1. IntroductionTransition metal oxide thin film, especially ZrO2, is a great active material due to its versatile nature, such as high chemical inertness, photochemical stability, good wear resistance, and wide band gap. Because of these properties, ZrO2 film is extensively applied to many fields, such as fuel cells, optical coatings, radiation shielding.[1–4] In particular, ZrO2 thin film is a promising alternative to SiO2 as the high κ gate dielectric in complementary metal–oxide–semiconductor technology, due to its high dielectric constant, wide energy band and offsets, and good stability with Si surface.[5–9] It is known that ZrO2 has three stable structures. The ZrO2 has monoclinic structure (m-ZrO2) at room-temperature, and it is transformed into a tetragonal structure (t-ZrO2) in a temperature range from 950 °C to 1170 °C and cubic structure (c-ZrO2) over 1170 °C.[10,11] The structure and properties of ZrO2 film strongly depend on the deposition technique and growth condition. Many researchers have tried different methods to deposit ZrO2 films, such as thermal decomposition,[12] electron-beam evaporation,[13] chemical vapor deposition,[14, 15] sol–gel method,[16] reactive sputtering,[17,18] etc. Among these methods, sputtering is known as a technique which has many merits suitable for industrial applications, such as easily depositing film in a large area, simple apparatus, low substrate temperature, high deposition rate, good adhesion to the substrate, etc.[19, 20]
As is well known, the addition of several elements such as Al,[21] Au,[16] Fe,[22] Tb,[23] Er,[24] etc. into ZrO2 film is a very effective method to influence its structure and phase composition and thereby its properties. To the best of our knowledge, there is little research about the preparation of Cu-doped ZrO2. Additionally, Cu in its Cu2+ ionic state has an ionic radius (0.72 Å) similar to that of Zr4+ (0.8 Å) and the similarities in their electronic structure, so Cu2+ ions can dope easily into the ZrO2 host lattice. Thus, we are to prepare Cu-doped ZrO2 (CZO) films with high crystalline quality and investigate the effects of Cu doping on the structural, morphological, electrical, and optical properties in this paper.
2. Experimental detailsCZO films for various doping content of Cu (0, 5.93, 7.18, and 8.07 at.%) were deposited on Si (100) and glass substrates by DC and RF magnetron co-sputtering. Zr target (pure, 99.99%) and Cu target (pure, 99.99%) were employed as source materials. All the RF depositions were performed at a fixed RF sputter power of 100 W applied to the Zr target. In order to produce CZO films with different Cu content, the DC sputtering power applied to the Cu target was varied from 0 W to 70 W. The substrate holder was fixed to a distance of 6 cm from the sputter targets and the substrate temperature was maintained around 600 °C. Prior to installing the substrates into the deposition chamber, they were ultrasonically cleaned in acetone, alcohol and deionized water for 20 min, then blown dry with purified nitrogen gas. The sputtering chamber was evacuated to a base pressure of 5 × 10−4 Pa using rotary and turbo molecular pump combination and the working pressure was set to be 2.0 Pa. High pure argon (99.99%) and high pure oxygen (99.99%) as working gas were introduced into the sputter chamber through mass flow controllers with Ar and O2 flow rates being 50 sccm and 30 sccm, respectively. In order to improve the crystalline quality and remove other contaminants present on the target surface, the pre-sputtering process was carried out for 15 min in argon atmosphere.
Various characterization techniques have been performed to study the influences of CZO films. The crystal structures of samples were characterized by x-ray diffraction (TonDa, Td-2500) using Cu Kα radiation (λ = 1.5406 Å). The chemical composition and surface morphology of CZO films were characterized by energy dispersive x-ray spectroscopy (EDX) and field emission scanning electron microscope (SEM) (FEI XL-30). The electrical resistivities of films were measured by using four-point probe technique (SZT-2A). Optical transmission spectra of the films were recorded in a specific wavelength range from 200 nm to 800 nm by using UV–vis NIR spectrophotometer (Shimadzu, UV-3600). All the measurements were performed at room temperature.
3. Results and discussion3.1. Elemental compositionThe Cu atom concentration in the CZO films has been identified by EDX analysis and presented in Fig. 1. As shown in EDX spectra, it can be confirmed that the deposited films are associated with characteristic Zr and O elements indicated by the high intensity peaks, whereas the low intensity peak attributed to Cu doping is evidence for the minor composition of Cu to be incorporated into ZrO2 film. Figures 1(a)– 1(c) show the EDX spectra of CZO films with Cu content of 5.93, 7.18, and 8.07 at.%, respectively. The observed Cu content in ZrO2 film increases from 0 at.% to 8.07 at.% with the increase of DC sputter power from 0 W to 70 W.
3.2. Structural propertiesFigure 2 shows the x-ray diffraction (XRD) patterns of the CZO films deposited on Si (100) substrates. From the XRD results, it is seen that all deposited films exhibit the (
) high intensity crystalline peak corresponding to ZrO2 monocline, which is in good agreement with the data of the standard JCPDS card No. 37-1484. When Cu-doping content is increased from 0 at.% to 7.18 at.%, the additional secondary peaks corresponding to Cu atom clusters or diverse metal oxides evidently disappear, which could be due to the homogenous mixing of Cu in the matrix of ZrO2, indicating the perfect solubility of Cu into ZrO2 host lattice. Moreover, the intensities of all peaks do not change much with Cu doping content increasing, indicating that the Cu2+ ions are perfectly incorporated into ZrO2 host lattice, without affecting the crystallinity of the ZrO2. However, when Cu-doping content is increased to 8.07 at.%, the intensity of (
) peak is weakened and a new diffraction peak is assigned to the copper oxide phase. Higher doping levels of Cu may energetically lead to coalescence of Cu atoms into metallic clusters. This observation is explained by the fact that Cu atoms first incorporate into the nanocrystal (50 W and 60 W) and then diffuse outward followed by the formation of crystalline CuO (70 W).[25,26] Mukhtar et al.[27] reported similar results that Cu-doped films have no secondary phases of Cu up to the moderate doping level of Cu 11 at.%. In addition, the full width at half maximum (FWHM) of preferential ZrO2 (
) peak decreases with the increase of Cu content in ZrO2 host lattice by substituting Cu2+ ions for Zr4+ ions. It can do so due to the ionic radius of Cu2+ (0.72 Å) being smaller than Zr4+ (0.8 Å) radius. It is also observed that with Cu content increasing, the location of the (
) diffraction peak shifts toward the high angle direction compared with that of the pure ZrO2 films as shown in Fig. 2(b), indicating important evidence of replacement of Zr4+ with Cu2+ ions.
All films are evaluated by using the structural parameters such as crystallite size. The average crystallite sizes of the samples are estimated from the broadening of the diffraction peaks (
) using the Scherrer’s formula[28]
where
D is the crystallite size,
K = 0.89 the correction factor,
λ the wavelength of the x-ray radiation (Cu K
α,
λ = 0.15405 nm),
β the corrected peak width at FWHM, and
θ the Bragg angle. The calculated values including diffraction angle, FWHM, crystallite size of the Cu-doped ZrO
2 (
![](cpb_26_10_106801/cpb_26_10_106801_ieqn8.gif)
) diffraction peak are summarized in Table
1. From Table
1, it can be seen that the crystallite size of the films first increases and then decreases as the Cu content increases. The crystallite size increases from 12.5 nm to 15.1 nm with increasing Cu content from 0 at.% to 5.93 at.%. We can speculate that the increase of crystallite size with Cu doping is due to Cu
2+ substituting for Zr
4+. However, further increasing Cu content to 8.07 at.%, the crystallite size slightly decreases to 13.2 nm. This may be attributed to the formation of stress by the difference in ionic size between Cu
2+ (0.72 Å) and Zr
4+ (0.8 Å) and the segregation of dopants in the grain boundaries for the higher doping content case. This can disturb the long-range crystallographic ordering, thereby reducing the crystallite size.
[29]Table 1.
Table 1.
![](https://cpb.iphy.ac.cn/html_resources/images/table-icon.gif) Table 1.
The values of Cu content, diffraction angle, FWHM, crystallite size of the CZO films.
.
Cu target power/W |
Cu/at.% |
2θ/(°) |
FWHM |
Crystallite size/nm |
0 |
0 |
28.18 |
0.646 |
12.5 |
50 |
5.93 |
28.38 |
0.535 |
15.1 |
60 |
7.18 |
28.40 |
0.603 |
13.4 |
70 |
8.07 |
28.42 |
0.613 |
13.2 |
| Table 1.
The values of Cu content, diffraction angle, FWHM, crystallite size of the CZO films.
. |
3.3. Morphological analysisThe surface morphologies of the films are further investigated by field emission scanning electron microscopy (SEM). Figures 3(a)–(3d) show the morphologies of CZO films deposited on Si (100) substrates at different Cu doping levels from 0 to 8.07 at.%, respectively. In Fig. 3(a), it can be observed that the pure ZrO2 film exhibits continuously, uniformly and without micro-cracks while small particles can be observed from the surface. At a lower doping level (5.93 at.%), there is more and more impurity clustering, leading to the increase of grain size and the more obvious protuberance as shown in Fig. 3(b). The enhanced grain size and surface roughness with Cu doping may be ascribed to the formation of clusters due to the aggregation of smaller grains. Kumaravel et al.[30] has also reported that the increase of grain size with the roughness of film increasing is due to the increase of density and size of metal particles on the surface of the film. However, when the Cu content is more than 5.93 at.%, the grain size decreases as shown in Figs. 3(c) and 3(d). The trend in the variation of grain size is in good agreement with that of the crystallite size calculated from XRD. The CZO film has larger grain size and surface roughness than the pure ZrO2 film, which indicates that the surface morphology is strongly influenced by the Cu-doping.
3.4. Electrical propertiesThe variation of electrical resistivity of CZO film with Cu doping content is shown in Fig. 4. Four-point probe method is a routine method of measuring the resistivity of film, due to its simple operation, easily reading and calculating, and no damage to sample surface. The resistivity (ρ) of film can be calculated from the following equation:[31]
where
C ≈ 6.28 ± 0.05 cm is the probe coefficient. We can improve measurement precision by averaging the testing values of repeated measurements (in Table
2). It could be seen from Fig.
4 that the electrical resistivity of CZO film is remarkably dependent on Cu doping content. The resistivity of CZO film is smaller than that of the pure ZrO
2 film, and first decreases then increases with Cu doping content increasing. The pure ZrO
2 film shows an electrical resistivity of 2.63 Ω⋅cm. At lower doping levels of Cu (5.93 at.% and 7.18 at.%), the resistivities of CZO film are about 1.78 Ω⋅cm and 1.48 Ω⋅cm, respectively. At a higher doping level of Cu 8.07 at.%, the resistivity increases to 2.57 Ω⋅cm. A similar trend of the variation in the electrical resistivity of Cu-doped film has been observed by Sreedhar
et al.
[32] The decrease of resistivity observed in the lower doped sample may be attributed to the holes transferring from Zr
4+ ions to the Cu
2+ ions located at the substitution sites, until ZrO
2 host lattices receive the maximum doping content of the Cu atoms. However, with the further increase of the Cu content (8.07 at.%), the resistivity is found to increase. The increase of the resistivity for a higher Cu doping content (8.07 at.%) may be due to the presence of the copper oxide, which can act as a scattering center, resulting in the increase of the resistivity. Presto
et al.
[33] reported the same results: the resistivity decreases with the Cu doping content increasing. However, as the CuO is added, the resistivity slightly increases.
Table 2.
Table 2.
![](https://cpb.iphy.ac.cn/html_resources/images/table-icon.gif) Table 2.
Values of Cu content, resistivity, transmittance, and band gap energy of the CZO films.
.
Cu target power/W |
Cu/at.% |
Resistivity/Ω⋅cm |
Transmittance/% |
Band gap/eV |
0 |
0 |
2.63 |
88.46 |
4.64 |
50 |
5.93 |
1.78 |
86.78 |
4.58 |
60 |
7.18 |
1.48 |
83.11 |
4.48 |
70 |
8.07 |
2.57 |
83.34 |
4.61 |
| Table 2.
Values of Cu content, resistivity, transmittance, and band gap energy of the CZO films.
. |
3.5. Optical propertiesIn order to further study the transmittances of the films, we prepare CZO films on glass substrates. Figure 5 shows the transmittances of the films with different Cu content in a wavelength range from 200 nm to 800 nm. As expected, all the films are transparent in the visible light region by exhibiting transmission above 81%, indicating that CZO films are of good transparency. High transmittance of the film indicates its low surface roughness and good homogeneity, which is in good agreement with SEM result of the film grown on glass substrates (not shown here). The average transmittance of the films decreases from 88.46% to 83.11% as the Cu content increases from 0 at.% to 7.18 at.%. The decrease in the transmittance value of the CZO film may be associated with the incident photons that are scattered by the metallic Cu crystals residing at substitutional positions of ZrO2 host lattice and it may also be due to the lattice defects existing in ZrO2 host lattice.[34] The similar observations on the different film samples with different dopants are made by the other researcher.[35] In addition, the sharp fall in transmittance observed at λ ~ 350 nm is due to the fundamental absorption edge of ZrO2 which is caused by the light and occurrences of inter-band transitions. Moreover a shift in the absorption edge towards higher wavelength region is also observed.
To understand the effect of Cu additions on the optical band gap of film, the (αhν)2 absorption coefficient (α2) versus the photon energy (hν) of Cu-doped ZrO2 films are plotted in Fig. 6. The optical band gap (Eg) of film is verified for direct transition and can be determined from the following relationship:[21]
where
α0 is the constant,
α ∝ ln (1/
T) the absorption coefficient, and T the transmittance of film.
Then, Eg can be evaluated from the plot of α2 versus hν, by using a linear fit of plots and extrapolating the linear part to photon energy axis at α2 = 0. The optical band gap values are 4.64, 4.58, 4.48, and 4.61 eV for 0, 5.93, 7.18, and 8.07 at.% CZO films, respectively (in Table 2). The optical band gap energies at low doping levels (5.93 at.% and 7.18 at.%), which could be due to band edge bending by Cu doping. In the case of Cu doping, Cu2+ ions are moved to the substituted sites in the ZrO2 matrix and may act as a donor impurity. The doped Cu2+ ions increase the donor density and thus result in the formation of a donor level below the conduction band (tail-like states below the conduction band), thereby reducing the band gap of ZrO2.[36] However, further increasing Cu content to 8.07 at.%, the optical band gap is found to increase. This may be attributed to the presence of CuO phase.
4. ConclusionsIn this work, Cu at various content of 0, 5.93, 7.18, and 8.07 at.% are successfully doped into ZrO2 (CZO) films by direct current (DC) and radio frequency (RF) magnetron co-sputtering. The structural, electrical and optical properties are investigated each as a function of the Cu content in ZrO2 film, determined by the power of Cu target. All these films have excellent crystalline quality with monocline structure, and crystal size becomes bigger after doping Cu. Moreover, both the average transmittance and band gap energy of the film decrease from 88.46% to 83.31% and from 4.64 eV to 4.48 eV, respectively, with the increase of doping content. The electrical resistivity of the film decreases with Cu doping increasing, indicating that the conductivity of ZrO2 film is improved. The obtained results presented in this study strongly suggest that the CZO film is very appealing candidate for functional and smart devices.