The Influence of Magnetron Sputtering Process Temperature on ZnO Thin-Film Properties

Abstract

The important research direction in surface engineering and photovoltaics is the development of new materials that can replace the previously used expensive films. A prospective compound is zinc oxide (ZnO), characterized by optical and electrical properties similar to ITO and a lower production cost. One of the key factors influencing the properties of the ZnO thin films is the technique and parameters of their production. The comprehensive investigation results of the influence of ZnO thin-films deposition process temperature on their structure, optical properties, and adhesion are presented in the paper. ZnO films were deposited by the magnetron sputtering method. The structural characteristics of the tested films were investigated by scanning electron microscopy (SEM), atomic force microscopy (AFM), X-ray diffractometry (XRD) and Raman spectroscopy, while the optical properties of the films were studied by the UV/VIS spectroscopy. Thickness and adhesion measurements of the obtained films were performed using the spectroscopic ellipsometry technique and the scratch test, respectively. The obtained research results showed the influence of the deposition process temperature on the morphology, crystallite size and adhesion of the thin films to the substrate. The effect of process temperature on optical properties, the value of the optical bandgap and crystal structures were analyzed and described. The results of this work have a meaning for the development of surface engineering and may serve as a clue in future studies in the field of modern photovoltaic structures.

1. Introduction

The transparent conductive oxides (TCO) are a group of materials from which a high electrical conductivity and high transparency for wavelength from the visible light range are required. The most commonly used type materials are Indium Tin Oxide (ITO) and Fluorine-doped Tin Oxide (FTO). The ITO layers are characterized by high transparency for wavelengths from the visible light range, amounting to approximately 85%, and by low resistance of approximately 10⁻⁴ Ω·cm. In the case of the FTO layers, the transparency reaches a value of about 80%, and the resistance is approximately 4 × 10⁻⁴ Ω·cm. Despite such good and meeting the expectations parameters, alternative materials with similar properties are sought after. The reason for this is the high cost of raw materials necessary for the production of ITO and FTO layers [1,2,3,4,5].

An interesting and promising alternative for ITO and FTO, much cheaper to produce, is zinc oxide (ZnO). This material belongs to the group of broadband n-type semiconductors. It is characterized by an energy bandgap of 3.37 eV and transparency higher than 70% for visible light waves range [1,5,6,7].

Thin films made of materials belonging to the TCO group are widely used in various industry branches, mainly as transparent conductive layers (TCL) applied, for example, in photovoltaic cells, displays or light-emitting diodes [1,3,4,5,6].

In addition to good electrical and optical properties, the homogeneous chemical composition in the entire volume of the film, high quality of the produced film, characterized by continuity throughout the entire volume, lack of cracks (both inside the film and on its surface) and as low as possible porosity are required from transparent conductive oxides. These features have a significant impact on the electrical conductivity of the material. Another factor that significantly influences the properties of the film is its morphology and structure [1,3].

The final properties of the deposited thin films can also be controlled and tuned by choosing the suitable method of their production. In the case of ZnO thin films, many techniques of its deposition mainly based on physical and chemical processes are used [6,8]. The most common are sol–gel [9,10,11], Chemical Vapor Deposition (CVD) [12,13,14,15] and Physical Vapor Deposition (PVD) [16,17,18,19].

One of the most interesting and advanced methods of producing nanometric ZnO thin films is one of the varieties of the PVD technique: magnetron sputtering. In this method, the source-target material is sputtered as a result of a direct current (DC) glow discharge or a flow of radio frequency (RF) current with a plasma assisting. Additionally, in order to amplify the intensity of the ionization process, a magnetic field is used that allows increasing the number of ions bombarding the source target. This process modification enables an increase in the speed of film formation on the surface of the substrate. The discussed method allows to control the thickness of the produced layers at the nanometric level and to produce high-quality coatings with a homogeneous material structure [6,8,20,21,22].

This work aimed to examine the influence of zinc oxide (ZnO) thin-films deposition process temperature produced by magnetron sputtering method on their structure, optical properties and adhesion.

2. Materials and Methods

The research materials were the glass samples and silicon wafers with ZnO thin films deposited on their surface by the magnetron sputtering process. The size of the samples was 20 mm × 20 mm. In order to compare how the production process affects the properties of the thin films, the deposition was performed at three different temperatures: 100, 200 and 300 °C and the process times were determined experimentally to obtain the thickness of the final films about 200 nm.

A zinc oxide target was used as the source material in the magnetron sputtering process, which was carried out in an argon atmosphere at a pressure of 5 mTorr and the gas flow adjusted to 7 sccm. The power on the magnetron was set to 75 W, and the substrate polarization to 60 V. The substrate temperature measurement was done using a Type K thermocouple enclosed in an Inconel. Feedback control of the substrate temperature was provided using a PID controller and the thermocouple signal.

The film thickness measurements were taken with the use of a spectroscopic ellipsometer SENTECH SE 850E and with SpectraRay/3 software (Sentech, Berlin, Germany). Measurements were made at angles of incidence of 50°, 60°, and 70°. In the case of ZnO thin films, the Cauchy model was used. This model is most often used for transparent oxide materials. In the case of the silicon substrate, layers of native SiO₂ having a thickness of 2.51 nm were included in the model. The thickness of the oxide films was determined based on the analysis of the uncovered silicon substrates. Thin films of ZnO deposited onto silicon wafers were fitted with a simple sandwich model Si/SiO₂/ZnO/air. The spectral range of the measurements was 200–2500 nm.

Scanning electron microscope (SEM) images were taken with a Supra 35 (Zeiss, Thornwood, NY, USA). The accelerating voltage was 3–5 kV. To obtain images of the surface topography, the secondary electron detector (by the in-lens detector) (Zeiss, Thornwood, NY, USA) was used. The microscope is additionally equipped with the X-ray energy dispersive spectroscopy detector from EDAX.

The surface morphology of the analyzed samples was evaluated using Park System’s XE-100 atomic force microscope (Park Systems, Suwon, South Korea). The study was conducted in a non-contact mode, in areas of 2 µm × 2 µm. The cantilever vibration frequency was 300 kHz, and the recorded test results were developed in the Park Systems XEI 4.3.0 program. The 2D and 3D images were recorded, and basic roughness parameters were calculated.

X-ray diffraction studies of the analysed films were carried out on an X’Pert PRO system made by the Panalytical Company (Malvern Panalytical, Malvern, UK) using the filter radiation of a cobalt anode lamp powered by a voltage of 40 kV at a filament current of 30 mA. The diffraction patterns were collected with a step of 0.05°.

Structural testing was performed using an inVia Reflex Raman spectrometer (Renishaw, New Mills, UK), equipped with an Arion laser with a 514.5 nm length for a spectral range of 50–3100 cm⁻¹.

The optical properties of the films were analyzed using the Thermo Scientific UV-VIS 220 Evolution spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The spectral range of the measurements was 300–800 nm. Based on the obtained measurements results, the value of an optical energy bandgap E_g was determined.

The adhesion of ZnO thin films was evaluated using a Nano-Combi-Tester open platform by CSM Instruments (CSM Instruments, Needham, MA, USA). During the test, a Rockwell diamond cone with a gradual increase in the indenter’s load was used to make a scratch. To estimate the critical load force (L_C), the friction force signals changes and optical microscope observations were used. Loading force was changed in the range of 0.03–30 N. The following operation parameters were applied: the scratch length—3 mm; loading speed—100 N/min, and the speed of the table displacement—10 mm/min.

3. Results and Discussion

The cross-section images of the investigated ZnO thin films are presented in Figure 1. The films produced by the magnetron sputtering technique characterize a compact structure without any delamination or defects and show columnar growth. In order to verify the thickness of the produced ZnO films, spectroscopic ellipsometry examinations were performed. The obtained tests results, summarized in

Table 1. Thickness measurement results of ZnO thin films.

Process Temperature	Spectroscopic Ellipsometry (nm)	Scanning Microscopy (nm)
100 °C	195	191
200 °C	197	194
300 °C	207	205

, confirmed the expected thickness of the produced films.

The grains were distributed randomly with no dominant arrangement direction; the grain size was irregular (Figure 2). The morphology of ZnO thin films deposited by magnetron sputtering is homogeneous and uniform. The surface does not show any discontinuities, cracks, pores, or defects.

EDS technique examination confirmed the chemical composition of the produced films (Figure 3). In the spectra, characteristic peaks for zinc and oxygen and silicon from the substrate material were present.

Using an atomic force microscope, ZnO films topography observations showed no significant process temperature influence on the size and quantity of particle aggregation on the surface of the samples (Figure 4, Figure 5 and Figure 6).

Based on the research performed using AFM microscopy, roughness parameters were determined (

Table 2. Summary results of roughness parameters values for ZnO thin films (AFM).

Process Temperature	Rq (nm)	Ra (nm)
100 °C	3.38	3.10
200 °C	4.42	3.39
300 °C	4.49	3.57

), such as Rq—the value of the root mean square average of the roughness profile, and Ra—the value of the arithmetic mean of the deviation from the roughness profile. It was found that the roughness parameters Rq and Ra of the tested films increased with the increasing temperature of the deposition process. The increase in roughness parameters may suggest the growth of grains and crystallites. Therefore, structural studies were also performed using X-ray diffraction.

The X-ray phase analysis studies (Figure 7) showed the existence of the hexagonal structure, characteristic of the zinc oxide crystallizing in the wurcite arrangement [22]. On the ZnO films diffractometers, three reflections for each of the samples were recorded: (002), (012), and (013). The highest intensity showed the reflection with index (002).

The crystallite size (D) was estimated from the Debye–Scherrer equation::

$$\mathrm{D}=(0.90\mathsf{\lambda })\backslash (\mathsf{\beta }\mathrm{c}\mathrm{o}\mathrm{s}\mathsf{\theta })$$

(1)

where D is the crystallite size, λ (1.79 nm) is the wavelength of the incident X-ray beam CoKa, β is the full width at half the maximum reflex (FWHM), and θ is the Bragg diffraction angle.

The results are summarized in

Table 3. Structural parameters of ZnO thin films.

Process Temperature	2θ (°)	Crystallite Size, D (nm)
100 °C	39.75	49
200 °C	39.95	51
300 °C	39.95	62

. Their analysis showed that the examined films were characterized by a nanocrystalline structure. The size of crystallites in the layers obtained at temperatures of 100 and 200 °C was very similar and was equal to 49 and 51 nm, respectively. In the case of films deposited at 300 °C, the crystallite sizes were larger than those obtained at other temperatures and were 62 nm.

In order to fully identify the structure of the produced ZnO films, Raman spectroscopy studies were performed (Figure 8). These are complementary XRD studies that confirm the previously identified structure. Raman spectroscopy is very sensitive to slight changes in the molecular environment and crystalline phase. RS can simultaneously measure Raman active phonon modes (in crystalline materials) and Raman active vibrational modes (in molecules). Thanks to that, it is possible to get a unique spectral fingerprint of different polymorphs of crystalline materials and spectral information from molecules in the measurement spot. The bands: 560, 570 and 1091 cm⁻¹ were identified as A₁ (LO) and E₁ (LO), for which the Raman shift values were very similar, and respectively A₁ (2LO) were characterized by the highest intensity. There have also been identified following bands: E₂^high at 436 cm⁻¹, A₁ (2LA) at 475 cm⁻¹, A₁ (LA + TO) and (LA + LO) at 771 and 791 cm⁻¹, A₁ (2TO) at 950 cm⁻¹ and A₁ (TO + LO) at 1000 cm⁻¹ [23,24,25,26].

The ZnO films transparency measurements for electromagnetic wavelengths from the visible light range did not show the influence of the production process temperature on their transparency (Figure 9). In the case of the tested samples, high transparency in the range of 80–90% for visible light, electromagnetic wavelengths above 400 nm was observed, which corresponds to the results achieved by earlier investigations [27,28]. Graph course for a sample deposited at a temperature of 300 °C is different from 100 °C and 200 °C. This is in line with the ellipsometer thickness results (the film produced by 300 °C is thicker than the rest) and the AFM results of the roughness parameters. The recorded hypochromic effect (downward shift) is associated with a change in thickness of films and a hypochrome effect (shift towards shorter wavelengths) which may be associated with a change in crystal size.

Based on the measurements of transparency and electromagnetic waves reflection and using the formula (2), the optical energy bandgap Eg was determined for the formed ZnO films (Figure 10):

$$\mathsf{\alpha }\mathrm{h}\mathsf{\nu }=\mathsf{\beta }(\mathrm{h}\mathrm{v}-{\mathrm{E}}_{\mathrm{g}}{)}^{\mathsf{\rho }}$$

(2)

where α is the absorption coefficient, h is the Planck constant, ν is the frequency of electromagnetic radiation, β is the constant dependent on the probabilities of electron transitions, and E_g is the energy bandgap. In the case of the ρ coefficient, the value of 0.5 was used, which corresponds to every allowed optical transition.

The obtained results are summarized in

Table 4. The optical energy bandgap of ZnO thin films.

Process Temperature	Optical Energy Bandgap, Eg (eV)
100 °C	3.64
200 °C	3.62
300 °C	3.56

. Their analysis leads to the conclusion that process temperature was slightly affected on the Eg value. The energy gap is 3.64, 3.62 and 3.56 for 100, 200, and 300 °C, respectively. In the case of investigated films, there was a tendency to increase the value of the energy bandgap with decreasing the process temperature, which is consistent with literature reports [29].

In order to characterize the adhesion of the examined ZnO films to the substrate materials, the critical load values L_C1 and L_C2 were determined by the scratch test method (Figure 11 and Figure 12). The load at which the first film defects appear is referred to as the first critical load L_C1. The critical load L_C1 is the point at which first damage is observed; the first appearance of microcracking, surface flaking outside or inside the scratch trace without exposure to the substrate material. The Lc₁ load value is related to the first cohesive damage of the film. L_C1 corresponds to the first small jump on the friction force curve on the graph. The second critical load L_C2 is the point at which complete delamination of the coatings starts; the first appearance of chipping, spallation, cracking and delamination outside or inside the scratch trace with the exposure of the substrate material. The Lc₂ load value is related to the first adhesive damage of the film. L_C2 corresponds to the point on the graph behind which the friction force curve has a disturbing run. The obtained research results are summarized in

Table 5. The adhesion of ZnO thin films was determined by a scratch test.

Process Temperature	Critical Load LC1 (N)	Critical Load LC2 (N)
100 °C	3.7	17.7
200 °C	7.4	22.6
300 °C	16.6	26.1

.

Two areas of damage could be distinguished in the resulting scratches that can use to determine the critical load values: crack Lc₁ and complete break Lc₂. The critical load value Lc₁ was determined based on light microscope optical observation of the made scratch. The critical load Lc₂ was determined based on the analysis of the friction force curve as well as light microscope optical observation.

Based on the analysis of the dependence of the friction force (F_t) on the load (F_n) (Figure 11) and most of all on the optical observation of scratch traces in ZnO thin films (Figure 12), it was ascertained that in the starting area of the scratch occurs cohesive cracks both inside and outside of the scratch, which corresponds to the value of the critical load L_C1 (Figure 12a). In the middle segment of the scratch, and across cracks (tensile cracks) and partial delamination at the edge of the scratch leading to total delamination of the film are visible, which corresponds to the value of the critical load L_C2 (Figure 12b).

Based on the determined critical loads Lc₁ and Lc₂ values, it was concluded that the temperature of the ZnO thin-films deposition process affects the degree of adhesion of the sputtered film to the substrate. As the process temperature increases, the adhesion of the film to the substrate increases. During each measurement, the acoustic emission signal was not recorded, which may prove that the bond energy value between the film and the substrate is too low.

4. Conclusions

The research results presented in this paper showed the influence of the deposition process temperature on the structure, topography, and morphology of the produced zinc oxide films, especially noticeable in the case of the films prepared at the temperature of 300 °C. This conclusion was confirmed by studies performed with the use of scanning electron microscopy, atomic force microscopy, and X-ray phase analysis. The structure of thin films sputtered at 300 °C was characterized by larger crystallites compared to the films produced at lower temperatures: 100 and 200 °C. EDS analysis authenticated the desired chemical composition of the prepared ZnO films. The test results of Raman spectroscopy additionally confirmed the correct identification of the materials. Examination of the optical properties showed a slight effect of the process temperature on it. Similarly, the process temperature was slightly affected on the energy bandgap value, which is 3.64, 3.62, and 3.56 eV for the films deposited by 100, 200, and 300 °C, respectively. An adhesion increase of the tested ZnO thin films to the substrate progressing with the increase of the process temperature was also observed.