shown in Fig. 2. Redshift in the absorption edge
is observed in the samples with the variation of the Co/Ni
dopants. The direct band gap energy (Eg) is calculated using the
Calculation of lattice parameters (Table 1) and indexing of
the diffraction peaks of all the samples are done by using the

3128
R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
Tauc equation
The exchange interactions between s–d and p–d orbitals give rise to
a negative and a positive correction to the conduction band and
valence band edges, respectively, resulting in a decrease in band
gap [17,18]. By increasing Ni and decreasing Co dopants, the band
gap energy varies and becomes less than that of pure ZnO (Table 2).
This reveals that the Ni and Co ions are incorporated into the lattice
sites. A direct–indirect transition has been proposed by Rakhshani
et al. [19] to explain the band gap narrowing effect. Many groups
have suggested that the alloying effect of parent compound with
some impurity phases may be responsible for the band gap
narrowing [20].
1=2
ð
a
h
nÞ2 ¼ Bðh
where h is the photon energy,
is a constant. The (
n
2EgÞ
ð2Þ
n
a
is the absorption coefficient and B
ah
n 2
)
plotted as a function of h
n
is displayed in
Fig. 3. Values of Eg are obtained by extrapolating the straight line
down to (
ahn
)2¼0 and are found to vary from 3.22 to 3.28 eV
(Table 2). It is well known that the Eg of pure ZnO is 3.35 eV at
300 K [5,6]. Therefore, the absorption edge of Zn0.9Co0.1O (x¼0.0)
sample shifts towards lower energy ꢂ3.28 eV (Redshift). This
shows that the Co2þ ions have got incorporated into the lattice
structure of ZnO. This is interpreted as mainly due to the sp–d
exchange interactions between the band electrons and the localized
d electrons of the Co2þ ions that are substituting Zn2þ ions [17].
3.3. FTIR spectra
Fourier transform infrared (FTIR) spectra were recorded in the
range of 400–4000 cmꢀ1 for all samples. Fig. 4 displays the FTIR
spectra for the two extreme (x¼0.0 and 0.09) samples. Inset of Fig. 4
shows the enlarged view of spectrum of x¼0.9 sample between 400
and 600 cmꢀ1. The distinct absorption band for ZnO appears around
464 cmꢀ1. The position and number of absorption bands not only
depend on crystal structure and chemical composition but also on
particle morphology [21,22]. Studies concerned with this morphology
dependency have shown that in case of spherical ZnO particles,
calculated as well as measured spectra show one distinct absorption
peak around 464 cmꢀ1 [23]. Reference spectra of ZnO powders often
2.0
Zn0.9Co0.1-XNixO
1.8
x=0.0
x=0.03
1.6
1.4
x=0.06
1.2
1.0
0.8
0.6
0.4
0.2
0.0
x=0.09
Table 2
Band gap energy and magnetization parameters of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03,
0.06 and 0.09) samples.
Samples Band gap
energy (eV)
HC (emu/g) MR (emu/g) Ms (emu/g) MR/MS (%)
300
400
500
600
700
800
x¼0.0
3.28
3.26
3.22
3.25
53.80
70.78
59.79
78.61
2.87 ꢃ 10–4
15.7 ꢃ 10–4
3.5 ꢃ 10–4
3.65 ꢃ 10–4
14.7 ꢃ 10–3
1.9
14.9 ꢃ 10–3 10.5
1.36 ꢃ 10–3 25.7
Wavelength (nm)
x¼0.03
x¼0.06
x¼0.09
8.1 ꢃ 10–3
4.5
Fig. 2. Room temperature optical absorption spectra of Zn0.9Co0.1ꢀxNixO (x¼0.0,
0.03, 0.06 and 0.09) samples.
20.0
19.5
37
36
35
34
33
32
x=0.0
Eg= 3.28 eV
x=0.03
Eg=3.26 eV
19.0
18.5
18.0
17.5
17.0
16.5
3.2 3.3 3.4 3.5 3.6 3.7
Binding Energy (eV)
3.2
3.3
3.4
3.5
3.6
Binding Energy (eV)
15.0
14.5
14.0
13.5
13.0
12.5
12.0
14.5
14.0
13.5
13.0
12.5
12.0
11.5
11.0
x=0.06
x=0.09
Eg=3.22 eV
Eg= 3.25 eV
3.2 3.3 3.4 3.5 3.6 3.7
Binding Energy (eV)
3.2 3.3 3.4 3.5 3.6 3.7
Binding Energy (eV)
2
Fig. 3. The (
ah
n)
versus. binding energy plots for Zn0.9Co0.1ꢀxNixO samples with x¼0.0 (a), 0.03 (b), 0.06 (c) and 0.09 (d). The band gap energy (Eg) is determined by
extrapolating the straight line and taking the intersection with x-axis.

R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
3129
demonstrate two absorption maxima at ꢂ512 and ꢂ406 cmꢀ1 [24].
In the present investigation, FTIR spectra of the samples illustrate the
characteristic peaks in the range of 420–620 cmꢀ1 (inset of Fig. 4)
and corresponding to the stretching vibration modes of ZnO [25].
The Zn–O stretching mode appears at 425 cmꢀ1, indicating that the
transformation from zinc nitrate to zinc oxide is completed in the
samples annealed at 500 1C. The bands occurring near 762 and
832 cmꢀ1 are attributed to the vibrations of ZnO–Ni local bonds
and defect states respectively. Absorption bands observed in the
ranges of 600–680 and 430–470 cmꢀ1 are attributed to the stretching
modes of ZnO [26–28] in the tetrahedral and octahedral coordination,
respectively. The gradual shift in the absorption frequency with
different Co/Ni doping is caused by the difference in the bond lengths
that occurs when Co2þ and Ni2þ ions replace Zn2þ ions, and confirm
the incorporation of Co and Ni into ZnO lattice structure.
120
Zn0.9Co0.1-xNixO
3.4. X-ray photoelectron spectra
x=0.09
105
90
75
60
45
30
15
0
X=0.0
X=0.09
The XRD patterns exclude the formation of secondary phases
in samples except minor NiO phase at high Ni concentration.
However, the nano-clustering of Ni and Co atoms are considered
as a source of magnetization. Therefore, XPS study was performed
on the sample with x¼0.03 (Zn0.9Co0.07Ni0.03O). The XPS aims to
investigate the oxidation states of ions present in the sample. The
peaks of Zn in XPS spectrum are shown in Fig. 5(a). The two peaks
are located at 1024.5 and 1047.7 eV, which correspond to the
binding energy of core levels 2p3/2 and 2p1/2, respectively. The
energy difference is 23.2 eV, which coincides with the findings for
Zn2þ bound to oxygen in the ZnO matrix [29]. The oxygen 1s
spectrum shows only a single peak that can be fitted into two
Gaussian peaks having different binding energy positions (Fig. 5b).
The peak centered at 530.97 eV (the lower binding energy) in the
O 1s spectrum is attributed to O2ꢀ ions on wurtzite structure of
hexagonal ZnO [30]. The peak centered at 532.31 eV (the higher
binding energy) is associated with O2ꢀ ions in the oxygen
deficient regions within the matrix of ZnO [30]. This indicates
550
450
400
500
O-H
ZnO
500
C=O
4000 3500 3000 2500 2000 1500 1000
Wavenumber (cm-1)
Fig. 4. FTIR spectra of Zn0.9Co0.1ꢀxNixO (x¼0.0 and 0.09) samples in the wave
number range of 400–4000 cmꢀ1. The inset shows the absorption peaks for ZnO in
the range of 400–600 cmꢀ1
.
O 1S
Zn 2p3/2
1024.5 eV
Data
Gaussian fit
Gaussian fit peaak 1
Gaussian fit peaak 2
Zn 2p1/2
1047.7 eV
532.31 eV
530.97 eV
1020 1025 1030 1035 1040 1045 1050
Binding Energy (eV)
526 528 530 532 534 536 538
Binding Energy (eV)
Co 2p3/2
Ni 2p3/2
855.74 eV
Ni 2p1/2
782.36 eV
871.93 eV
Co 2p1/2
797.75 eV
780
785
790
795
800
855
860
865
870
Binding Energy (eV)
Binding Energy (eV)
Fig. 5. XPS spectra of Zn0.9Co0.07Ni0.03O sample for (a) Zn, (b) O 1s with two Gaussian peaks, (c) Co and (d) Ni.

3130
R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
that there are some oxygen vacancies in the sample [31].
The valence states of cobalt in the system are shown in
Fig. 5(c). The Co 2p peaks are located at ꢂ782.36 eV for 2p3/2
and at ꢂ797.75 eV for 2p1/2 electrons. Lakshmi et al. [32] have
reported the value of difference in binding energy to be
ꢂ15.05 eV for pure Co metal and 15.5 eV for Co2þ ions. For the
present sample, the energy difference is found to be ꢂ15.39 eV
which is close to that of Co2þ. Therefore, we suggest that the
Co ions exist in the þ2 valence state within the matrix of ZnO.
Neither Co cluster metal nor CoO was detected, within the
resolution of the technique. XPS spectra of Ni 2p3/2 and Ni 2p1/2
in the Zn0.9Co0.07Ni0.03O nanocrystalline powder is displayed in
Fig. 5(d). The peak of Ni 2p3/2 is located at 855.4 eV which is
associated with Ni2þ ions in the sample [33]. It indicates that the
Ni2þ ions are substituted for the Zn2þ ions in the ZnO lattice.
hysteresis curve for x¼0.06 sample. All other samples show
similar behavior. The hysteresis loops indicate that all samples
have weak RTFM. For the sample of x¼0.0 (Zn0.9 Co0.1O), the
observed HC is ꢂ53.80 Oe, the remanent magnetization MR is
ꢂ2.87 ꢃ 10ꢀ4 emu/g and the saturation magnetization MS
isꢂ14.7 ꢃ 10–3 emu/g at 10 KOe. For Zn0.9Co0.07Ni0.03O (x¼0.03)
sample, the magnetization increases as compared to x¼0 sample
and shows very clear hysteresis loop. This can be attributed to
the reduction of AFM interaction between Co2þ and Co2þ ions.
The codoping of Ni ions in the samples effectively hampers the
formation of metallic clusters, suggesting that it mediates the
distribution of Co2þ ions in the ZnO host. The samples with
x¼0.06 and 0.09 show a decrease in magnetization where an
appearance of NiO secondary phase is observed (Fig. 1). The bulk
NiO is antiferromagnetic with a Neel temperature (TN) ofꢂ520 K
[17] and the nanaocrystalline NiO shows weak ferromagnetic or
super-paramagnetic (SPM) behavior at low temperature [18].
Fig. 7a and b illustrates M–T curves for samples of x¼0.0 and
0.06, respectively, during zero-field cooling (ZFC) and field cooling
(FC) modes in an applied magnetic field of 1000 Oe. The insets
demonstrate the inverse magnetic susceptibility (wꢀ1) plotted as
a function of temperature (T). The ZFC and FC curves do not show
ferromagnetic transition in the temperature range of 10–300 K.
The magnetization increases with the decrease in T and the rapid
increase occurs below 60 K. This behavior is similar to paramag-
netic one. However, the remanent magnetization at room tem-
perature and the rapid increase of magnetization at low
temperature can be considered as ferromagnetic properties. No
difference between ZFC and FC in sample of x¼0.0 is observed
(Fig. 7a), while the sample with x¼0.06 (Fig. 7b) displays small
disparity between ZFC and FC curves which indicates the pre-
sence of antiferromagnetic properties. Inverse magnetic suscept-
ibility (wꢀ1) is given by
3.5. Magnetic properties
Fig. 6 illustrates the magnetization versus applied magnetic
field (M–H) curves of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and 0.09)
samples at room temperature. The inset of Fig. 6 displays the
0.02
Zn0.9Co0.1-XNixO
X=0.0
X=0.03
X=0.06
X=0.09
0.01
0.00
-0.01
wꢀ1 ¼ ðTꢀ
where T is the absolute temperature,
yÞ=C
ð3Þ
y
is the Curie–Weiss
-0.02
temperature and C is the Curie constant. The linear fit of the
wꢀ1–T) curves show negative Curie–Weiss temperature (
ꢂ–62
and –174 K for x¼0.0 and 0.06 respectively). The negative value
of is usually observed in the case of local atomic antiferromag-
-15000 -10000 -5000
0
5000
10000 15000
(
y
Applied Magnetic Field (KOe)
y
Fig. 6. M–H curves for Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and 0.09) samples at
300 K. The inset shows the clear hysteresis as well as Hc and MR for the sample of
x¼0.06. Other samples also show similar behavior.
netic interaction. However, the magnetization does not become
zero and the hysteresis loop shows FM at room temperature.
0.05
Zn0.9Co0.1
O
6x109
5x109
Zn0.9Co0.04Ni0.06
O
0.05
0.04
0.03
0.02
0.01
0.00
2.0x109
0.04
0.03
0.02
0.01
0.00
4x109
3x109
2x109
1x109
0
1.5x109
1.0x109
5.0x108
0.0
0
50
100 150 200 250 300
Temperature (K)
0
50 100 150 200 250 300
Temperature (K)
FC
FC
ZFC
50
ZFC
0
100
150
200
250
300
0
50
100
150
200
250
300
Temperature (K)
Temperature (K)
Fig. 7. M–H curves for x¼0.0 (a) and 0.06 (b) samples in zero-field cooling (ZFC) and field cooling (FC) modes at a magnetic field of 1000 Oe. The insets show inverse of
magnetic susceptibility (1/
¼H/M) versus temperature from 10 to 300 K.
w

R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
3131
This may be interpreted as an indication that only minor Co spins
are ferromagnetically coupled and the predominant Co spins
remain uncoupled or even antiferromagnetically coupled [34].
in magnetization by codoping of Ni and Co may be due to
Ni-induced acceptor state localized within the gap region, which
reduces the acceptor mobility. Oxygen vacancies (VO) have also
been proposed to play a significant role for the origin of ferro-
magnetism in oxide DMS [38]. An electron locally trapped in VO
occupies an orbit and overlaps the d shells of Co2þ neighbors.
This has an important role in the spin orientations of neighboring
Co2þ ions. Based on the Hund’s rule and Pauli’s exclusion
principle, spin orientations of trapped electrons and neighboring
Co2þ ions should be parallel in the same direction and the
ferromagnetism should be achieved. More oxygen vacancies lead
to more Co2þ–VO–Co2þ groups, which increases the ferromagnet-
ism in the system [43].
Effective magnetic moment
magnetic susceptibility as given by
m can be calculated from the molar
w
¼ Nb2
m
2=3kT
ð4Þ
where
the number density of doping ions. Substituting the values of N,
b is the Bohr magneton and k is Boltzmann’s constant. N is
b
and k, we may get
m as
1=2
m
¼ 2:827ð
w
TÞ ¼ 2:827 C1=2mB
ð5Þ
where the Curie constant C can be obtained from the slope of wꢀ1
versus. T curves using Eq. (3). The calculated value of
m
for sample
of x¼0.0 is found to be 2.47
mB which is smaller than the value of
spin magnetic moment of Co2þ ions (ꢂ3.87
mB) in tetrahedral
4. Conclusion
symmetry. It confirms that all Co ions do not contribute to the
magnetization of the sample. For x¼0.06 sample, the value of
m is
Nanoparticles of Zn0.9Co0.1ꢀxNixO (x¼0.0, 0.03, 0.06 and 0.09)
DMS were synthesized by sol–gel method with annealing tem-
perature of 500 1C. XRD pattern reveal that all investigated
samples are nanocrystalline powder of hexagonal wurtzite struc-
ture (space group P63mc). The average of nanoparticles size is
19.31–25.71 nm. FTIR and UV–vis spectroscopic data show the
evidence for the incorporation of Co2þ and Ni2þ ions into the
nonmagnetic host of ZnO lattice without any modification in
the structure. The band gap energy decreases with the increase of
Ni and decrease of Co dopants concentration. The RTFM has been
presented in terms of vacancy in the frame of BPM model, where
the Co ions at surface of the nanoparticles can be ferromagneti-
cally coupled and mediated by oxygen vacancies. Because of low
carrier concentration in the system, we cannot use the RKKY
theory to explain the RTFM. As there is no Co metallic cluster in
the present DMS, the cluster as source of ferromagnetism is
ruled out.
2.39 mB which is also smaller than the effective magnetic moment
of Co2þ (ꢂ3.87
m
B) and Ni2þ (ꢂ2.83
mB).
Though the secondary phases are a concern in any DMS system
as a source of spurious magnetic signals, the origin of ferromag-
netism in the present system cannot be attributed to the forma-
tion of secondary phases. The possible secondary phases may be
Co metallic cluster, CoO or NiO. But CoO and NiO are antiferro-
magnetic with small positive susceptibility having TN ꢂ293 and
ꢂ520 K, respectively. The Co metallic cluster can provide the
dominant source of the ferromagnetism [35]. However the XRD
and XPS data do not support the presence of Co metal in the
samples within the sensitivity of XRD and XPS. Therefore, the
possibility of ferromagnetism due to Co metallic cluster can be
ruled out in the present study.
The weak RTFM cannot be explained by Ruderman–Kittel–
Kasuya–Yosida (RKKY) theory [36]. According to this theory, the
magnetization is due to exchange interaction between local spin-
polarized electrons of Co2þ ions and conduction electrons. The
interaction leads to the spin-polarization of conduction electrons.
Subsequently, the spin-polarized conduction electrons perform an
exchange interaction with local spin-polarized electrons of other
Co2þ ions. Consequently, the long-range exchange interactions
lead almost to all the Co2þ ions to have same spin direction. The
conduction electrons act as media for the interaction among
the Co2þ ions, where the free carrier is a necessary condition
for the appearance of ferromagnetism. But for confirming the
existence of free carriers we have studied the resistivity of the
Acknowledgments
We are very thankful to Dr. Mukul Gupta and Dr. T. Shripathi,
UGC–DAE-CSR, Indore, for giving the XRD and XPS characterization
facilities, respectively. The authors are grateful to Dr. A. Banerjee,
UGC–DAE-CSR, Indore, for providing magnetization measurement
facility. Facility provided by the Center of Excellence in Material
Science (Nanomaterials), Department of Applied Physics, AMU,
Aligarh, is highly acknowledged.
samples which is found to be very high (ꢂ105–107
O cm). This
indicates to a very low carrier concentration in the present
samples. A defect mediated ferromagnetism model based on
the bound magnetic polaron (BMP) theory has been predicted
theoretically [37,38] and confirmed experimentally [39,40]. For
instance, surface defects have been implicated to be essential for
the generation of ferromagnetism in DMS oxide [39,41]. In the
present study, the oxygen vacancies have been detected by XPS
and would likely be generated on the surfaces of the nanoparti-
cles. According to BMP theory, when the concentration of surface
defects exceeds the percolation threshold, the surface defects can
overlap many dopant ions as well as adjacent defects, inducing a
ferromagnetic coupling between dopant spins. Thus, we can
speculate that the Co ions at surface of the nanoparticles can be
ferromagneticlly coupled and mediated by oxygen vacancies. If
the ferromagnetism arises from the surface of Co2þ ions, only a
few of them may contribute to the ferromagnetism because most
of the Co2þ ions reside in the core of the nanoparticles. However,
due to the insufficiency of oxygen vacancies in the core of the
nanoparticles, these Co2þ spins are distributed as uncoupled
spins or mediated through oxygen ions [34,42]. The enhancement
References
[1] J.J. Lu, T.C. Lin, S.Y. Tsai, T.S. Mo, K.J. Gan, J. Magn. Magn. Mater. 323 (2011) 829.
[2] H. Ohno, Science 281 (1998) 951.
[3] T. Dietl, H. Ohno, F. Matsukura, J. Cibert, D. Ferrand, Science 287 (2000) 1019.
[4] K. Saito, H. Katayama-Yoshida, Jpn. J. Appl. Phys. 39 (2000) L555.
[5] B. Pandey, S. Ghosh, P. Srivastava, D. Kabiraj, T. Shripati, N.P. Lalla, Physica E
41 (2009) 1164.
[6] U. Ozgur, Y.I. Alivov, C. Liu, A. Teke, M.A. Reshchikov, S. Dogan, V. Vrutin,
S.J. Cho, H. Morkoc, J. Appl. Phys. 98 (2005) 041301.
[7] M. El-Hilo, A.A. Dakhel, A.Y. Ali-Mohamed, J. Magn. Magn. Mater. 321 (2009)
2279.
[8] X. Xu, C. Cao, Z. Chen, J. Magn. Magn. Mater. 323 (2011) 1886.
[9] Y. Liu, J. Yang, Q. Guan, L. Yang, H. Liu, Z. Wang, Y. Wang, D. Wang, Z.Q. Chen,
D.D. Wang, N. Qi, J. Gong, C.Y. Cao, Z. Tang, Appl. Surf. Sci. 256 (2010) 3559.
[10] S. Ramachandran, A. Tiwari, J. Narayan, Appl. Phys. Lett. 84 (2004) 5255.
[11] W. Yan, Z. Sun, Q. Liu, T. Yao, Q. Jiang, F. Hu, Y. Li, J. He, Y. Peng, S. Wei, Appl.
Phys. Lett. 97 (2010) 042504.
[12] N.N. Lathiotakis, N.A. Andriotis, M. Menon, Phys. Rev. B 78 (2008) 193311.
[13] D. Gao, Z. Zhang, J. Fu, Y. Xu, J. Qi, D. Xue, J. Appl. Phys. 105 (2009) 113928.
[14] G. Lawes, A.S. Risbud, A.P. Ramirez, R. Seshadri, Phys. Rev. B 71 (2005)
045201.
[15] R.N. Aljawfi, A. Azam, S. Mollah, J. Phys. D: Appl. Phys., 2011, submitted for
publication.

3132
R.N. Aljawfi, S. Mollah / Journal of Magnetism and Magnetic Materials 323 (2011) 3126–3132
[16] T. Wakano, N. Fujimura, Y. Morrinaga, N. Abe, A. Ashida, T. Ito, Physica E
10 (2001) 260.
[30] M. Chen, X. Wang, Y.H. Yu, Z.L. Pei, X.D. Bai, C. Sun, R.F. Huang, L.S. Wen, Appl.
Surf. Sci. 158 (2000) 134.
[17] M. Bouloudenine, N. Viart, S. Colis, J. Kortus, A. Dinia, Appl. Phys. Lett.
87 (2005) 052501.
[18] P. Koidl, Phys. Rev. B 15 (1977) 2493.
[31] J.C.C. Fan, J.B. Goodenough, J. Appl. Phys. 48 (1977) 3524.
[32] K.Y. Lakshmi, K. Srinivas, B. Sridhar, M.M. Raja, M. Vithald, P.V. Reddy, Mater.
Chem. Phys. 113 (2009) 749.
[19] A.E. Rakhshani, Y. Makdisi, H.A. Ramazaniyan, J. Appl. Phys. 83 (1998) 1049.
[20] L. Chun-Ming, F. Li-Mei, Z. Xiao-Tao, Z. Wei-Lie, Chin. Phys. 16 (2007) 95.
[21] F.A. Sigoli, M.R. Davolos, M.J. Jafelicci, J. Alloys Compd. 292 (1997) 262.
[22] S. Hayashi, N. Nakamori, H. Kanamori, J. Phys. Soc. Jpn. 46 (1979) 176.
[33] C. Xia, C. Hu, Y. Tian, B. Wan, J. Xu, X. He, Physica E 42 (2010) 2086.
[34] Z.M. Tian, S.L. Yuan, J.H. He, P. Li, S.Q. Zhang, C.H. Wang, Y.Q. Wang, S.Y. Yin,
L. Liu, J. Alloys Compd. 466 (2008) 26.
[35] S. Maensiri, P. Laokul, J. Klinkaewnarong, J. Magn. Magn. Mater. 302 (2006) 448.
[36] M.A. Ruderman, C. Kittel, Phys. Rev 96 (1954) 99.
´
´
[23] M. Andres Verges, A. Mifsud, C.J. Serna, J. Chem. Soc. Faraday Trans. 86 (1990)
959.
[37] A. Kaminski, S.D. Sarma, Phys. Rev. Lett. 17 (2002) 247202.
[38] J.M.D. Coey, M. Venkatesan, C.B. Fitzgerald, Nat. Mater. 4 (2005) 173.
[39] K.R. Kittilstved, W.K. Liu, D.R. Gamelin, Nat. Mater. 2 (2006) 291.
[40] X.F. Wang, J.B. Xu, N. Ke, J.G. Yu, J. Wang, Q. Li, H.C. Ong, R. Zhang, Appl. Phys.
Lett. 88 (2006) 223108.
[24] Sadtler Research Laboratories (Ed.), The Infrared Spectra Handbook of
Inorganic Compounds, Heyden & Son Ltd., London, 1984.
[25] H. Kleinwechter, C. Janzen, J. Knipping, H. Wiggers, P. Roth, J. Mater. Sci.
7 (2002) 4349.
[26] C.J. Conga, J.H. Honga, K.L. Zhanga, Mater. Chem. Phys. 113 (2009) 435.
[27] K. Yosida, Phys. Rev. 106 (1957) 893.
[28] H. Kumagai, Y. Oka, S. Kawata, M. Ohba, K. Inoue, M. Kurmoo, H. Okawa,
Polyhedron 22 (2003) 1917.
[41] H.S. Hsu, J.C.A. Huang, S.F. Chen, C.P. Liu, Appl. Phys. Lett. 90 (2007) 102506.
[42] R.K. Singhal, A. Samariya, Y.T. Xing, S. Kumar, S.N. Dolia, U.P. Deshpande,
T. Shripathi, E.B. Saitovitch, J. Alloys Compd. 496 (2010) 324.
[43] P. Long, Z. Huai-Wu, W. Qi-Ye, S. Yuan-Qiang, S.U. Hua, J.Q. Xiao, Chin. Phys.
Lett. 25 (2008) 1438.
[29] Y.J. Kwon, K.H. Kim, C.S. Lim, K.B. Shim, J. Ceram. Proc. Res. 3 (2002) 146.