Copper and nickel co-doping effects on the structural, optical and electrical properties of tungsten trioxide nanoparticles prepared by co-precipitation technique

The pristine WO3 (tungsten trioxide) and Cu/Ni Co-doped WO3 nanoparticles were synthesized by the co-precipitation synthesis methods with fixed wt.% of nickel, i.e., 5 wt.%, and different wt.% of copper, i.e., 2, 3, 5 wt.%. The structural, optical and electrical properties, as well as the surface morphology were investigated thoroughly by various characterization techniques. All the doped/co-doped and un-doped samples had an orthorhombic structure, according to the XRD measurements. The XRD results further confirmed that the average crystalline size ranged from 46 to 25 nm after the dopant concentration was introduced. Reflectance spectroscopy was used to investigate the optical properties of the prepared nanomaterials and revealed that the optical band gap varied from 2.80 to 2.63 eV with dopant concentrations. The room temperature photoluminescence study showed that the emission peaks were observed in the visible region with slight peak shift towards short wavelength with dopant concentrations. The FTIR studies described the different mode of band related to a functional groups present in the materials and the stretching mode of O–W–W observed from 550 to 1050 cm−1. The I-V studies confirmed that the prepared samples had a good ohmic contact behavior and the resistivity decreased from 6.3 × 10−3Ω.cm to 0.63 × 10−3 Ω.cm with doping concentrations.


Introduction
Metal oxide semiconductor nanoparticles have recently received a lot of consideration due to their unique potential for different applications. Tungsten trioxide has been highly studied because of its outstanding optical, structural and electrical properties [1]. It has reasonable amount of electrons in the valence band with band gap energy of 2.7 to 3.5 eV [2] and with significant absorption in the far-ultraviolet and visible regions. It has transition metal oxide semiconductor which is an excellent property for various applications, such as photovoltaic [3], electrochromic [4], photocatalyst [5], and photochromic [6], and gas sensor [7]. Various researchers have prepared tungsten trioxide nanostructures such as nanowires, nanotubes, nanoribbons, nanorods, and nanoplates which may influence their physical, optical and electrical properties [8][9][10]. Controlling the size, surface morphology, and dopant concentration of WO 3 are important to for different applications. Doping is one of the most essential mechanisms to increase material's optical and electrical properties. This means that increasing the materials' conductivity and the optical band gap of the host materials reduces or increases their energy. According to a recent study, substitutional doping of metal ions is a good way to adjust the electronic structure and electrical properties of WO 3 [11,12]. So far, various researchers have been investigating transition metal ions as dopants to study the structural, electrical and band edge properties of WO 3 , including Mo [13], Co [14], Ni [15], Fe [16]. Cu [17,18] Mn, [19] and Cd [20]. Moreover, various studies have been conducted co-doping transition metal ions tomodify Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI. optoelectricalproperties of WO 3 nanoparticles. For instant, S Mohammadi et al investigated the effects of pure, Zn, Cu, co-doping Zn and Cu in WO 3 nanoparticles prepared by using both precipitation and co-precipitation methods for the application of photocatalytic and antibacterial activity. They found that the co-doping WO 3 nanoparticles have higher photocatalytic degradation of gentamicin than other sample [21]. Albalshi et al investigated the electrochemical properties of Ti and Zn co-doped WO 3 thin films synthesized by the sol-gel method [22]. Furthermore, compared to a single doped WO 3 thin film, co-doped film has improved electrochemical and electrochromic properties as well as higher electrochemical stability [22]. N Boonprakob et al investigated photocatalytic for removal of hexavalent chromium using Ag and Cu co-doped WO 3 nanoparticles synthesized by using a simple co-precipitation approach [23]. This co-doping effect has higher photocatalytic reduction under visible light irradiation than un-doped [23].
To synthesize WO 3 nanostructures, researchers have used different synthesis methods, for example sol-gel [24], microwave irradiation [25], co-precipitation [26], spray pyrolysis [27], and hydrothermal [28]. Among them, the co-precipitation method is the simplest, cost-effective, the highest purity, and has good uniformity. To our knowledge, no studies have examined pure WO 3 , 5% Ni:WO 3 and (2%Ni+5%Cu):WO 3 , (3%Ni+5%Cu):WO 3, and 5% (Cu:Ni) co-doped WO 3 nanoparticles by using environmental friendly coprecipitation method. The objective of this work is to give an insight on how the mono doped and co-doped can affect the structure, electrical and optical properties of WO 3 nanoparticles. The obtained results are suggested that the prepared materials are good candidates for solar cell applications.
2. Experimental procedure 2.1. Synthesis of pure, mono and Co-doped (Cu,Ni) WO 3 nanoparticles (NPs) Pure WO 3 , 5%Ni:WO 3 (mono) and (2% Cu+5%Cu):WO 3, (3% Cu+5%Ni):WO 3 and 5%(Cu:Ni) co-doped WO 3 nanoparticle were synthesized by simple chemical co-precipitation method. The precursors used in the synthesis were sodium tungstates dehydrate, copper (ii) nitrate trihydrate, and nickel nitrate hexahdrate as starting materials., while aqueous HCl used to adjust the pH of the solution and NaCl as capping agent. In a typical synthesis appropriate amount of sodium tungstate dihydrate was dissolved in 100 ml of distilled water. In a closed-necked flask, the solution was vigorously stirred for 30 min with a magnetic stirrer. After 30 min appropriate amount of NaCl was added to the solution then stirred until a homogeneous solution was formed. Then 37% aqueous HCl solution was also added to the above solution drop wise under constant stirring until the pH was attained to 1 and then the solution continues stirred and heated for one hrs at 100°C. Afterwards, the solution aged for 24 h and then centrifuged to the obtained precipitates. The obtained precipitates were washed several times with distilled water and ethanol to remove excess sodium and chlorine ions then dried in an electric oven at 100°C overnight. Finally, the obtained powder was calcined at 500°C for 3 h in order to obtain WO 3 nanoparticles. The 5% Ni:WO 3 and (2% Cu+5%Cu):WO 3 , (3% Cu+5%Ni):WO 3 and 5%(Cu:Ni) co-doped WO 3 nanoparticles have synthesized by chemical co-precipitation method with different weight percentage of 2%, 3%, and 5% for Cu and fixed amount 5% for Ni using the same procedure explained above. The all the obtained samples were used for different characterization.

Characterization techniques
The phase purity and structure of the pure, mono and co-doped WO 3 nanoparticles were analyzed by XRD-7000 x-ray difractometer maxima with Cu Kα radiation (λ=0.15406 nm). The presence of different functional groups in the sample was confirmed by FTIR spectrophotometer (PerkinElmer, Model: Spectrum Two). SEM and UV-Vis Spectroscopy (Model: UV-3600 Plus) were used to characterize the morphological information and optical properties of the samples. Room temperature photoluminescence (PL) spectra were performed by a Fluoromax-4 Spectrofluorometer in the excitation wavelength of 350 nm using a 40 W Xenon arc lamp. The current-voltage characteristics of the junctions were measured using a Keithley source meter unit (model: 2400) in the dark at room temperature.

XRD analysis
The prepared samples were analyzed by x-ray diffraction using the CuKα wavelength of 0.15406 nm. Figure 1(a) shows the XRD of pure WO 3 , 5% Ni:WO 3 (2% Cu+5%Cu):WO 3 , (3% Cu+5%Ni):WO 3 and 5%(Cu:Ni) codoped WO 3 nanoparticles. The XRD result confirmed that the prepared samples had an orthorhombic crystal structure of with space group P and lattice parameter a=7.3840 Å, b =7.5120 Å, and c =3.8460 Å and α=β=γ=90°C. The result is in a good agreement with the standard JCPDS card No. 020-1324. The incidence of sharp and number of diffraction peaks revealed the polycrystalline nature of the prepared nanomaterials. No impurity peaks of Ni and Cu were observed from the XRD diffraction pattern which implies a successfully incorporations of Ni 2+ and Cu 2+ ions into the WO 3 lattice. As demonstrated in figure 1 the intensity of co-doped (Ni:Cu) WO 3 nanoparticles increased in the plane (001) when compared to 5% Ni:WO 3 and pure WO 3 nanoparticles. This is due to the fact that (Ni:Cu) co-doping improves the development rate of WO 3 nanoparticles and gives them a higher degree of crystallinity than 5% Ni:WO 3 and pure WO 3 , resulting in sharper diffraction peaks in XRD. The intensity of 5% Ni doped WO 3 nanoparticles is lower than that of pure WO 3 nanoparticles. As result clearly indicate that the successful incorporation of 5% Ni has the WO 3 matrix.
The crystallite sizes for the prepared NPs were analyzed from the three most prominent diffraction peaks of (001), (020) and (200) planes using the Debye Scherer relation shown in equation (1).
Where, D is average crystallite size obtained from the three dominant peak, K is constant for the shape factor nearly equal to 0.9, λ is the wavelength of x-ray used (CuKα, 0.15406 nm), β is the full width at half maximum and θ is the angel of diffraction of three do peak. The average crystallite size was calculated by using the (001), (020) and (200) peaks of WO 3 to be 46 nm; 25 nm for 5% Ni:WO 3 furthermore, the crystalline size, it was decreased from 35 nm to 40 nm for various concentration of Cu at 2 wt.%,3 wt.%, 5 wt.% and fixed in Ni of 5 wt.% doped WO 3 respectively. It is indicated that Cu 2+ and Ni 2+ ions substituted in the W 6+ ions lattice site of orthorhombic structure and then the crystallite size was reduced by dopant concentration [29]. The expanded version of XRD (figure 1(f)) clearly shows the slight peak position shifts of (001), (020), (200) planes to the lower diffraction angles when compared to the pure WO 3 NPs. This peak shift to the lower angle for doped NPs indicates that the lattice parameters are slightly varied than the pure WO 3 and a successful substitution Ni 2+ and Cu 2+ ions onto the host lattice [30]. This can be associated to the smaller ionic radii difference between the host W (0.62 Å) and dopant 0.60 Å for Ni and 0.73 Å for Cu ions [31,32].
As shown in table 1, the unit cell volume of WO 3 NPs co-doped with fixed nickel and varying copper amounts decreased with dopant concentration implies that substitution of Cu and Ni ions in the lattice site, as similar phenomena have been reported [33]. It is also observed that the value of the dislocation density of undoped WO 3 NPs has the lowest value as compared to the doped sample, shown in table 1. This indicates that Cu 2+ and Ni 2+ co-doping deteriorate the WO 3 crystal structure. Table 1 also reveals that the micro-strain and crystallite size have a strong relationship. The differing ionic radius of W 6+ , Cu 2+ , and Ni 2+ ions, which enhance the micro-strain during the substitution of W ions by both Cu and Ni ions, could explain these changes in micro-strain with crystal size. As a result, dopant concentrations diminish the development rate and crystallinity of WO 3 NPs, and crystal size increases produce a decrease in the strain [34,35].
The from the XRD peaks the lattice parameters for the orthorhombic crystal structure was calculated by using equation (2):  structure of WO 3 nanoparticles is measured by the dislocation, and the presence of this strongly influences the properties of the sample. The dislocation density can also be calculated from the equation (3): Where δ is dislocation density and D is the average crystallite size of the sample. The micro-strian (ε) can be calculated by the following equation [36].
The UV-vis spectroscopy measurements were taken to assess the influence of 5% Ni:WO 3, 2% Cu +5% Ni :WO 3 , 3% Cu+5% Ni :WO 3 and 5% (Cu :Ni) co-doped on WO 3 NPs. Figures 2(a)-(f) shows the reflectance spectra and optical band gap of all WO 3 samples. Figure 2(f) shows all WO 3 samples have minimum reflectance in the UV region. In the entire sample in the visible region, the reflectance increases, but all the co-doped sample has higher reflectance than the 5% Ni:WO 3 and pure WO 3 nanoparticles. The optical band gap energy of WO 3 nanoparticles were analyzed by using Kubelka-Munk function relation equation (5) [26,37].
Where F(R) is known as kubelka-munk function and R is the reflection of the materials. The indirect band gap energy of the sample can be found by plotting (F(R)hυ) 1/2 versus energy (hυ) as shown in the figures 2(a)-(e). The figures 2(a)-(e) indicates that pureWO 3 nanoparticles has the optical band gap of about 2.80 eV which is decreased to 2.63 eV for 5% Ni: WO 3 , 2.75 eV for (2% Cu+5% Ni):WO 3 , 2.77 eV for (3% Cu+5% Ni):WO 3 and 2.73 eV for 5% (Cu: Ni) co-doped WO 3 nanoparticles. The crystal size and dopant content increases the optical band gap energy decrease. Therefore, the reduction in the optical band gap energy of WO 3 is due to the induced defect energy levels near to the conduction and valence band by Cu and Ni ions substitutions in lattice site of orthorhombic structure. However, the electrical transition from the filled valence band to the bottom of the conduction band occurs due to a higher number of defects [38]. This implies that as (Ni,Cu) is co-doped into the WO 3 lattice, defect states are formed, causing the band gap to narrow [39]. The band gap narrow due to there is more light absorption in the visible area, and which can improve the performance of optoelectronic devices. Figure 3 shows the room-temperature PL emission spectra of the prepared samples at an excitation wavelength of 350 nm. The emission spectra of pure WO 3 , 5% Ni, WO 3 , (2% Cu+5% Ni): WO 3 , (3% Cu+5% Ni): WO 3 and 5% (Cu: Ni) co-doped WO 3 nanoparticles are all in the visible region. The greater emission spectrum in the (Ni,Cu) co-dopant has a higher percentage of surface oxygen vacancy. The emission spectra of pure WO 3 , 5%Ni:WO 3 ,(2% Cu+5%Ni):WO 3 ,(3% Cu+5% Ni):WO 3 and 5%( Cu:Ni) co-doped WO 3 nanoparticles were exhibited strong emission peaks centered at 431 nm related to violet emission, 440 nm and 445 nm related to blue emission, 494 nm, 500 nm related to blue -green emission, 512, 518, 523, 525, 532, 535, 536, and 547 nm related to green emission, 610 nm related to orange emission and 621, 662 nm and 672 nm related to red emission. These different emission peaks may be due to an electron transition in WO 3 's band gap mediated by the defect energy level generated from oxygen vacancies and tungsten interstitials [40]. The broad and high intensity peaks observed at 431 nm, can be assigned to recombination free excitons, which are referred to as near band edge emission (NBE) [41]. This originated from the band-to-band transition and corresponds band gap energies of 2.87 eV, respectively, which are consistent with band gap energies derived from UV reflectance spectra. The blue emission peaks at 440 nm and 445 nm are attributed to singly ionized tungsten vacancies [42]. The blue-green emission peaks at 494 and 500 nm due to a surface defects in the WO 3 NPs corresponding to the transition between oxygen vacancy and oxygen interstitial defect or lattice defects related to oxygen and tungsten vacancies [43].   impurity and, oxygen antisites [44][45][46][47][48]. The orange-red emission originated from by transitions associated with interstitial atomic defects such as oxygen vacancy and tungsten interstitial [42]. These findings are similar works reported in the literatures [49]. Pure WO 3 and 5% Ni:WO 3 nanoparticles exhibit lower intensity than (Ni: Cu) co-doped WO 3 samples due to a lack of crystallinity [50]. The lower the PL intensity; the lower the recombination rate of photo-induced electron-hole pairs [51]. The increased PL emission intensity was attributed to electron-hole recombination caused by surface imperfections or oxygen vacancies [52]. As a result of the higher growth and more intense PL emission peak observed in the Cu and Ni co-doped WO 3 samples. The PL emission peak of co-doped WO 3 samples shifts to the lower wavelength side and has a distinct intensity because of the substitution of W + 6 by Ni 2+ and Cu 2+ ions into the host WO 3 unit cell lattice site. As the WO 3 nanoparticles were synthesized in an aqueous solution a high density of oxygen vacancies might be expected. The oxygen vacancy itself may be neutral, singly ionized or doubly ionized [49,53]. These oxygen vacancies and interstitial atomic defects form a number of trap states which are responsible for the various PL emissions in the visible region. similarly, as shown in figure 3, the incorporations of Ni 2+ and Cu 2+ ions into the WO 3 matrix resulted as a development of more new trapping centers. Based on the above discussions the proposed emission mechanisms of the generalized band gap structure and specific defect levels emissions of the prepared samples are illustrated in figure 4.

Fourier transforms infra-red (FTIR) analysis
The FTIR spectroscopy was studied to the information about the surface chemistry and chemical quality of WO 3 nanoparticles. Figure 5 shows the FTIR analysis of pure WO 3 , 5% Ni:WO 3 , (2%Cu+5% Ni):WO 3 , (3% Cu+5% Ni):WO 3 and 5% (Cu:Ni):WO 3 NPs. The observed band located at 1409 cm −1 represented to bending modes of O-H groups and the band located at 2955 to 3683 cm −1 maybe represented to the stretching modes of O-H group in water or hydroxyls [32,54]. Furthermore, the broad band observed at 550 to 1050 cm −1 is linked with the O-W-W stretching mode [32,54]. The presence of WO 3 stretching mode and absence of Ni and Cu oxide related mode further support that Ni 2+ and Cu 2+ ions are successfully doped WO 3 crystal structure. The orthorhombic crystal WO 3 nanoparticle had no impurities detected in the FTIR investigation, which was verified by the XRD data.

Surface morphology analysis
The morphology and histogram results of pure WO 3 , 5% Ni:WO 3 , and 5% (Ni: Cu) co-doped WO 3 nanoparticles are shown in figures 6(a)-(f). As demonstrated in figures 6(a)-(c), the morphology of all samples is uniform and densely packed. The particles were clearly agglomerated in the SEM images. From the SEM images it is easily observed that doping with Ni and Cu plays a significant role on the morphological nature of the nanocrystalline of the prepared samples. The surface morphology is also impacted by the dopant element, indicating that W 6+ has been replaced in the WO 3 lattice by transition metal ions, as shown by SEM micrographs. The average particle size distribution of WO 3 , 5% Ni, and 5% (Ni: Cu) co-doped WO 3 was obtained by analyzing the SEM image using the software ImageJ as shown in figures 6(d)-(f). The histograms represent WO 3 nanoparticles with an average diameter of about 225 nm. However, the addition of Ni 2+ influence the size morphology in which of average particles size distribution of the nanoplates is decreased from 225 nm to 138 nm due to divalent nickel ions restrict the growth of WO 3 and lead to growing in the nanoplate direction shown in figure 6(b) [55]. This decreased the average diameter which is a good agreement with average crystalline size calculated from the XRD. When both Ni 2+ and Cu 2+ the dopant precursor concentration the morphology may be attributed to resulting in the change of surface feature from nanoplates shape to both nanorods and nanoplates shows in figure 6(c). The Ni and Cu co-doped WO 3 nanoparticles increase the diameter from 225 to 252 nm shown in figure 6(f). These co-doping change both the shape and size of the WO 3 nanoparticles.

I-V characteristic studies
The electrical properties were carried outunder dark condition at room temperatureof pure WO 3 , 5% Ni:WO 3 and (2% Cu+5% Ni):WO 3 ,(3% Cu+5% Ni):WO 3 and 5% (Cu/Ni) co-doped WO 3 thin film. The prepared samples were used to deposit on glass substrate by simple spin coating method after dissolving in DFM. Then the spin coated film was dried at 100°C for 2 h in an oven, and the metal contact was made over the prepared film using sliver past. The result of the I-V characteristics measured for the sample exhibits a good ohmic behavior as shown in figure 7. The resistivity of the sample was calculated by using equation (5) and as shown table 2.
Where, ρ is the resistivity of the film (Ω. cm), R is the resistance (Ω), A is the area of active layer (cm −2 ) and L is the length in (cm).
The resistivity of calculated value found to decrease from 6.3×10 −3 Ω.cm for WO 3 film to 2.9×10 −3 Ω.cm for 5% (Cu:Ni):WO 3 and 0.63×10 −3 Ω.cm for 5% Ni: WO 3 thin film. This demonstrates that altering the dopant concentration in the lattice increases the electrical conductivity of WO 3 thin films by increasing carrier charge mobility. The amount of dopant in the WO 3 unit cell lattice can generate crystallographic defects, alter the electronic structure, and reduce the electrical characteristics of the (2% Cu+5% Ni):WO 3 and (3% Cu+5% Ni):WO 3 thin films related to pure WO 3 .
These findings of similar work with previous literature results were obtained by [56][57][58]. As a result, the electrical properties of 5% Ni:WO 3 NPs were found to better performance than Ni and Cu co-doped WO 3 and undoped samples. It is also observed that the resistance of the synthesized thin films decreases as the dopant concentration increases. The prepared samples show good ohmic properties and are good candidates for solar cell window applications.

Conclusions
Pure WO 3 , 5%Cu:WO 3 , (2% Cu+5% Ni): WO 3 , (3% Cu+5% Ni): WO 3 , and 5% (Ni:Cu):WO 3 nanoparticles were successfully synthesized by co-precipitation methods. The structural, optical, and electrical properties of the prepared nanomaterials were investigated. The formation of the nanocrystalline orthorhombic structure of WO 3 was confirmed by XRD. Moreover, average crystalline size, microstrain, unit cell volume and lattice parameters were significantly influenced by the dopant concentrations. The UV-Vis analyses revealed that the band gaps of the prepared samples decreased as the dopant concentrations increased. Various PL emissions in visible region were observed due to defects from a number of trap states. The increased intensity of the PL emission peak at 431 nm, 440 nm, 512 nm, and 523 nm of pure WO 3 and (Ni: Cu) co-doped WO 3 samples was attributed to the high electron-hole recombination caused by nanoplates with little rod morphology. The SEM micrographs confirmed that the prepared samples have both nanoplate and nanorod shapes and are agglomerated irrespective of the dopants. The result of the I-V characteristics measured for the sample exhibits a good ohmic behavior and the resistivity decreased with dopant concentration. The obtained results revealed that the prepared materials are suitable for solar cell devices as the window layer.

Data availability statement
The data that support the findings of this study are available upon reasonable request from the authors.