Materials Science in Semiconductor Processing 66 (2017) 149–156
Contents lists available at ScienceDirect
Materials Science in Semiconductor Processing journal homepage: www.elsevier.com/locate/mssp
Effect of calcination temperature on Cu doped NiO nanoparticles prepared via wet-chemical method: Structural, optical and morphological studies K. Varunkumara, Rafikul Hussaina, Gurumurthy Hegdeb, Anita S. Ethiraja, a b
MARK
⁎
Centre for Nanotechnology Research, VIT University, Vellore 632014, Tamil Nadu, India BMS R & D, BMS College of Engineering, Basavanagudi, Bengaluru 560019, India
A R T I C L E I N F O
A B S T R A C T
Keywords: Nanoparticles Chemical method NiO Raman Calcination Tauc plot
In the present study, NiO and Cu-doped NiO nanoparticles were successfully synthesized by wet chemical method at room temperature using sodium hydroxide (NaOH) as precipitating agent. The as-prepared Cu-doped NiO powder samples were subjected to three different calcination temperatures such as, 350 °C, 450 °C and 550 °C in order to investigate the impact of calcined temperatures on the phase formation, particle size and band gap evolution. The phase formation and crystal structure information of the prepared nanomaterials were examined by X-ray powder diffraction (XRD). XRD revealed the face-centered cubic (FCC) structure. Average crystalline size of pure and doped samples estimated using Scherer formula was found to be 15 nm and 9 nm respectively. With increase in the calcination temperature from 350 °C to 550 °C for the Cu doped NiO samples the particle size of the nanoparticles was found to increase from 4 nm to 9 nm respectively. The optical study for both pure and doped NiO nanoparticles was performed using an UV–Vis spectrophotometer in the wavelength range of 200–800 nm. The strong absorption in the UV region confirms the band gap absorption in NiO and was estimated from the UV–Vis diffuse reflectance spectra via Tauc plot. Systematic studies were also carried out to study the effect of calcination on the optical transmittance. Samples were also investigated using Raman and Fourier Transform Infrared Spectroscopy (FTIR). Furthermore, morphology of the pure NiO and Cu-doped NiO Nanoparticles were examined by scanning electron microscope (SEM).
1. Introduction Nanometer sized nickel oxide materials has received great attention in research and various application purposes. Nickel oxide (NiO), a ptype oxide semiconductor is an important transition metal oxide which exhibits a wide band gap of 3.6–4.0 eV [1]. The nanomaterial system also serves as an alternative material for energy applications, such as ptype semiconductor material for solar cells, electrochemical capacitors, photocatalysts, smart windows and organic light emitting diodes [2,3]. However, high resistivity of NiO has always been a concern in many applications. To overcome this limitation, doping of NiO with monovalent impurities, such as copper (Cu) and lithium (Li) has been done. This kind of doping helps to improve the properties of NiO making it suitable in a variety of applications. For instance, enhanced electrochromic behaviour leads to better smart windows and good catalysis. NiO nanoparticles (NPs) can be synthesized by different techniques, such as hydrothermal, microemulsion, solvothermal, chemical precipitation, sol-gel, thermal decomposition, combustion and microwave irradiation [4–10]. Using solvothermal technique, Anandan group
⁎
obtained different morphologies of NiO NPs by using single precursor (Nickel acetate) in different solvents. The obtained nanoparticles were found to be around 25–30 nm [11]. Another work showed that Ni(OH)2 precursor was first prepared by using microwave assisted hydrothermal technique and subsequently thermal decomposition was used to prepare NiO NPs [12]. NiO NPs prepared by using sol-gel method at pH 11 and calcination temperature 450 °C, revealed amorphous phase before calcination and crystalline phase after calcination (~32.9 nm average diameter) in XRD results [13]. Annealing study of NiO carried out by Kisan B et al. showed no changes in XRD peak upto 300 °C, whereafter, the peak width decreased. According to them, the annealing process not only increased the average crystallite size, but dislocations/defects were also reduced with increase in the calcinations temperature [14]. Nadeem et al. reported structural and magnetic study of NiO with respect to annealing temperature from 400 °C–800 °C. Results showed that at lower annealing temperature Ni phase was dominant and oxidation did not occur to form NiO. Moreover the particles were found to remain separated. At higher temperature, the particles became agglomerated and XRD data revealed
Corresponding author. E-mail address:
[email protected] (A.S. Ethiraj).
http://dx.doi.org/10.1016/j.mssp.2017.04.009 Received 13 November 2016; Received in revised form 20 March 2017; Accepted 5 April 2017 Available online 26 April 2017 1369-8001/ © 2017 Elsevier Ltd. All rights reserved.
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
rate of 75 mAg−1 [33]. In addition capacity retention of ~84% (784 mAh g−1) was observed even at the 100th cycle. In this work, we have synthesized pure NiO and Cu doped NiO NPs (NiO:Cu) by wet chemical method successfully. Although researchers have used NiO and Cu doped NiO films prepared by other methods for different studies, such as physical, optical, electrochemical, magnetic; [34–37] no proper investigation has been done on the effect of calcination temperature on NiO NPs and NiO:Cu NPs. In the present work, a systematic study has been done on the effect of calcination temperature on the structural, morphological and optical properties of NiO and Cu doped NiO nanoparticles.
increase in NiO phase with temperature [15]. NiO synthesized using chemical precipitation method indicated blue shift in optical study with increase in calcination temperatures. Thermogravimetric study revealed weight loss of precursor at calcinations temperatures of 50°C350°C and complete decomposition of precursor to form NiO at 400 °C [16]. NiO/RGO composites prepared by hydrothermal method and treated at different calcinations temperatures (250 °C, 300 °C, 400 °C and 500 °C) showed that the intensity of diffraction peak in XRD increased with temperature. The particles size of NiO in NiO/RGO was 2.62, 3.60, 5.75 and 9.94 nm at 250 °C, 300 °C, 400 °C and 500 °C, respectively. Additionally, lower temperature calcined composites exhibited enhanced electrochemical performance due to small crystallite size and high surface area [6]. Research work was also carried out to prepare 1-D NiO-RuO2carbon composite nanofibers using thermal assisted electrospinning technique [17]. The fabricated composite nanofibers were further utilized in the application of asymmetric supercapacitors and Li-cyclic studies. Preparation of vertically aligned single crystalline NiO nanowalls on a Ni foil using simple and efficient method of plasma assisted oxidation method was reported by Sow et al. group [18]. They have demonstrated the application of these NiO nanowall films in electron field emission and Li-ion battery application (turn-on field of 7.4 V/µm, maximum current density of ~160 µA/cm2). G. R. Balakrishna group [19] utilized gel combustion technique synthesized NiO nanoparticles using various weight ratios of oxidizer (O), nickel nitrate hexahydrate and fuel (F) such as cassava starch respectively. Their result showed that when the oxidizer to fuel ratio is 1:1, the photocatalytic degradation of MB dye performed in sunlight exhibited higher degradation efficiency (94%) as compared to UV light. The prepared nanoparticles were tested for energy storage where electrochemical studies depicted a reversible capacity of 940 and 785 mA h/g when O:F ratio was taken as 1:0.5 and 1:1 respectively. Antibacterial activity was also performed against a fungal strain and two bacterial strains. A few researchers synthesized NiO and Cu doped NiO NPs by using chemical co-precipitation technique. The reported Ni1−xCuxO (6.52–13.4 nm) exhibited decreased particle size with increasing doping concentration. They showed that doping at lower concentration was suitable electrode material for super capacitor application [20,21]. Cu doped ZnO co-doped Ni (Zn 0.96−X Ni 0.04 CuX O) NPs synthesized by solgel method showed well agglomerated rod like morphology [22]. In literature, un-doped, Fe, Mn and Zn doped NiO NPs synthesis by chemical co-precipitation method were also documented [9,23,24]. Recently researchers have also reported the application of electrospun NiO-SnO2 nanofibers in the electrical and humidity sensors [25]. In addition to NiO, few other Binary Oxides MO (M=Co, Fe, Cu, Mn) systems of significant interest is also reported in literature. M. V. Reddy et al. [26] utilized urea combustion method nitridation and cabothermal reduction methods to prepare Co3O, CoN and CoO respectively. Their results of CV and galvanostatic cyclic tests indicated that the prepared materials showed negligible capacity fading. Also CoN proved to be the best anode material with higher theoretical capacity of 1100 mAh g−1 due to formation of Co metal and Li3N.The same group also reported nano/submicrometer and micrometer sized CuO prepared by molten salt method at different temperatures for Liion battery (LIB) applications. They concluded that at the end of 40th cycle the CuO prepared at 750 °C exhibited high and stable capacity of ~620 mAh g−1 [27]. Other application of energy storage studies on CuO is also well documented [28–31]. Formation of exfoliated graphene oxide (EG)/Iron (II) oxide FeO composite using a novel and new graphenothermal reduction process is also reported in literature [32]. This one pot synthesis process is eco-friendly, scalable and the synthesized composite material proved to be a suitable alternate anode material for LIB's. The capacity and cyclic stability was much higher when compared with their individual counterpart. The same technique was utilized even to prepare EG/MnO Manganese (II) oxide composite which exhibited a high reversible capacity of 936 mAh g−1at a current
2. Experimental details 2.1. Synthesis of NiO and Cu doped NiO nanoparticles All chemicals used in this experimental work were of analytical reagent grades and used without any further purification. NiO and NiO:Cu NPs were prepared by chemical co-precipitation method [8]. NiCl2·6H2O (2.37 g) and CuSO4·5H2O (2.49 g) were used as the sources of Ni and Cu, respectively. For pure NiO NPs synthesis, NiCl2·6H2O and NaOH(0.39 g) were taken in desired molarity of 0.1 M. Then aqueous solution of NiCl2·6H2O was stirred for 20 min. The pH value was adjusted to 10 by using NaOH solution. Once pH value of solution has attained 10, allow this solution to stir for 4 h. Samples were washed by DI water several times and dried for 8 h at 80 °C in hot air oven. Finally, dried powders were calcined at 550 °C to obtain NiO NPs. Cu doped NiO samples were prepared using the same procedure as described for NiO except desired quantity of CuSO4·5H2O solution were added into NiCl2·6H2O to prepare 4% doping samples. As dried NiO:Cu powder samples were calcined for three different temperatures 350 °C, 450 °C and 550 °C respectively. 2.2. Sample characterization The crystallinity and phase formation of NiO and NiO:Cu NPs were identified by using a Bruker X ray diffractometer with Cu Kα (1.54 Å) radiation source over a range of 2θ from 20° to 80°. Optical properties were analyzed by Specord 210 plus UV–visible absorption spectra instruments in the range of 200–800 nm. Transmittance and Diffuse reflectance study was performed by V-670 JAASCO. Surface morphology of the samples was analyzed by using Scanning Electron Microscope (SEM; FEI Quanta FEG 200–High Resolution Scanning Electron Microscope) coupled with energy dispersive X-ray spectroscopy (EDS). Atomic Absorption Spectroscopy (AAS) was utilized to estimate the amount of Cu doping in NiO. Raman spectra were measured at room temperature using 532 nm green laser beam (ENWaveoptronicsezRaman pro) while Fourier Transform Infrared (FTIR; IRAffinity-1 Shimadzu) were also utilized to characterize the prepared NPs. 3. Results and discussion 3.1. Optical properties The UV-Vis absorption spectra for the NiO and NiO:Cu samples recorded in the wavelength range of 200 nm to 800 nm is shown in Fig. 1. Both the samples showed strong absorption peaks in the UV region which is blue shifted from the absorption edge of bulk NiO [38]. Pure NiO exhibited the absorption peak at 344 nm while Cu doped NiO NPs showed peak at 323 nm (Fig. 1(a)). This lower wavelength shift or blue shift in the latter case is attributed to Burstein – Moss effect, which confirms the quantum confinement effect and Fig. 1(b) shows the absorption data for the NiO:Cu samples calcined at different temperatures viz. 350 °C, 450 °C and 550 °C. The absorption peaks in the UV region appeared at wavelengths of about 262 nm, 280 nm and 328 nm, 150
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
Fig. 1. UV visible absorbance spectra of NPs (a) NiO and NiO:Cu (b) NiO:Cu calcined at 350 °C, 450 °C and 550 °C.
photon energy (hv ) are related by expression as given below, where A is an energy independent constant, Eg is an optical bandgap of material and n denotes the type of transitions. Here n is equal to 1 indicates 2 direct transitions while n equal to 2 corresponds to indirect transitions. An extrapolation of the linear region of the plot (αhʋ)2 vs. Energy (hʋ) gives the optical bandgap value Eg [8,42,43].
respectively. These absorptions in the UV region are due to the band gap absorption in NiO semiconductor nanoparticles [8]. The obtained data for the NiO:Cu samples clearly shows that with increase in the calcination temperature there is an obvious red-shift to the visible region indicative of decreased bandgap and increased crystallite size. This work is in very well agreement with the literature report [39]. Using the AAS technique the amount of Cu incorporated in the NiO samples were found to be approximately 4 at%. Further the optical band gaps of synthesized nanoparticles have been calculated from the absorption spectrum using the expression of Tauc relation [40,41]. The absorption coefficient (α ) and incident
αhv = A (hv − Eg )n Fig. 2 shows Tauc plot for prepared nanoparticles before calcination while Fig. 3 represents the condition of after calcination. Bandgap energy of pure NiO and Cu doped NiO nanoparticles before calcination
Fig. 2. Tauc plot of NPs before calcination (a) Pure NiO (b) Cu doped NiO.
Fig. 3. Tauc plot of NPs after calcination (a) Pure NiO (b) Cu-doped NiO.
151
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
phonon (2 P) 2TO at 685.7 cm−1. The two magnon (2 M) bands at 1469.4 cm−1 and 1529.8 cm−1are due to a two-magnon (2 M) scattering. The sharp band at 493.5 cm−1 is attributed to a one-phonon (~440 cm−1) plus one-magnon (~50 cm−1) excitation [48]. The frequency and shape of the phonon bands did not vary with temperature, whereas the magnon scattering intensities were strongly temperature dependent. Therefore, for Cu incorporated samples they shifted to lower frequencies and decreased in intensity with increasing temperature, disappearing completely close to the Néel point (TN=523 K). The disorder-induced 1 P band at ~334.7 cm−1 has very small intensity indicating good quality of single-crystal [49]. The phonon related part of the Raman spectra (1 P and 2 P bands in Fig. 4) in nanosized NiO powders is similar to that in the single-crystal. However, the 1 P band became more pronounced in powders due to the presence of defects or surface effect. At the same time, the twomagnon (2 M) band experiences dramatic decrease of intensity in nanopowders, becoming undetectable for 100 nm crystallites size at RT [50]. The results clearly indicated that due to low concentration of dopant in the synthesized NiO nanoparticles no signal from Cu was observed. Our XRD data also supports the same. Hence, it did not bring about drastic changes in the fundamental phonon peaks of NiO.
Fig. 4. Raman spectra (λex=532 nm) of nickel oxide and NiO:Cu nanoparticles at room temperature, having Peaks due to one-phonon (1 P), two-phonon (2 P) and two-magnon (2 M) scattering. Samples S1(NiO nanoparticles), S2 (NiO:Cu nanoparticles), S3(NiO:Cu calcined at 350 °C) and S4(NiO:Cu calcined at 550 °C) respectively.
shows 3.94 eV and 3.97 eV whereas the same samples after calcination at 550 °C, the optical band gap value was found to be 3.22 eV and 3.24 eV respectively. In both the cases the optical bandgap value was found to be lower than the reported bulk value of NiO i.e. 4.0 eV [44]. Moreover, the Cu doped NiO samples exhibited higher bandgap than the pure NiO which indicated formation of smaller particle size due to quantum confinement effect [45]. This consequence can be due to chemical defects or vacancies nearby intergranular regions, creating new energy levels to reduce the bandgap [46]. The results clearly indicate that the process of doping increased the band gap of synthesized NiO samples while calcination process decreased the bandgap. Ponnusamy et al. [8] and Sharma et al. [23] also reported widening of NiO bandgap with Fe and Mn dopant. Compared to NiO (3.958 eV), Fe doping in NiO increased the bandgap to 4.026 eV and with increase in Mn concentration the bandgap increased from 3.79 eV to 3.95 eV. In addition to NiO, other systems like Mg doped ZnO [43] also shows increased bandgap of 3.47 eV when compared to ZnO (3.23 eV). The equivalent range of energy bandgap (Eg) materials can be utilized for the solar cell applications [47].
3.3. Transmittance analysis The optical transmittance spectra of NiO and NiO:Cu NPs recorded in the wavelength range from 300 nm to 800 nm are depicted in Fig. 5(a). It was observed that the transmittance of pure NiO NPs reduces after the incorporation of the Cu in NiO NPs. The obtained transmittance was found to be about 55% for pure NiO samples while 35% of transmittance was exhibited in case of NiO:Cu samples which could be attributed to light scattering phenomenon by large amount of grain boundaries present in the latter case. Our results obtained are much in consistent with the reported results of Chen group [51]. Chen et al.’s work was based on Cu doped NiO composite films deposited on the Corning 1737F glass substrates using RF magnetron sputtering technique. The film thickness was about 100 nm. Their experimental results indicated reduction in transmittance values from 96% to 43% with increase in the Cu concentration from 2.29 at% to 18.17 at%. Recently, similar observation of reduction in transmittance was also reported for potassium doped NiO films as compared to pure NiO prepared by spray pyrolysis technique [52]. The transmittance data procured for NiO:Cu NPs at different calcination temperature has been highlighted in Fig. 5(b). We noticed that with increase in the calcination temperature i.e. 350 °C, 450 °C and 550 °C the transmittance % was found to decrease as 65%, 40% and 32%, respectively. The possible reason which could be thought for the present observation may be due
3.2. Raman analysis The room temperature Raman spectra of NiO (S1), NiO:Cu (S2), NiO:Cu calcined at 350 °C (S3) and NiO:Cu calcined at 550 °C (S4) samples measured using 532 nm green laser beam exhibited multiple bands above 300 cm−1 (Fig. 4). Two vibrational bands could be seen: one-phonon (1 P) TO at 334.7 cm−1 and LO at 493.5 cm−1 modes, two-
Fig. 5. Transmittance spectra (a) NiO and NiO:Cu (b) NiO:Cu calcined at 350 °C, 450 °C and 550 °C.
152
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
Table 1 Information obtained from XRD patterns for pure NiO and Cu-doped NiO nanoparticles. Samples
hkl
2θ (deg)
d-spacing (nm)
Lattice parameter, a (nm)
Pure NiO
111 200 111 200
37.33 43.34 37.34 43.39
0.2406 0.2085 0.2405 0.2083
0.41673 0.41698 0.41662 0.41660
Cu-doped NiO
NiO without any peaks originating from impurity elements. XRD obtained for calcination temperature of 350 °C shows broad peak corresponding to small particle size. The intensities of the diffraction peaks were found to become intense and sharper with increase in the calcination temperature. This clearly indicated an increase in crystallite size as well as its crystallinity for doped samples. The average crystallite size of NiO:Cu NPs calcined at 350 °C, 450 °C and 550 °C, determined by the Debye Scherrer's formula were 4 nm, 6 nm and 9 nm respectively. The calcination results obtained matched very well with the literature report of Karthik et al. [57]. According to them, when NiO nanoparticles were subjected to calcination temperature in the range of 500 °C to 800 °C the particle size of the nanocrystallites were found to increase from 16 nm to 25 nm [57]. Chen et al.’s work on calcination of NiO/RGO composites also supports our data [6]. We have also calculated the variation of lattice parameter values for pure NiO and Cu doped NiO (NiO:Cu) samples. The variation of lattice constant is listed in Table 1. Our XRD data clearly shows no peaks corresponding to metallic Cu or CuO which may be due to the fact that Cu ions can be easily substituted for Ni ions as their ionic radii are very similar in dimension. The ionic radii of Cu and Ni are 0.082 nm and 0.078 nm respectively. Hence no significant lattice distortion was observed for Cu doped NiO nanoparticles and Pure NiO. This data is well in agreement with the literature report of Cu doped NiO thin films [58]. Variation of lattice parameters as obtained from the XRD data for NiO:Cu samples subjected to different calcination temperature is tabulated in Table 2. It was observed that with increase in calcination temperature there is slight decrease in lattice parameters. Typical SEM image of synthesized pure and doped NiO NPs are shown in Fig. 8. It can be seen that pure NiO NPs reveal homogenous spherical shaped particles (Fig. 8(a)) while the doped sample (Fig. 8(b)) shows comparatively small sized particles which are agglomerated. This result is in well agreement with the XRD data which showed formation of smaller particles in Cu doped NiO NPs. The average size of pure NiO NPs was found to be about 12 nm. According to the literature report, the value obtained using SEM is smaller than the value obtained using XRD technique [42]. Our data also matches with the same. The elemental composition of NiO samples was performed using EDS analysis. The sample indicated presence of Ni and O while NiO: Cu samples (Fig. 8(c)) confirmed the presence of Cu in addition to the elements present in pure NiO NPs. Further investigation of the morphology of calcined NiO:Cu samples at 350 °C, 450 °C and 550 °C using FESEM are shown in Fig. 9. The results clearly indicated morphological changes in doped samples with
Fig. 6. XRD pattern of nanoparticles (a) NiO:Cu (b) Pure NiO.
to oxygen deficiency. The process of calcination may eliminate or remove the interstitial oxygen and this effect may be more pronounced with increase in calcination temperature [53,54]. 3.4. Structural and morphological study Fig. 6 highlights the X-ray diffraction pattern of prepared pure and doped NiO samples obtained from the NiCl2·6H2O precursor. The diffraction peak positions appearing at 2θ=37.2°, 43.3°, 62.7°, 75.6° and 79.4° can be readily indexed as (111), (200), (220), (311) and (222) crystal planes of NiO. These observed peaks are well assigned to the face-centered cubic (FCC) structure, which matches well with the standard data (JCPDS – 73-1519) [25,55]. Compared with pure NiO the peaks observed in case of copper doped samples are little broad which clearly indicates formation of smaller sized nanoparticles. The average crystallite size of NiO and NiO:Cu was found to be 15 nm and 9 nm respectively as determined by the Debye Scherrer's relation [56]. This reduction in the particle size for the doped nanoparticles could be attributed to the internal microstructural strain and disorder introduced in the NiO lattice due to incorporation of the Cu ions. Recently Ponnusamy et al. reported similar size results found for the Fe doped NiO nanoparticles when compared with their undoped counterpart. In the present case no peaks were found due to any other impurities or from Ni precursor. This confirms formation of pure phase of NiO and NiO:Cu NPs. The sharpness and the intensity of the peaks indicate formation of well crystalline nanoparticles. The XRD patterns of NiO:Cu NPs calcined at 350 °C, 450 °C and 550 °C as indicated in Fig. 7, exhibits peaks corresponding to pure FCC
Table 2 Information obtained from XRD patterns for Cu-doped NiO samples (NiO:Cu) at different calcination temperatures. Samples at different calcinations Temp.
hkl
2θ (deg)
d-spacing (nm)
Lattice parameter, a (nm)
NiO:Cu (350 °C)
111 200 111 200 111 200
36.74 42.72 37.20 43.30 37.34 43.39
0.2443 0.2114 0.2414 0.2087 0.2405 0.2083
0.42319 0.42282 0.41813 0.41742 0.41662 0.41660
NiO:Cu (450 °C) NiO:Cu (550 °C)
Fig. 7. XRD pattern of Cu doped NiO NPs calcined at (a) 550 °C (b) 450 °C and (c) 350 °C.
153
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
Fig. 8. FESEM images of (a) pure NiO (b) Cu doped NiO (c) EDS analysis of NiO:Cu NPs.
literature [11,63]. In case of Cu doped NiO nanoparticles in addition to the 455 cm−1 peak due to Ni-O stretching the 628 cm−1band was observed which indicated the Cu-O stretching vibration band [64]. The other peaks at 531 cm−1 and 1096 cm−1 seen in doped sample can be assigned to the different modes of Cu-O bond bending vibrations which is in well agreement with the reported literature [65]. The FTIR spectra recorded for NiO:Cu NPs at different calcination temperatures are shown in Fig. 11. In all the samples the characteristic band of Ni-O stretching vibration mode was observed at 455 cm−1 which confirmed the formation of NiO nanoparticles and not the bulk NiO [66]. It was observed that with increase in the calcination temperature from 350 °C to 550 °C the O-H stretching vibrations peak intensity (3427 cm−1) reduced [61] and became broader which is due to the fact that calcined nanomaterials tend to physically absorb water as, explained by the Kemary et al. The 2360 cm−1 band indicated CO2 vibration which totally disappeared at calcination temperature of 550 °C. This result of ours is consistent with the literature reports [56]. The intensity of the 1632 cm−1 band was also found to decrease with increase calcination temperature. 1764 cm−1 peak due to C=O stretching is observed only at 550 °C and was found to be absent at lower calcinations temperature. Doublet peak observed at 1030 cm−1 and 1128 cm−1, identified the existence of carbonates at low calcination temperature of 350 °C which diminished at 550 °C [67]. 1030 cm−1 and 1128 cm−1 strong carbonates peak observed at
increase in the calcination temperature. For the case of NiO:Cu sample calcined at 350 °C (Fig. 9(a)), spherical shaped particles are formed as for the case of pure NiO samples but of smaller size of approximately 9 nm. Further increase in the calcination temperature of 450 °C and 550 °C makes the NPs to aggregate and form big clusters as shown in Fig. 9(b) and (c) respectively. 3.5. FTIR Analysis Fourier transform infrared spectroscopy is one of the promising techniques mainly to identify the functional groups present in the molecules [59]. Fig. 10 represents the FTIR spectra of pure and doped NiO NPs recorded in the wavenumber range of 400 cm−1 to 4000 cm−1. Compared to pure NiO sample the Cu doped NiO samples show broad peak in the higher wavenumber region above 3000 cm−1. This band was found to be centered at 3427 cm−1 which is assigned to hydroxyl groups and O-H stretching vibration of the interlayer water molecules [60]. In case of pure NiO this hydroxyl group band is found at higher wavenumber [61,62]. 1650 cm−1 and 1632 cm−1 band indicates the OH bending vibrations of water molecules [56] in NiO and NiO:Cu samples. 1300 cm−1 peak may be due to O-C=O asymmetric stretching [60] and 1764 cm−1 peak due to C=O stretching. The 571 cm−1and 455 cm−1peaks observed in pure NiO sample undeniably indicated to the Ni-O stretching mode, which was further confirmed by 154
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
Fig. 9. FESEM images of NiO:Cu NPs calcined at (a) 350 °C (b) 450 °C and (c) 550 °C.
Fig. 10. FTIR Spectra of nanoparticles (a) pure NiO (b) Cu-doped NiO.
Fig. 11. FTIR spectra of NiO:Cu NPs calcined at (a) 350 °C (b) 450 °C and (c) 550 °C.
4. Conclusion
350 °C, weak carbonates peak observed at 550 °C. At higher calcination temperature of 550 °C, two peaks at 519 cm−1 and 628 cm−1 band were observed which corresponded to Cu–O bonds, and was in wellagreement with the literature [64,68].
To conclude, we have reported successful synthesis of pure NiO and NiO:Cu (4 at%) nanoparticles obtained by wet chemical method using NiCl2·6H2O as a precursor source. Our systematic studies performed in 155
Materials Science in Semiconductor Processing 66 (2017) 149–156
K. Varunkumar et al.
(2015) 194–202. [20] K. Sathishkumar, N. Shanmugam, N. Kannadasan, S. Cholan, G. Viruthagiri, J. Sol.Gel Sci. Technol. 74 (3) (2015) 621–630. [21] M. Aliahmad, A. Rahdar, Y. Azizi, JNS 4 (2014) 145–152. [22] D. Theyvaraju, S. Muthukumaran, Phys. E 74 (2015) 93–100. [23] S. Sankar, S.K. Sharma, NamhyunAn, H. Lee, D.Y. Kim, Y. Bin Im, Y.D. Cho, R. Sankar Ganesh, S. Ponnusamy, P. Raji, L.P. Purohit, Optik 127 (2016) 10727–10734. [24] C. Thangamani, K. Pushpanathan, JCHPS Spec. Issue 11 (2015) 111–113. [25] P. Pascariu, A. Airinei, N. Olaru, IulianPetrila, V. Nica, L. Sacarescu, F. Tudorache, Sens. Actuators B222 (2016) 1024–1031. [26] M.V. Reddy, G. Prithvi, Kian Ping Loh, B.V.R. Chowdari, ACS Appl. Mater. Interfaces 6 (1) (2014) 680–690. [27] M.V. Reddy, Cai, Yu, F. Jiahuan, Kian Ping Loh, B.V.R. Chowdari, ACS Appl. Mater. Interfaces 5 (10) (2013) 4361–4366. [28] M.V. Reddy, G.V. SubbaRao, B.V.R. Chowdari, Chem. Rev. 113 (2013) 5364–5457. [29] R. Sahay, P. Suresh Kumar, V. Aravindan, J. Sundaramurthy, W. Chui Ling, S.G. Mhaisalkar, S. Ramakrishna, S. Madhavi, J. Phys. Chem. C 116 (2012) 18087–18092. [30] K. Krishnamoorthy, Sang Jae Kim, Mater. Res. Bull. 48 (2013) 3136–3139. [31] L. Wang, K. Zhang, Zhe Hu, W. Duan, F. Cheng, Jun Chen, Nano Res. 7 (2) (2014) 199–208. [32] S. Petnikota, S.K. Marka, A. Banerjee, M.V. Reddy, V.V.S.S. Srikanth, B.V.R. Chowdari, J. Power Sources 293 (2015) 253–263. [33] S. Petnikota, V.V.S.S. Srikanth, P. Nithyadharseni, M.V. Reddy, S. Adams, B.V.R. Chowdari, ACS Sustain. Chem. Eng. 3 (2015) 3205–3213. [34] P.S. Patil, L.D. Kadam, Appl. Surf. Sci. 199 (2002) 211–221. [35] L. Zhao, Ge Su, Wei Liu, L. Cao, J. Wang, Z. Dong, M. Song, Appl. Surf. Sci. 257 (2011) 3974–3979. [36] A.H. Ali, R.R. Kadhim, Int. J. Appl. Or. Innov. Eng. Manag. 4 (3) (2015) 23–28. [37] V. Gowthami, M. Meenakshi, N. Anandhan, C. Sanjeeviraja, Solid State Phys. AIP Conf. Proc 1591 (2014) 884–886. [38] A. Barakat, M. Al-Noaimi, M. Suleiman, A.S. Aldwayyan, B. Hammouti, T.B. Hadda, S.F. Haddad, A. Boshaala, I. Warad, Int. J. Mol. Sci. 14 (2013) 23941–23954. [39] Z.N. Kayani, F. Saleemi, I. Batool, Appl. Phys. A 119 (2015) 713. [40] J. Tauc, R. Grigorovici, A. Vancu, Phys. Status Solid 15 (1966) 627–637. [41] R.K. Pandey, S.N. Sahu, S. Chandra, Handbook of Semiconductor Electrodepositions, Mercel Dekker, inc, NewYork, 1996, p. 142. [42] N.M. Hosny, Polyhedron 30 (2011) 470–476. [43] M. Arshad, M. Meenhaz Ansari, A.S. Ahmed, P. Tripathi, S.S.Z. Ashraf, A.H. Naqvi, Ameer Azam, J. Lumin. 161 (2015) 275–280. [44] A.J. Varkey, A.F. Fort, Thin Solid Films 235 (1993) 47–50. [45] M.S. Mohseni, S.A. Hosseini, M. Rezaee, S.K. Sadrnezhaad, D. Mohammadyani, Ultrason. Sonochem. 19 (2012) 841–845. [46] X. Song, L. Gao, J. Phys. Chem. C112 (2008) 15299–15305. [47] M. Lai, Fu, G. CongGuo, Xi Liu, Bin Liu, Li ZhenCai, J.S. Huang, Inorg. Chem. Commun. 8 (1) (2005) 18–21. [48] N. MironovaUlmane, A. Kuzmin, T. Dizhbite, I. Sildos, M. Pars, J. Grabis, 〈http:// www.physics.by/e107_files/mono/2book2009_pdf/3st_33.pdf〉, 2009. [49] N. MironovaUlmane, A. Kuzmin, I. Steins, J. Grabis, I. Sildos, M. Pars, J. Phys.: Conf. Ser. 93 (012039) (2007) 1–5. [50] N. MironovaUlmane, A. Kuzmin, J. Grabis, I. Sildos, V.I. Voronin, I.F. Berger, V.A. Kazantsev, Solid State Phenom. 168 (2011) 341–344. [51] S.C. Chen, T.Y. Kuo, Y.C. Lin, C.L. Chang, Adv. Mater. Res. 123–125 (2010) 181–184. [52] A. Loukil, A. Boukhachem, M. Ben Amor, M. Ghamnia, K. Raouadi, Ceram. Int. 42 (7) (2016) 8274–8289. [53] J. Mathew, G. Varghese, J. Mathew, SOP Trans. Nano-Technol. 1 (3) (2014) 1–11. [54] Y. Ashok kumarReddy, A. SivasankarReddy, P. SreedharaReddy, J. Mater. Sci. Technol. 29 (7) (2013) 647–651. [55] N. Dharmaraj, P. Prabu, S. Nagarajan, C.H. Kim, J.H. Park, H.Y. Kim, Mater. Sci. Eng. B 128 (2006) 111–114. [56] K.K. Purushothaman, G. Muralidharan, Sol. Energy Mater. Sol. Cells 93 (2009) 1195–1201. [57] K. Karthik, G. KalaiSelvan, M. Kanagaraj, S. Arumugam, N. Victor Jaya, J. Alloy. Compd. 509 (2011) 181–184. [58] K. Ho Kim, C. Takahashi, Y. Abe, M. Kawamura, Optik 125 (2014) 2899–2901. [59] N. Li, R. Bai, Ind. Eng. Chem. Res. 44 (17) (2005) 6692–6700. [60] M. El-Kemary, N. Nagy, I. El-Mehasseb, Mater. Sci. Semicond. Process. 16 (2013) 1747–1752. [61] N. Bayal, P. Jeevanandam, J. Alloy. Compd. 537 (2012) 232–241. [62] M.Z.B. Hussein, Z. Zainal, C.Y. Ming, J. Mater. Sci. Lett. 19 (2000) 879–883. [63] F. Davar, Z. Fereshteh, M. Salavati- Niasari, J. Alloy. Compd. 476 (2009) 797–801. [64] M. Faisal, S. BahadarKhan, M.M. Rahman, A. Jamal, A. Umar, Mater. Lett. 65 (9) (2011) 1400–1403. [65] P.K. Raul, S. Senapati, A.K. Sahoo, I.M. Umlong, R.R. Devi, A.J. Thakur, V. Veer, RSC Adv. 4 (2014) 40580–40587. [66] A. Kalam, A. Al-Sehemi, A. Al-Shihri, Gaohui Du, T. Ahmad, Mater. Charact. 68 (2012) 77–81. [67] A. Sharma, Pallavi, S. Kumar, S. Dahiya, N. Budhiraja, Adv. Appl. Sci. Res. 4 (1) (2013) 124–130. [68] S. Al-Amri, M. Shahnawaze Ansari, S. Rafique, M. Aldhahri, S. Rahimuddin, A. Azam, A. Memic, Curr. Nanosci. 11 (2) (2015) 191–197.
the present work clearly indicated that calcination temperature is one of the major factors that have significant effect on the optical, structural and morphological changes observed in Cu doped NiO NPs system. UVVisible absorption spectroscopy showed blue shift due to quantum confinement effect in NiO:Cu as compared to pure NiO. With increase in calcination temperature from 350 °C to 550 °C, NiO:Cu samples showed an obvious red-shift to the visible region which indicated decreased bandgap and increased crystallite size. Bandgap estimation done using Tauc plot illustrated that the process of doping increased the band gap of synthesized NiO samples while calcination process has reverse effect. The XRD results showed that the resultant NPs prepared in the present case were highly crystalline pure NiO with FCC structure. The average crystallite size of NiO:Cu NPs calcined at 350 °C, 450 °C and 550 °C, determined by the Debye Scherrer's formula were 4 nm, 6 nm and 9 nm respectively. AAS and EDS data confirmed the presence of Cu element in doped NiO samples. Raman data confirmed the fundamental phonon and magnon peaks characteristic of NiO in pure and doped samples. Optical transmittance data exhibits reduction in transmittance due to Cu doping (55% to 35%) and with increase in the calcination temperature the NiO:Cu samples exhibited reduced transmittance i.e. 65–32% which is in well-agreement with published reports. Morphological studies using FESEM revealed spherical shaped particles for pure NiO and NiO:Cu samples calcined at 350 °C while changes were observed for other calcination temperatures. FTIR results supported the XRD data and confirmed the formation of NiO and NiO:Cu nanoparticles with high purity. Acknowledgments We sincerely thank Divyashree A for helping us in carrying out some of the analysis mentioned in this paper. We also thank Mr. Gadhadar, NoPo nanotechnologies for analyzing Raman data for our samples. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References [1] M. Tadic, D. Nikolic, M. Panjan, G.R. Blake, J. Alloy. Compd. 647 (2015) 1061–1068. [2] G. AnandhaBabu, G. Ravi, T. Mahalingam, M. Kumaresavanji, Y. Hayakawa, Dalton Trans. 44 (2015) 4485–4497. [3] K.C. Wang, Po.S. Shen, M. Hsien, Li,S. Chen, M.W. Lin, P. Chen, G.T. Fang, ACS Appl. Mater. Interfaces 6 (15) (2014) 11851–11858. [4] Y. Huang, Y. Zhang, Sai Lin, Lu Yan, R. Cao, R. Yang, X. Liang, W. Xiang, J. Alloy. Compd. 686 (2016) 564–570. [5] H. Hu, G. Chen, C. Deng, Y. Qian, M. Wang, Q. Zheng, Mater. Lett. 170 (2016) 139–141. [6] G. Chen, G. Hongtao, D. Chengjun, X. Xuechun, Y. Wang, J. Phys. Chem. Solids 98 (2016) 209–219. [7] S. Lee, S. Park, C.W. Kim, D. Lee, L. Chongmu, C. Jin, Thin Solid Films 598 (2015) 33–38. [8] P.M. Ponnusamy, S. Agilan, N. Muthukumarasamy, T.S. Senthil, G. Rajesh, M.R. Venkatraman, Dhayalan Velauthapillai, Mater. Charact. 114 (2016) 166–171. [9] R. Krishnakanth, G. Jayakumar, A. Albert Irudayaraj, A. Dhayal Raj, Mater. Today.: Proc. 3 (2016) 1370–1377. [10] Y. Wang, J. Zhu, X. Yang, L. Lu, X. Wang, Thermochim. Acta 437 (2005) 106–109. [11] K. Anandan, V. Rajendran, Mater. Sci. Semicond. Process. 14 (2011) 43–47. [12] Z. Zhu, Na Wei, Hui Liu, Zuoli He, Adv. Powder Technol. 22 (2011) 422–426. [13] N.N.M. Zorkipli, N.H. MohdKaus, A. AzminMohamad, Procedia Chem. 19 (2016) 626–631. [14] B. Kisan, P. Saravanan, S. Layek, H.C. Verma, David Hesp, Vinod Dhanak, S. Krishnamurthy, A. Perumal, J. Magn. Magn. Mater. 384 (2015) 296–301. [15] K. Nadeem, AsmatUllah, M. Mushtaq, M. Kamran, S.S. Hussain, M. Mumtaz, J. Magn. Magn. Mater. 417 (2016) 6–10. [16] P.A. Sheena, K.P. Priyanka, S.N. Aloysius, S. Boby, V. Thomas, Nanosyst.: Phys. Chem. Math. 5 (3) (2014) 441–449. [17] Y. Wu, R. Balakrishna, M.V. Reddy, A. Sreekumaran Nair, B.V.R. Chowdari, S. Ramakrishna, J. Alloy. Compd. 517 (2012) 69–74. [18] B. Varghese, M.V. Reddy, Zhu Yanwu, Chang Sheh Lit, T.C. Hoong, G.V. SubbaRao, B.V.R. Chowdari, A.T. Shen Wee, C.T. Lim, C.H. Sow, Chem. Mater. 20 (2008) 3360–3367. [19] A.K. Ramasami, M.V. Reddy, G.R. Balakrishna, Mater. Sci. Semicond. Process. 40
156