3.1 XRD analysis
Figure 3 displays the XRD patterns of the deposited Ru-doped ZnO films. All peaks were indexed well with the zinc oxide (pdf number: 01-075-0576) or zincite (pdf number: 01-075-1526) structure with preferential orientation along (002). No diffraction peaks that belong to ruthenium are observed in all samples. The peak is a little bit wider for low content of Ru doping. But the peaks’ positions slightly shift to a little bit higher 2-theta values with ruthenium entering into the matrix, and it becomes sharper with ruthenium doping. Further, an increasing tendency in the intensity of (002) and (101) peak has occurred with the increasing of Ru doped for B and C films, while an increase in the intensity of (100) and (101) peak of A films has been observed. It is known that increases in concentration attribute to an increase in film thickness [32]. The primary XRD peaks were used to derive the lattice constants a and c, which are presented in Table 2. These parameters have shown a slight difference for Ru doped samples as compared to undoped ZnO. Ghosh et al. [33] expressed that this was due to differences in the ionic radii of Zn2+ and Ru3+. Furthermore, the results shown in Fig. 3 clearly show that the Ru complex of the L1 ligand gave higher peaks compared to those with the L2 ligand.
The complete width at half-maximum of the (002) reflection has been evaluated to calculate the mean crystallite size by using conventional DebyeScherrer's formula (D). The average crystallite size of the intense diffraction peak (002) is found to be 178.8, 198.3, 244.68, 244.71, 86.9, 121.1, 200.5, 85, 141.2, and 213.4 nm for Z0, A2, A4, A6, B2, B4, B6, C2, C4 and C6, respectively. Particle size increases with an increase of Ru amount. During crystal growth, different kind of dislocations or defects can be formed. The existence of defects in the formation, such as vacancies and orientation disorder, affects the lattice parameters [34]. Furthermore, samples with greater concentrations may age faster and reach peak grain size faster than those with lower amounts [35]. The Rietveld refinement has been performed by using the Zincite and Zinc Oxide structure as starting model structure. It has occurred an increase in a and c lattice parameter after Rietveld refinement. The data is given in Table 2.
The Young’s modulus of films is given in the range 76–257 GPa in the literature [36]. The interval of the Young modulus can be attributed to geometrical effects result from microfabrication. Young modulus can be calculated via Eq. 1.
Here, a and c are lattice parameters, s11, s13, s33, and s44 are elastic compliances of ZnO with values 7.858×10− 12, − 2.206×10− 12, 6.940×10− 12, and 23.571×10− 12 m2 N− 1 respectively [37]. The Young modulus has been found around 48 GPa for (002) plane by using Eq. 1. in our study.
Residual stress is a significant characteristic of films that has a direct impact on the conductivity and surface morphology of the film [38]. Thus, Residual stress values and properties of films have been investigated in this study. As the ruthenium content increased from 0–6%, the residual stress values ranged from 0.21 GPa to 2.117 GPa, and the measured stress was determined to be compressive except for A4. Also, 4% doped A and B films have shown an increase and, shown a decrease in 6% decrease, whereas, C film have shown opposite. It can be due to stability problems and decreasing fiber radius of C6 film. Compare to %6 doping, C6 has demonstrated the lowest residual stress value. For ZnO films, Al-Khawaja et al. [39] have found that when crystalline quality improves with film thickness, the stress concentration reduces rapidly in increasing thickness and stays unchanged. The influence of dopants on crystallinity, according to Pham et al. [38], may be described by the growing process, which can be clarified by the desorption possibility of dopant atoms, and the varied residual stresses and crystal sizes reflect the dopant size impact on the host ZnO film structure. The thermoelectric capabilities of the ZnO thin films with the least compressive stress are enhanced by its higher Power Factor value.
Residual stress values have shown a lower value than pure ZnO film. It has been discovered that as the thickness of the produced films grows, the crystalline quality improves and the residual stress decreases. These results indicate the positive effect of the perimidine ligand derivatives for this Ru-doped ZnO thin films.
Table 2
The lattice constant, young modulus and residual stress of ZnO thin films
Thin Films
|
Unit cell parameter
(Å)
|
Rietveld
Refinement
(Å)
|
Young Modulus (GPa)
|
Residual Stress (GPa)
|
Film thickness (µm)
|
|
Z0
|
a = 3.24, c = 5.19
|
a = 3.24, c = 5.19
|
48.46
|
-2.177 (1.5)
|
1.045
|
|
A2
|
a = 3.22, c = 5.15
|
a = 3.24, c = 5.19
|
48.51
|
-0.5 (1)
|
1.261
|
|
A4
|
a = 3.23, c = 5.16
|
a = 3.25, c = 5.21
|
48.62
|
1.4 (1.1)
|
1.436
|
|
A6
|
a = 3.21, c = 5.13
|
a = 3.25, c = 5.20
|
48.62
|
-0.9 (2)
|
1.714
|
|
B2
|
a = 3.23, c = 5.17
|
a = 3.24, c = 5.20
|
48.51
|
-0.8 (0.9)
|
1.458
|
|
B4
|
a = 3.22, c = 5.15
|
a = 3.24, c = 5.20
|
48.51
|
-1.7 (0.8)
|
1.919
|
|
B6
|
a = 3.22, c = 5.16
|
a = 3.24, c = 5.20
|
48.41
|
-0.4 (0.1)
|
2.256
|
|
C2
|
a = 3.22, c = 5.16
|
a = 3.25, c = 5.22
|
48.41
|
-1.75 (0.2)
|
2.014
|
|
C4
|
a = 3.22, c = 5.12
|
a = 3.24, c = 5.20
|
48.92
|
-0.21 (0.9)
|
2.322
|
|
C6
|
a = 3.22, c = 5.16
|
a = 3.24, c = 5.20
|
48.41
|
-0.29 (1.3)
|
2.586
|
|
3.2. SEM analysis
The surface structure of Ru-doped ZnO thin film was shown in SEM images in Fig. 4. Images were taken at 10.00 K magnification with EHT value at 15.00 kV. The Ru-doped ZnO thin film was mechanically stable and demonstrated very strong adherence to the substrate at room temperature. After deposition, no peel or crack were observed of the thin films except for C films. In these films, a slight decrease in stability was observed from C4 to C6 films as the addition rate increased. As can be seen from the SEM images, while there is a leafy structure in undoped thin film, a nano fiber like structure was observed in Ru-doped ones and the density of this structure increased as the amount of additive increased.
The nano fiber like structures were shown in the upper right corner for each sample SEM image in Fig. 4. The average diameters of these fibers calculated by taking different measurements were shown in Table 3. The coating thicknesses of thin films were obtained from the SEM image of the cross-sectional area. The coating thickness and fiber thickness were demonstrated in Fig. 5. Both increased as the doping rate increased except for C6 film. In this sample, it is was seen that as the coating thickness increased, the fiber thickness decreased. This reveals that the [Ru(L2)(p-cymene)Cl]Cl compound used for the ruthenium doping process has a stability problem at 6 wt.%.
Table 3
Coating and fiber thickness of Ru-doped ZnO thin films
Thin film
|
Coating thickness (nm)
|
Fiber
thickness (nm)
|
Z0
|
1 045
|
----
|
A2
|
1 261
|
318
|
A4
|
1 436
|
388
|
A6
|
1 714
|
475
|
B2
|
1 458
|
743
|
B4
|
1 919
|
969
|
B6
|
2 256
|
1 471
|
C2
|
2 014
|
259
|
C4
|
2 322
|
728
|
C6
|
2 586
|
627
|
3.3. UV-Vis analysis
The optical property of pure ZnO and Ru-doped ZnO films have been determined by UV–vis spectrometer from 300 to 800 nm wavelength range The UV–vis absorption spectra of various Ru doped ZnO thin films shows with a broad absorption in the ultraviolet region around 300–370 nm (Fig.6All of the films have a relatively high transmittance of more than 70% in the visible spectrum. The absorption edges have been found between 392–425 nm for all the films. The dramatic rise in absorbance at wavelengths of 300 nm, according to Ghosh et al. [40], is due to inter-band transitions near the fundamental edge. Undoped ZnO thin film has lower absorption in the visible range of the spectrum and the absorption edge increases for B and C films depending on Ru dopant. Further, the absorption wavelength increases from 403 to 417 nm and (403–420 nm) and shifted to blue. Blue emissions are often caused by impurities and imperfections in the ZnO lattice [41].
The bandgap is a key indicator in a solid's electrical conductivity and is influenced by layer thickness, degree of crystallinity, and temperature. Tauc's relationship was used to calculate the optical bandgaps of the coatings (Eq. 2).
Here, A is constant, hv is photon energy, and a is absorption coefficient. Optical bandgap of all films prepared with different Ru concentrations 2%, 4%, and 6% for all films are given in Fig. 7. The optical bandgap is 3.20 eV for pure ZnO and all films have bandgap between 2.76–3.20 eV. The bandgaps of the A and B films have decreased with the increase of Ru content. The s–d and p–d contacts that lead to the bandgap cause the bandgap to shrink, making these materials more suitable for nano-optoelectronic devices [33, 42]. Because the wave patterns of the electrons coupled to the impure atoms begin to overlap as the Ru doping concentration level grows, the bandgap decreases at high doping densities. Allowable shallow states in the bandgap are created by impurities such as ruthenium dopant, and these shallow levels have very weak ionization energies [37]. Also, the film thickness and particle size have an effect on bandgap of films. Lower particle size has a higher bandgap due to quantization. The reason for the contraction in the Eg bandgap of Ru doped ZnO films may be that these films doping formed new recombination centers with lower emission energy. According to Kumar et al. [42], the smaller bandgap value in Ru doped ZnO nanorods can be related to the materials' oxygen-deficient non-stoichiometric nature. Whereas, the bandgap is a bit increased at C4, and then decreased at C6 film. The reason for the increase in Eg values in C films may be that the increased carrier concentration prevents low energy states in the transmission band and causes the Burstein-Moss effect [43, 44]. This explains the decrease in fiber thickness at the transition from C4 to C6, although fiber sizes increase as ruthenium contribution increases.
3.4. Photoluminescence analysis
Photoluminescence (PL) analysis is a non-destructive and sensitive analysis method used to examine defects in semiconductors and the energy levels of these defects. Room-temperature PL spectra of pure and Ru-doped ZnO films are given in Fig.8. The interface, surfaces, contamination levels, interface smoothness, and structural flaws in crystals can all affect photoluminescence (PL) intensity vs. wavelength [45].
Photoluminescence (PL) spectroscopy has been used to investigate the luminescence characteristics of individual and collective ZnO nanostructures [46]. It has been obtained three bands in PL spectra in our study. One excitation band have been observed around 326 nm and, it has been observed that the low-intensity and broad excitation peak centered at 471 nm corresponding to blue emission for B films. This peak level decreases depending on the Ru increase. This excitation peak is slightly shifted with the effect of annealing according to the literature [47]. At 650 nm, a red emission peak has been seen. The red PL band has attributed either to the oxygen intermediate atoms (Oi) or the hydroxyl group (OH) due to the excess oxygen at the ZnO surface. The change from Zn interstitial (Zni) to oxygen interstitial (Oi) states caused the red emission. These kind of occurrences are most common in oxygen-rich environments. Other reports show that receiver defects associated with VZn are assumed to be responsible for deep-level emission close to 650 nm [48]. Crystal defects in ZnO are caused by the substitution of divalent Zn2 + ions with trivalent Ru3 + ions, such as zinc interstitials (Zni), zinc spaces (VZn), oxygen interstitials (Oi), and oxygen vacancies (VO) [47]. As the concentration of Ru ions increased, the intensity of UV emission increased due to the ZnO increasing the crystal's quality. The strong emissions have been observed in 6% Ru doped ZnO A, C, and B, respectively. In addition, it is thought that the study may have an effect on rapidly developing borate and obsidian glasses and improve the properties of these glasses [49, 50].
3.5 FT-IR analysis
FT-IR spectra of all synthesized thin films have been recorded in the range of 400–4000 cm− 1 using KBr pellets and are shown in Fig. 9. The commercial ruthenium additive (A films) has caused to form new absorption peaks around 3672 cm− 1 attributed to O–H stretching vibrations, and 1394 cm− 1 bending vibration of O–H groups [51]. The peak observed around 2849 and 2918 cm− 1 at Z0 film, which is attributed to CH2 and CH groups and other associations by H–bond, has become more intense with the commercial ruthenium doping, and have shifted to 2920 and 2988 cm− 1, respectively. The υ(C = O) ester peak, seen around 1730 cm− 1 for Z0 and 2% films, has disappeared with the increase in ruthenium amount. A shift occurred in the peaks observed around 1581, 1139 and 872 cm− 1 with ruthenium addition. Sing et al. [52] have expressed that the reason why occur a shift in peaks of Zn–O stretching vibrations with the increasing doping of Ru, might be due to the increased replacement of Zn2+ ions by Ru3+ ions as the percentage of Ru increases in the ZnO lattice. The peaks have been observed at 2920, 2846, 1571, 1080, and 867 cm− 1 in the spectrum of C films, similar to Z0 films. Unlike Z0 films, there has been an increase in the intensity of the peaks seen at 2920 and 2846 cm− 1, and the peak at 1137 cm− 1 has shifted towards 1080 cm− 1. The new peaks have been observed around 3672, 1377, 1056, and 665 cm− 1 in the spectrum of B films compared to Z0 films. A shifting for both organic ruthenium doped film has occurred in peaks seen at 2923, 2853, 1581, and 892 cm− 1 of pure film.