Physica E 74 (2015) 93–100

Contents lists available at ScienceDirect

Physica E
journal homepage: www.elsevier.com/locate/physe

Preparation, structural, photoluminescence and magnetic studies of Cu

doped ZnO nanoparticles co-doped with Ni by sol–gel method
D. Theyvaraju, S. Muthukumaran n,1
PG & Research Department of Physics, H.H. The Rajah's College (Autonomous), Pudukkottai 622 001, Tamilnadu, India

H I G H L I G H T S G R A P H I C A L A B S T R A C T

Ni, Cu co-doped ZnO nanoparticles Secondary phases, CuO and Zn were noticed at higher Cu content. Tuning of energy gap and PL emission
were successfully synthesized by band were discussed based on defect formation.
sol–gel method.

Secondary phases, CuO (111) and Zn
(101) were noticed at higher Cu
content.

Energy gap was tuned by Cu content.

UV and visible emission band were
discussed based on defect formation.

art ic l e i nf o a b s t r a c t

Article history: Zn0.96  xNi0.04CuxO nanoparticles have been synthesized by varying different Cu concentrations between
Received 17 May 2015 0% and 4% using simple sol–gel method. X-ray diffraction studies confirmed the hexagonal structure of
Received in revised form the prepared samples. The formation of secondary phases, CuO (111) and Zn (101) at higher Cu content is
7 June 2015
due un-reacted Cu2 þ and Zn2 þ ions present in the solution which reduces the interaction between
Accepted 15 June 2015
Available online 27 June 2015
precursor ions and surfaces of ZnO. Well agglomerated and rod-like structure noticed at Cu ¼4% greatly
de-generate and enhanced the particle size. The nominal elemental composition of Zn, Cu, Ni and O was
Keywords: confirmed by energy dispersive X-ray analysis. Even though energy gap was increased (blue-shift) from
X-ray diffraction Cu ¼0–2% by quantum size effect, the s–d and p–d exchange interactions between the band electrons of
Energy gap
ZnO and localized d electrons of Cu and Ni led to decrease (red-shift) the energy gap at Cu ¼4%. Presence
Photoluminescence
of Zn–Ni–Cu–O bond was confirmed by Fourier transform infrared analysis. Ultraviolet emission by band
Magnetic property
to band electronic transition and defect related blue emission were discussed by photoluminescence
spectra. The observed optical properties concluded that the doping of Cu in the present system is useful
to tune the emission wavelength and hence acting as the important candidates for the optoelectronic
device applications. Ferromagnetic ordering of Cu ¼ 2% sample was enhanced by charge carrier con-
centration where as the antiferromagnetic interaction between neighboring Cu–Cu ions suppressed the
ferromagnetism at higher doping concentrations of Cu.
& 2015 Elsevier B.V. All rights reserved.

1. Introduction
n
Corresponding author. Fax: þ 91 4322 230490. In recent years, nanotechnology and nanomaterials have at-
E-mail address: drsmk123@yahoo.co.in (S. Muthukumaran).
1
Presently working in Government Arts College, Melur 625106, Tamilnadu, tracted several researchers from different fields due to their un-
India. ique properties and potential applications in different areas such

http://dx.doi.org/10.1016/j.physe.2015.06.012
1386-9477/& 2015 Elsevier B.V. All rights reserved.
94 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
as spintronics devices, transparent electronics, piezoelectricity,
optoelectronics, etc. [1–4]. ZnO is known as ionic semiconductor
with a wide band gap of about 3.36 eV at room temperature and a
high excitation binding energy of 60 meV, which give rise to its
outstanding properties and potential applications in diverse fields
[5–8].
Doping is a widely used method to improve electrical, optical
and magnetic properties of semiconductor compounds, facilitating
the construction of many electronic and optoelectronic devices.
Transition metal (TM) ion alters the electronic and magnetic
properties of ZnO due to the exchange interaction between s and p
electron of host ZnO and d electron of TM ions [9]. Nickel (Ni) ion
is an important with rich optical properties in different transpar-
ent materials and also has the advantage of Ni2 þ ion is its sensi-
tivity towards their legend field environment. Some researches on
Ni doped ZnO have been reported and results showed that the
luminescence properties of ZnO were changed after doping with
Ni [10,11]. Liu et al. [12] noticed that the preparation details of Ni-
doped ZnO has great influence on their magnetic properties, so
that annealing at 800 °C in Ar gas atmosphere significantly in-
creases the magnetizable of Zn1  xNixO material. To avoid the
secondary formation Ni concentration is limited to 4%.
Among the different TM elements, Cu gets much interest due to
its similar electronic shell structure, physical and chemical prop-
erties to those of Zn [13]. Sharma et al. [14] have studied the
doping dependent room temperature ferromagnetism (RTFM) and
structural properties of ZnO:Cu2 þ nano-rods. Cu-doped ZnO na-
noparticles with 40 nm size have been synthesized by solution
combustion method which showed a red-shift and narrowing of
band gap [15]. The enhanced ferromagnetic order was observed in
Ni and Cu doped ZnO powder where Cu acted as an acceptor [16].
The fabrication and magnetic characteristics of Ni and Cu co-
doped ZnO nanorods are reported by Tang et al. [17]. They ex-
plained the room temperature ferromagnetism by the exchange
interaction between free delocalized carriers (holes from valence
band) and the localized d spins of Ni and Cu ions.
Different physical or chemical synthetic methods have been
used to prepare the ZnO nanoparticles such as thermal decom-
position [18], chemical vapor deposition, sol–gel [19], co-pre-
cipitation [20] and hydrothermal [21]. Among the different phy-
sical and chemical methods [18–21], sol–gel is one of the most
important methods to prepare the nanopowders. Sol–gel method
is a simple and easily reproducible method [22].
Until now, some articles have been published about Ni-doped
ZnO nanoparticles and Cu-doped ZnO nanoparticles. Even though
some of the literature focussed on Ni and Cu-doped ZnO nano-
particles, the detailed study about Ni, Cu co-doped ZnO nano-
particles is nearly scanty. The main goal of the present work is to
prepare Ni, Cu co-doped ZnO nanoparticles by sol–gel method and
the characterization studies such as structure, crystallite size,
surface morphology, band gap and photoluminescence behavior of
the nanoparticles.
2. Experimental
2.1. Preparation of Zn0.96–xNi0.04CuxO (x ¼0.00, 0.02 and 0.04)
nanoparticles
During the preparation of Zn0.96  xNi0.04CuxO (x ¼0.00, 0.02 and
0.04) nanoparticles, the high purity chemicals (499% purity) such
as Zinc acetate dihydrate (Zn(CH3CO2)2  2H2O), Nickel acetate
tetrahydrate (Ni(CH3CO2)2  4H2O), Copper acetate monohyrate
((CH3COO)Cu  H2O), N,N dimethyl-formamide (DMF) were used
without further purification. The appropriate amount of Zinc and
Nickel acetates were dissolved in N,N dimethyl-formamide (DMF)
one by one under constant stirring for 1 h using magnetic stirrer.
Followed by Zinc and Nickel acetates, the appropriate amount of
Copper acetate was added into the solution under constant stirring
for another 1 h.
The prepared homogeneous solution was kept at 60 °C for 1 h
under constant stirring to ensure the complete reaction. Then, the
resulting sols were evaporated using muffle furnace. The collected
precipitates were dried by using micro-oven for two hours. The
dried precursors were collected and ground in an agate mortar.
Finally, the collected nanopowders were annealed at 500 °C under
air atmosphere for 2 h followed by furnace cooling. The same
procedure is repeated to the remaining samples synthesized with
nominal compositions of Zn0.96  xNi0.04CuxO (x ¼0.00, 0.02 and
0.04).
2.2. Characterization techniques
The crystal structure of Zn0.96  xNi0.04CuxO (x ¼0.00, 0.02 and
0.04) nanoparticles was determined by powder X-ray diffraction
technique (XRD). XRD patterns were recorded by RigaKu C/max-
2500 diffractometer using Cu Kα radiation (λ ¼1.5406 Å) at 40 kV
and 30 mA from 2θ ¼30–70°. The topological features and com-
position of Zn, O, Ni and Cu were determined by energy dispersive
X-ray (EDX) spectrometer using K and L lines. The surface mor-
phology of Zn0.96  xNi0.04CuxO nanoparticles were studied using a
scanning electron microscope (SEM, JEOLJSM 6390). SEM techni-
que is used to study the particle size, surface morphology and
distribution of the particles of the sample.
The UV–visible optical absorption and transmittance studies
were carried out to explore their optical properties. The optical
absorption and transmittance were determined using UV–visible
spectrometer (Model: lambda 35, Make: PerkinElmer) in the wa-
velength ranging from 300 nm to 700 nm at room temperature
using cm  1 quartz cuvettes. Halogen and deuterium lamp are used
as sources for visible and UV radiations, respectively at room
temperature.
The chemical bonding existing in Zn0.96  xNi0.04CuxO nano-
particles was studied by Fourier transform infra red (FTIR) spec-
trometer (Model: PerkinElmer, Make: Spectrum RX I) from 400 to
4000 cm  1. The sample used for FTIR studies is in the form of
pellets prepared by mixing the nanoparticles with KBr at 1 wt%.
The photoluminescence (PL) spectra of Zn0.96  xNi0.04CuxO nano-
particles have been carried out between the wavelength ranging
from 320 nm to 520 nm under the excitation of Xenon lamp laser
with 310 nm line using a fluorescence spectrophotometer (F-2500,
Hitachi) at room temperature. The magnetization (M) and mag-
netic hysteresis (M–H) loops were measured at room temperature
using vibrating sample magnetometer (VSM, Make: Lake shore,
Model: 7404).
3. Results and discussion
3.1. X-ray diffraction (XRD): structural studies
The typical XRD patterns of Cu-doped and Ni, Cu co-doped ZnO
nanoparticles synthesized by sol–gel method are shown in Fig. 1a.
All the diffraction peaks can be indexed as pure ZnO phase with a
hexagonal symmetry. Fig. 1b illustrates the magnified and high
resolution XRD spectra along (101) plane which shows the varia-
tion of peak position and peak intensity as a function of Cu be-
tween 0% and 4%. The XRD spectra of Zn0.96Ni0.04O nanoparticles
prepared using N,N-dimethyl formamide (DMF) as a solvent
clearly shows the nanocrystalline nature corresponding to the
diffraction angles 31.59° (100), 34.25° (002), 36.09° (101), 47.39°
(102), 56.46° (110), 62.77° (103), 66.24° (200), 67.79° (112) and
 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
Fig. 1. X-ray diffraction pattern of Zn0.96Ni0.04CuxO nanoparticles with x ¼0, 2% and 4% (a) between 30° and 70° and (b) enlarged view from 35.7° to 36.9°.
68.93° (201) planes.
The observed characteristic peaks are in close agreement with
the pure ZnO (JCPDS 36-1451, a ¼3.2488 Å, c¼5.2061 Å). All peaks
match the hexagonal ZnO structure having space group P63mc with
preferred orientation along (101) plane in all the samples. In ad-
dition, there are no traces of impurity or secondary phases such as
Ni, NiO, Cu, CuO, Cu2O up to Cu¼ 2% within the detection limit of
XRD. It is noticed from Fig. 1 that the peak intensity increases with
Cu concentration which means that the crystalline quality is im-
proved by Cu doping and also Cu2 þ ions are understood to have
occupied Zn2 þ without changing crystal structure. It can be seen
from Fig. 1 that at higher doping percentages of Cu2 þ (4%), the
new phases emerges at 38.73° and 42.43° in the XRD spectra. The
first new phase at 38.73° in the XRD spectra corresponds to CuO
(111) (matched with JCPDS 05-0661), which may be due to the
formation of CuO from remaining un-reacted Cu2 þ ions present in
the solution. The secondary new phase at 42.43° in the XRD
spectra corresponding to Zn (101) (JCBDS card no.: 04-0831) which
may be due to the un-reacted Zn2 þ ions. The presence of CuO and
un-reacted Zn2 þ in Cu–O at Cu¼ 4% is due to the low polarity of
the ions which reduces the interaction between precursor ions and
surfaces of ZnO.
From XRD, a slight lattice deformation was observed for Cu-
doped Zn–Ni–O due to differences in ionic radii. This could be due
to the shorter CuZn–O bonds and smaller [CuZnO4] units in Cu-
doped Zn–Ni–O material [23]. Meanwhile, diffraction peaks of
Cu ¼2% doped nanoparticles shift towards higher angles (∼0.10°)
compared to the undoped one (Fig. 1b), due to the smaller radius
of Cu2 þ ions (0.054 nm) than that of Zn2 þ ions [24]. No con-
siderable change in peak positions by the further increase of Cu
concentration. No change in the crystalline structure is detected,
which suggests that the most Cu atoms are in the ZnO wurtzite
lattice. All the available reflections of the present XRD phases have
been fitted with Gaussian distribution. The average crystallite size
of the nanoparticles is calculated after appropriate background
correction from X-ray line broadening using Debye Scherrer's
formula [25]
0.9λ
Average crystallite size (D) =
β cos θ
where λ is the wavelength of X-ray used (1.5406 Å), β is the an-
gular peak width at half-maximum in radian along (101) plane and
95
θ is Bragg's diffraction angle. The micro-strain (ε) can be calcu-
lated using the formula [25]
β cos θ
Micro−strain (ε) =
4 (2)
Table 1 shows the full width at half-maximum (FWHM) value,
average crystallite size (D) and micro-strain (ε) of Zn0.96  xNi0.04Cu
xO (0 rx r0.04) nanoparticles. The average crystallite size of
Zn0.96Ni0.04O is 21.3 nm. From XRD pattern, it was also observed
that there is no significant change in crystallite size for the initial
doping of Cu (2%), while a decrease in crystallinity was observed
with doping percentage. The suppression of crystallite size is due
to distortion in the host ZnO lattice because of the introduction of
a foreign impurity i.e. by Cu doping. Moreover, the nucleation and
also growth is decreased by Cu-doping due to the size difference of
Zn and Cu ions. The lattice distortion by Cu doping enhance the
strain moderately. The further increase of Cu increases the crys-
tallite size due to the increase in crystallinity and decrease of
micro-strain.
Table 2 shows the variation of peak position (2θ), d-value, cell
parameters ‘a’ and ‘c’, c/a ratio, bond length (Ɩ) and volume (V) of
Zn0.96  xNi0.04CuxO nanoparticles. The lattice constants a and c are
calculated using the following formula used for hexagonal systems
[26]:
1 4 ⎛ h2 + hk + k 2 ⎞ l2
= ⎜ ⎟+
d2 3⎝ a2 ⎠ c2 (3)
where d is lattice spacing, h, k, l are miller indices. Lattice para-
meters ‘d’, ‘a’ and ‘c’ of Ni-doped ZnO is always higher than Ni, Cu
co-doped ZnO. The decrease in the lattice parameters with the
introduction of can be assigned to the substitution of Zn2 þ ions by
smaller Cu2 þ ions and their mismatch in ionic radii.
The Ni, Cu co-doping mechanism of ZnO can be described by
Cu-doping into Zn–Ni–O host lattice, Cu2 þ ions are substituting
the Zn2 þ sites [27,28]. The observed constant c/a ratio revealing
that there is no change in hexagonal wurtzite structure by Cu
value.
(1)
Bond length (Ɩ) and volume (V) are calculated manually from
cell parameters by using Eqs. (4) and (5). The Zn–O bond length
has been calculated using the relationship [29]
96 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100

Table 1 Fig. 3. The peaks corresponding to Ni, Cu, Zn and O are clearly
Full width half-maximum (β), average crystallite size (D) and micro-strain (ε) of observed at their normal energy. It is observed from Fig. 3b and c
Zn0.94  xNi0.04Cux0 (0 rx r 0.04) nanoparticles.
that the Cu/(Zn þNiþ Cu) ratio is derived to be 2.05%, 4.01% for
Samples FWHM, β Average crystallite size, Micro-strain, ε Cu¼ 2% and 4%, respectively. The estimated amount of Cu2 þ ions is
(deg) D (nm) (10  3) slightly greater than the expected value such as 2% and 4%. This
could be attributed due to the contribution arising from the cop-
Zn0.96Ni0.040 0.3931 21.3 1.630
per grid used for the measurement. The atomic percentage of Zn is
Zn0.94Ni0.04Cu0.020 0.4060 20.6 1.683
Zn0.92Ni0.04Cu0.040 0.3454 24.2 1.432 decreased from 47.48% to 41.06% where as O percentage is in-
creased from 49.29% to 55.18% when Cu is increased from 2% to 4%.
The observed higher oxygen % at Cu¼4% represents the formation
⎛ a2 ⎛ 1 ⎞2 ⎞ of Zn–O; also encouraged Ni/Cu–O phases at higher Cu con-
Bond length (l) = ⎜⎜ + ⎜ − u⎟ c 2⎟⎟ centration. However, the lower Zn% at Cu ¼4% is due to the dis-
⎝3 ⎝2 ⎠ ⎠
(4) solution of Zn.
a2
where u = + 0.25 is the potential parameter of the hexagonal 3.4. UV–visible spectra – optical properties
3c 2
structure. The volume of unit cell of hexagonal system has been
calculated from the equation [29] UV–visible optical absorption spectra of the undoped and Cu
and Zn0.96Ni0.04O nanoparticles have been carried out for different
Volume (V ) = 0.866 × a2 ×c (5)
Cu concentration from 0% to 4% at room temperature using UV–
The behavior of volume of the unit cell and Zn–O bond length visible spectrometer between 315 nm and 700 nm and are shown
are directly related to the lattice parameters and hence possesses in Fig. 4. The inset of Fig. 4 shows the enlarged picture of ab-
the behavior similar to them. The iono-covalent bonds in solid sorption changes from 315 nm to 370 nm. It is noticed from Fig. 4
state lattice of ZnO are formed from 3d Zn and 2p oxygen orbitals. that the absorption of Ni-doped ZnO is always higher than Ni, Cu
When copper is substituted to Zn, the Cu–O bond is formed from co-doped ZnO. The initial increase of Cu concentration decreases
3d copper and 2p oxygen orbitals which reduce the bond length. the absorption abruptly from Cu ¼0–2% where Cu ¼2% doped
sample has minimum absorption. At lower Cu concentration, the
3.2. SEM-morphology analysis nucleus gets sufficient time to grow. Due to the slowness of the
reactions the formation of nucleation centers and hence the ab-
The surface morphologies of Ni doped ZnO and Ni, Cu co-doped sorption is low. At higher Cu concentration beyond 2%, the ab-
ZnO are characterized by SEM and presented in Fig. 2. Fig. 2a and b sorption found to increase. The increasing nucleation centers and
shows the surface morphology of Zn0.96Ni0.04O nanoparticles de-generating states and vacancies are responsible for the increase
which are quasi-spherical and agglomerated. A more uniform of absorption.
grain size and dense surface structure are observed from Fig. 2b. The typical room temperature transmittance spectra of
Slightly degraded surface structure is seen from Fig. 2c and d Zn0.96  xNi0.04CuxO (x ¼0.00, 0.02 and 0.04) nanoparticles between
which represents the surface morphology of Zn0.94Ni0.04Cu0.02O 315 nm and 700 nm are shown in Fig. 5. The transmission spectra
nanoparticles. It is observed from Fig. 2d that the size of the par- of the synthesized nanoparticles show just opposite trend of the
ticle get reduced and the particles are agglomerated each other optical absorption spectra. All the samples have semi-transparent
with other grain size around 10–30 nm. The present degradation is in the visible region. The better transmittance observed in the
mainly because of a decrease in nucleation and subsequent growth visible region among all the samples represent the good optical
rate due to increasing Cu2 þ doping percentage. quality of the crystals with low scattering or absorption losses.
The mixture of spheroid-like, rod-like and the cluster form of Moreover, the higher transmittance (more than 65%) noticed at
uneven surface morphology are observed from Fig. 2e and f which Cu¼ 2% leads to the industrial applications especially as transpar-
represent the surface morphology of Zn0.92Ni0.04Cu0.04O nano- ent electrode.
particles. Fig. 2f shows the well agglomerated and rod-like particle According to Tauc rule in the high absorbance region, the op-
structure (dominated) with size around 10–38 nm. The existence tical band gap Eg could be estimated by the following equations by
of more defects greatly de-generates and enhances the particle means of fitting between the contribution (αhυ)2 and hυ of the
size. incident photon [30]:

3.3. Energy dispersive X-ray (EDX) – compositional characterization
Analysis by EDX was performed to estimate the presence of Cu
content in Zn0.96  xNi0.04CuxO (0 r xr0.04) nanoparticles. The
existence of Ni and Cu in ZnO was confirmed by EDX and the re-
sults are displayed in Fig. 3. The atomic percentage of the com-
positional elements in Ni–ZnO and Ni–Cu–ZnO is given as inset of
(
ahυ = A hυ − Eg )n (6)
where A as a constant, Eg is optical band gap of the material and
the exponent n depends upon the type of transition. The values of
n for direct allowed, indirect allowed and direct forbidden are 1/2,
2, 3/2. In the present case, n is taken as 1/2. For a direct transition,
the square form of αhυ is chosen, since it gives the best linear
fitting curve in the band edge region. The relationship between

Table 2
Peak position (2θ), d-value, cell parameters, c/a ratio, bond length (Ɩ) and volume (V) of Zn0.94  xNi0.04Cux0 (0 r xr 0.04) nanoparticles.

Samples Peak position, 2θ (deg) d-value (Ǻ) Cell parameters (Ǻ) c/a ratio Bond length, Ɩ (Ǻ) Volume, V (Ǻ)3

a¼ b c

Zn0.96Ni0.040 36.29 2.4733 3.2473 5.1998 1.601 1.9759 47.48

Zn0.94Ni0.04Cu0.020 36.39 2.4670 3.2374 5.1864 1.602 1.9702 47.07
Zn0.92Ni0.04Cu0.040 36.38 2.4672 3.2377 5.1858 1.602 1.9703 47.08
 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
Fig. 2. SEM images of (a), (b) Zn0.96Ni0.04O, (c), (d) Zn0.94Ni0.04Cu0.02O and (e), (f) Zn0.92Ni0.04Cu0.04O nanoparticles.
(αhυ)2 and hυ is shown in Fig. 6. The values of Eg are obtained by
extrapolating the linear region near the onset in a plot of (αhυ)2
versus hυ.
The energy gap (Eg) of Zn0.96  xNi0.04CuxO nanoparticles is
steeply increased from Cu ¼0% (3.65 eV) to Cu ¼2% (3.83 eV, ΔEg
¼0.18 eV). The present blue-shift of Eg is due the increase of carrier
density generated by Cu substitution which is supported by XRD
and EDX spectra. The increase of carrier density shifts the Fermi
level close to conduction band and thus energy gap increases
(Burstein–Moss effect) [31]. It is also in good agreement to the
quantum confinement effect of the nanoparticles [32]. Addition of
Cu after 2% decreases the Eg. Diouri et al. [33] explained this red-
shift by the p–d spin-exchange interactions between the band
electrons and the localized d electrons of Cu2 þ ion. In our samples,
when Cu and Ni enter into Zn–O lattice, the s–d and p–d exchange
interactions between the band electrons of ZnO and localized d
electrons of Cu and Ni give rise to change in the energy band
structure and leading to reduction in the band gap.
3.5. Fourier transform infrared (FTIR) analysis
The band positions and numbers of absorption peaks are de-
pending on crystalline structure, chemical composition and mor-
phology. FTIR is one of the important techniques used to obtain
information about the chemical bonding in a material. It is used to
identify the elemental constituents of a material. The character-
istic peaks exhibited by FTIR spectra of Zn0.96  xNi0.04CuxO nano-
particles synthesized with x ¼0%, 2% and 4% are shown in Fig. 7a.
The broad absorption band around 3400–3600 cm  1 corresponds
to the O–H stretching vibrations of water present in Zn–O [34]. The
absorption bands around 1600 cm  1 arises due to the bending
vibration of O–H [35]. A sharp absorption peak between at
2336 cm  1 is because of an existence of CO2 molecule in air [36].
The absorption peaks around 1370 cm  1 are due to the stretching
vibration of carboxyl group (C ¼O) [37].
Evaluation of IR peaks below 800 cm  1 is interesting to study
Zn–O/Ni–O/Cu–O/Zn–O–Cu bonds. In order to study their oxide
bonds, IR characteristic spectra between the wave number
400 cm  1 and 800 cm  1 are focussed and shown in Fig. 7b. The
absorption band around 770 cm  1 is associated with Cu2 þ /Ni2 þ
97
occupation at Zn2 þ sites. The change in intensity and FWHM
corresponding to the frequency around 760–796 cm  1 represents
a change in density of defect states surrounding to Cu/Ni ions in
Zn–O. The vibrational bands observed from 400 to 600 cm  1 are
the characteristics of stretching vibrational modes of ZnO [38]. The
stretching mode of Zn–O bond is assigned to 559 cm  1 for un-
doped Zn0.96Ni0.04O. When Cu is introduced into Zn–Ni–O
(Cu ¼2%), the broad band corresponding to Zn–O bond is shifted to
lower side. In addition to that band intensity is also decreased.
Further increase of Cu (Cu ¼4%) increase the IR peak intensity and
also shift towards the higher side. The intrinsic host lattice defects
and activated impurities cause this shift in the vibrational mode at
higher Cu concentration.
3.6. Photoluminescence (PL) spectra
Generally, the densities of defects and oxygen vacancies affect
significantly the optical properties of oxide nanostructures. ZnO is
a well known optoelectronic material. However, the luminescence
properties of Cu and Ni doped ZnO have rarely been studied. The
correlation between structure and property is investigated by
photoluminescence (PL) spectra which originate from the re-
combination of the surface states [39]. The presence of defects
such as electron hole recombination centers and vacancies are
responsible for optical absorption and emission spectra.
Fig. 8 shows the room temperature PL spectra of Ni-doped ZnO
and Ni, Cu co-doped ZnO nanoparticles with different Cu doping
concentrations. Zn0.96Ni0.04O sample exhibit a strong near band
(NB) ultra-violet (UV) emission centered at 345 nm (3.6 eV). Even
though there is no considerable shift in UV emission intensity is
modified by Cu concentration where Cu¼ 2% doped sample have
higher intensity. The strong UV emission is originated from the
excitonic recombination corresponding to the band edge emission
of ZnO. The presence of UV emission at room temperature de-
monstrates the high crystal quality of the Cu and Ni codoped ZnO
crystals. The steady decrease of intensity and the broadening of UV
emission at higher Cu concentration (4%) are due to the change in
the crystal size and quality [40].
Electrons in the valence band are excited to the conduction
band by absorbing a photon and then the electronic transition
98 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
Fig. 3. Energy dispersive X-ray (EDX) spectra of (a)
Zn0.94Ni0.04Cu0.02O and (c) Zn0.92Ni0.04Cu0.04O nanoparticles.
Fig. 4. UV–visible absorption spectra of Zn0.96  xNi0.04CuxO (x ¼ 0, 0.02 and 0.04)
nanoparticles with different Cu concentrations from 0% to 4%.
Fig. 5. Transmittance spectra of Zn0.96  xNi0.04CuxO (x ¼ 0, 0.02 and 0.04) nano-
particles with different Cu concentrations from 0% to 4%.
Fig. 6. The (αhυ)2 versus hυ curves of Zn0.96  xNi0.04CuxO (x ¼0, 0.02 and 0.04)
nanoparticles with different Cu concentrations from 0% to 4% for the optical energy
gap calculation.
from conduction band to valence band gives rise to the emission at
394 nm in Zn0.96Ni0.04O sample. The difference between absorp-
tion energy and emission energy may arise from electron–phonon
couplings, lattice distortions, and localization of charge carriers
[41]. When Cu ions are added into ZnO lattice, they become donors
Zn0.96Ni0.04O, (b) and multi-emission centers are formed which makes the emission
of the donors to the valence band and hence the intrinsic transi-
tion of Cu2 þ ions. The blue-shift of violet emission from 394 nm
(Cu ¼0%) to 383 nm (Cu ¼4%) and a band-edge modification are
also observed by Cu-doping.
Zn0.96  xNi0.04CuxO nanoparticles exhibit a moderate defect re-
lated blue band (BB) emission between 477 nm and 481 nm. The
blue emission might come from the intrinsic defects and donor–
acceptor pair recombination involving Cu impurity acceptor [42].
Zn interstitials and oxygen vacancies were main donor defects
while Zn vacancies and oxygen interstitials were main acceptor
defects in intrinsic ZnO [43]. Kohan et al. [44] have calculated the
energy levels of defects in ZnO by the full-potential linear muffin-
tin orbit method and they have shown that the energy interval
from the interstitial Zn levels to Zn vacancies was about 2.57 eV
which were well consistent with the energy of the blue emission
around 477–481 nm. Therefore, blue emission peak is assigned to
the electron transition from the interstitial Zn levels to Zn va-
cancies, respectively. The doping of Cu in the present system is
useful to tune the emission wavelength and hence acting as the
important candidates for the optoelectronic device applications.
 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
Fig. 7. (a) FTIR spectra of Zn0.96  xNi0.04CuxO (x¼ 0, 0.02 and 0.04) nanoparticles
with different Cu concentrations from 0% to 4% at room temperature in the wave
number from 400 cm  1 to 4000 cm  1 at room temperature. (b) The enlarged FTIR
spectra in the wave number range 400–800 cm  1 at room temperature.
Fig. 8. The room temperature PL spectra of Zn0.96  xNi0.04CuxO (x¼ 0, 0.02 and 0.04)
nanoparticles with different Cu concentrations from 0% to 4% as a function of
wavelength between 320 nm and 520 nm.
3.7. Magnetic studies
The magnetization versus magnetic field (M–H) curve mea-
sured from  6000 Oe to 6000 Oe at room temperature for Ni-
doped and Ni, Cu co-doped ZnO nanoparticles are carried out by
VSM. The hysteresis curves (Fig. 9) exhibits the ferromagnetic
nature at room temperature. The room temperature ferro-
magnetism (RTFM) in pure ZnO is due to the Zn vacancies induced
spin polarization. RTFM observed in TM-doped ZnO nanoparticles
is originated from two factors; one is the increase of the number of
defects and oxygen vacancies; the other is the exchange interac-
tions between the TM ions and the O ion spin moment [45]. The
existence of ferromagnetism in Zn0.96Ni0.04O may be due to the
clustering of metallic Ni and/or due to intrinsic ferromagnetism
from the charge carriers [46]. XRD results clearly indicate the
99
Fig. 9. Magnetization versus magnetic field (M–H) curves for Zn0.96  xNi0.04CuxO
(x ¼0.00, 0.02 and 0.04) nanoparticles at room temperature.
absence of metallic Ni phase in the sample therefore the observed
ferromagnetism at room temperature is an intrinsic property Ni-
doped ZnO. It is believed that the existence of ferromagnetism in
ZnO:Ni could be due to the exchange coupling between localized
‘d’ spins on the Ni ions mediated by free delocalized carriers [47].
Saturation magnetization (Ms) of Ni, Cu co-doped ZnO samples
have different values which indicating that room temperature
ferromagnetism is induced by the introduction of Ni and Cu. It is
worthy to note that Cu ¼2% doped sample has a large Ms which
can be attributed to the enhanced ferromagnetic coupling. The
ferromagnetic behavior of this sample can be attributed to the
presence of small magnetic dipoles located at the surface of na-
nocrystals, which interacts with their nearest neighbors inside the
crystal. Consequently, the interchange energy in these magnetic
dipoles making other neighboring dipoles oriented in the same
direction. In nanocrystals, surface to volume ratio increases
(crystallite size of this sample (20.6 nm) is smaller than others), so
the population of magnetic dipoles oriented in the same direction
will increase at the surface. Thus, the sum of the total amount of
dipoles oriented along the same direction will increase subse-
quently. In short the crystal surface will be usually more magne-
tically oriented.
Moreover, the simultaneous doping of Ni and Cu into ZnO in-
creases carrier (hole) concentration of these materials. Free car-
riers play an important role in establishing the magnetic phase
and hence the ferromagnetism in ZnO-based DMSs [5]. According
to the RKKY exchange mechanism, the ferromagnetic interaction
in DMS materials originates from the exchange interactions be-
tween free delocalized carriers (holes from valence band) and the
localized d spins of Ni and Cu ions local [48]. The presence of free
carriers by simultaneous substitution of Ni and Cu is suggested as
the main cause for the enhanced ferromagnetic properties. When
Cu doping is increased after 2% the ferromagnetic behavior is
suppressed. The enhanced antiferromagnetic interaction between
neighboring Cu–Cu ions suppressed the ferromagnetism at higher
doping concentrations of Cu.
4. Conclusions
Followings are the conclusions drawn from the present studies:

Cu-doped and Ni, Cu co-doped ZnO nanoparticles have been
synthesized successfully for different Cu concentrations be-
tween 0% and 4% using sol–gel method.

XRD studies confirmed the hexagonal structure of the prepared
samples.

The formation of secondary phases such as CuO along (111)
plane and Zn along (101) plane at higher Cu content (4%) was
due un-reacted Cu2 þ and Zn2 þ ions present in the solution
100 D. Theyvaraju, S. Muthukumaran / Physica E 74 (2015) 93–100
which reduces the interaction between precursor ions and
surfaces of ZnO.

The nominal elemental composition of Zn, Cu, Ni and O was
confirmed by EDX studies.

Even though energy gap was increased (blue-shift) from
Cu¼ 0–2% by quantum size effect, the s–d and p–d exchange
interactions between the band electrons of ZnO and localized d
electrons of Cu and Ni led to decrease (red-shift) the energy gap
at Cu ¼4%.

Presence of Zn–O and Zn–Cu–Ni–O bonds was confirmed by
FTIR analysis.

The change in UV emission and blue emissions were discussed
based on defect formation induced by Cu concentration.

Structural and optical properties of the samples concluded that
Cu-doping is useful to tune the emission wavelength and hence
acting as the important candidates for the optoelectronic de-
vice applications.

The enhanced ferromagnetic ordering at Cu ¼2% sample is due
to the presence of free charge carriers where as the anti-
ferromagnetic interaction between neighboring Cu–Cu ions
suppressed the ferromagnetism at higher Cu concentration
(Cu ¼4%).
References
[1] I. Malajovich, J.J. Berry, N. Samarth, D.D. Awshalom, Nature 411 (2001)
770–772.
[2] J.K. Furdyna, J. Appl. Phys. 64 (1988) R29.
[3] S.J. Pearton, C.R. Abernathy, M.E. Overberg, G.T. Thaler, A.F. Hebard, Y.D. Park,
F. Ren, J. Kim, L.A. Boatner, J. Appl. Phys. 93 (2003) 1.
[4] S.A. Wolf, D.D. Awschalom, R.A. Buhrman, J.M. Daughton, S. von Molnar, M.
L. Roukes, A.Y. Chtchelkanova, D.M. Treger, Science 294 (2001) 1488–1495.
[5] K.R. Kittilstved, W.K. Liu, D.R. Gamelin, Nat. Mater. 5 (2006) 291–297.
[6] C. Zapata, M. Khalid, G. Simonelli, M. Villafuerte, S.P. Heluani, P. Esquinazi,
Appl. Phys. Lett. 99 (2011) 112503.
[7] V. Bhosle, J. Narayan, Appl. Phys. Lett. 93 (2008) 021912.
[8] Z.B. Xia, Y.W. Wang, Y.J. Fang, Y.T. Wan, W.W. Xia, J. Sha, J. Phys. Chem. C 115
(2011) 14576–14582.
[9] Y.S. Wang, P.J. Thomas, P.O. Brien, J. Phys. Chem. B 110 (2006) 21412–21415.
[10] C.W. Cheng, G.Y. Xu, H.G. Zhang, Y. Luo, Mater. Lett. 62 (2008) 1617–1620.
[11] S. Ghosh, P. Srivastava, B. Pandey, M. Saurav, P. Bharadwaj, D.K. Avasthi,
D. Kabiraj, S.M. Shiva prasad, Appl. Phys. A 90 (2008) 765–769.
[12] E. Liu, P. Xiao, J.S. Chen, B.C. Lim, L. Li, Curr. Appl. Phys. 8 (2008) 408–411.
[13] Z.B. Bahsi, A.Y. Oral, Opt. Mater. 29 (2007) 672–678.
[14] P.K. Sharma, R.K. Dutta, A.C. Pandey, J. Magn. Magn. Mater. 321 (2009)
4001–4005.
[15] A. Jagannatha Reddy, M.K. Kokila, H. Nagabhushan, J. Alloy. Compd. 509 (2011)
5349–5355.
[16] Y. Wei, D. Hou, S. Qiao, C. Zhen, G. Tang, Physica B 404 (2009) 2486–2488.
[17] G. Tang, X. Shi, C. Huo, Z. Wang, Ceram. Int. 39 (2013) 4825–4829.
[18] M.S. Niasari, F. Davar, A. Khansari, J. Alloy. Compd. 509 (2011) 61–65.
[19] J. Yang, L. Feia, H. Liua, Y. Liu, M. Gaoa, Y. Zhanga, L. Yanga, J. Alloy. Compd. 509
(2011) 3672–3676.
[20] S. Muthukumaran, R. Gopalakrishnan, Opt. Mater. 34 (2012) 1946–1953.
[21] R. Savu, R. Parra, E. Joanni, B. Jancar, S.A. Elizario, R. de Camargo, P.R. Bueno, J.
A. Varela, E. Longo, M.A. Zaghate, J. Cryst. Growth 311 (2009) 4102–4108.
[22] J. Yang, L. Fei, H. Liu, Y. Liu, M. Gao, Y. Zhang, L. Yang, J. Alloy. Compd. 509
(2010) 3672–3676.
[23] Q. Ma, D.B. Buchholz, R.P.H. Chang, Phys. Rev. B 78 (2008) 214429.
[24] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751–767.
[25] S. Sivaselvan, S. Muthukumaran, M. Ashokkumar, Opt. Mater. 36 (2014)
797–803.
[26] O. Lupan, L. Chow, G. Chaic, B. Roldan, A. Naitabdi, A. Schulte, H. Heinrich,
Mater. Sci. Eng. B 145 (2007) 57–66.
[27] T.S. Herng, S.P. Lau, S.F. Yu, H.Y. Yang, L. Wang, M. Tanemura, J.S. Chen, Appl.
Phys. Lett. 90 (2007) 032509.
[28] O. Lupan, T. Pauporté, L. Chow, B. Viana, F. Pellé, B. Roldan Cuenya, L.K. Ono,
H. Heinrich, Appl. Surf. Sci. 256 (2010) 1895–1907.
[29] D. Anbuselvan, S. Muthukumaran, M. Ashokkumar, J. Mater. Sci.: Mater.
Electron. 25 (2014) 2004–2015.
[30] S. Anandan, S. Muthukumaran, M. Ashokkumar, Superlattices Microstruct. 74
(2014) 247–260.
[31] S. Suwanboon, T. Ratana, W.T. Ratana, J. Sci. Technol. 4 (2007) 111–121.
[32] T. Takagahara, K. Takeda, Phys. Rev. B 46 (1992) 15578.
[33] J. Diouri, J.P. Lascaray, M.E. Amrani, Phys. Rev. B 31 (1985) 7995.
[34] M. Ashokkumar, S. Muthukumaran, Opt. Mater. 37 (2014) 671–678.
[35] M. Ashokkumar, S. Muthukumaran, Superlattices Microstruct. 69 (2014)
53–64.
[36] M. Arshad, A. Azam, A.S. Ahmea, S. Mollah, A.H. Naqvi, J. Alloy. Compd. 509
(2011) 8378–8381.
[37] S. Senthilkumar, K. Rajendran, S. Banerjee, T.K. Chini, V. Sengodan, Mater. Sci.
Semicond. Process. 11 (2008) 6–12.
[38] T. Long, S. Yin, K. Takabatake, P. Zhnag, T. Sato, Nanoscale Res. Lett. 4 (2009)
247–253.
[39] N. Chestony, T.D. Harris, R. Hull, L.E. Brus, J. Phys. Chem. 90 (1986) 3393–3399.
[40] T. Yamamoto, H. Katayama-Yoshida, Jpn. J. Appl. Phys. 38 (1999) L166–L169.
[41] C.L. Yang, J.N. Wang, W.K. Ge, L. Guo, S.H. Yang, D.Z. Shen, J. Appl. Phys. 90
(2001) 4489.
[42] S.J. Pearton, D.P. Norton, K. Ip, Y.W. Heo, T. Steiner, Prog. Mater. Sci. 50 (2005)
293–340.
[43] S.B. Zhang, S.H. Wei, A. Zunger, Phys. Rev. B 63 (2001) 075205.
[44] A.F. Kohan, G. Ceder, D. Morgan, Phys. Rev. D 61 (2000) 15027.
[45] B. Babu, G.R. Sundari, K. Ravindranadh, M.R. Yadav, R.V.S.S.N. Ravikumar, J.
Magn. Magn. Mater. 372 (2014) 79–85.
[46] B. Pandey, S. Ghosh, P. Srivastava, D.K. Avasthi, D. Kabiraj, J.C. Pivin, J. Magn.
Magn. Mater. 320 (2008) 3347–3351.
[47] M. El-Hilo, A.A. Dakhel, A.Y. Ali-Mohamed, J. Magn. Magn. Mater. 321 (2009)
2279–2283.
[48] J.D. Priour, E.H. Hwang, S.D. Sarma, Phys. Rev. Lett. 92 (2004) 117201.