Abstract

As we move towards a decarbonized society, transparent solar cells have recently become an important means of increasing power generation in familiar locations. Copper oxide has been attracting attention as a candidate material for the production of transparent solar cells. In order to further expand the range of applications, it is believed that it will be necessary to develop flexible, transparent plastic substrates. For the purpose of optical applications, copper oxide thin films with CuO, Cu2O, or both phases were fabricated on highly transparent plastics such as poly-methyl methacrylate (PMMA) and polycarbonate (PC), as well as on a glass substrate as a comparable sample. The helicon plasma DC magnetron sputtering method was used to fabricate thin films at a low deposition temperature due to the low heat resistance of plastic substrates. In addition, the helicon plasma DC magnetron sputtering method is capable of depositing films at low vacuum pressure, which enables the preparation of thin films with high crystallinity. The structure, surface shape, and optical properties of fabricated films were investigated using X-ray diffraction (XRD), a laser microscope, and a UV-Vis spectrophotometer, respectively. In this study, the formation of copper oxide thin films on plastic substrates at low temperatures has been verified, as have the crystal structure and optical properties, which is considered a rare study in this field.

1. Introduction

In recent years, as we move towards a decarbonized society, transparent solar cells have been gaining importance as a means of increasing power generation in familiar locations. Copper oxide, which was used as a rectifying element in the 1920s, is attracting renewed attention as a candidate material for transparent solar cells. In order to further expand the range of applications to buildings and automobiles [1], it is believed that it is necessary to develop flexible, transparent plastic substrates.

Copper oxides with CuO, Cu2O, or both phases are one of the more promising oxide semiconductors assessed by many researchers due to their good optical characteristics, such as the narrow bandgap (1.2–2.3 eV) or high transparency, and copper oxides are also a nontoxic and low-cost material. These features make copper oxide a promising material to be used in numerous applications, such as optical computers, optical memory, optical limiting devices, or optical switching elements [2, 3]. Moreover, copper oxides are used as sensors [4] or electrochromic devices [5], and their distinctive features can act as an absorber layer in hetero-junction thin-film solar cells [6].

For the above film formations, various fabrication methods have been utilized to achieve high crystallinity, such as plasma ion-assisted deposition (plasma-IAD) [7], thermal evaporation [8], activated reactive evaporation [9], molecular beam epitaxial growth [10], solution growth [11], the sol-gel process [12], electrodeposition [13], and RF magnetron sputtering. Most of those methods require processing at high temperatures. There are no known methods for producing copper oxide thin films with high crystallinity at low deposition temperatures.

In this study, copper oxide thin films with various phases were fabricated at low temperatures via reactive DC magnetron sputtering with helicon plasma. Because the helicon wave provides electromagnetic and magnetron power to the plasma, the helicon plasma can be maintained at a low gas pressure without contacting the substrate, a process referred to as long-throw sputtering [14]. Therefore, this system can fabricate thin films with an independent substrate temperature without influencing sputtering power. In addition, the helicon plasma DC magnetron sputtering method is capable of depositing films at low vacuum pressure, which enables the preparation of thin films with high crystallinity. The working pressure of helicon DC magnetron sputtering is an order of magnitude lower than that of normal magnetron sputtering. The characteristics of the fabricated thin films are greatly influenced by the fabrication parameters, such as the O2 gas flow rate, the Ar gas flow rate, or the sputtering power, which allows the fabrication of thin films with different crystalline phases [15].

In optical applications, the transparency of a substrate plays a significant role in the effectivity of optical devices [16], which can obtain wider applicability by using a high-transparency substrate. In particular, glass made of an inorganic material has high transparency and heat resistance, while organic materials with high transparency are limited but have a low melting point. Despite this, the benefits of thin films made on organic materials include flexibility and lightness. In thin film fabrication on organic materials, some problems arise, such as low crystallinity or poor adhesion between the film and material due to the limited heating temperature [7].

The copper oxide thin films of Cu2O, CuO, or both phases have been fabricated on poly-methyl methacrylate (PMMA) and polycarbonate (PC), which are key materials with transparency and are categorized as transparent thermoplastic. PMMA, also known as acrylic or acrylic glass, achieves higher transparency than PC. PC in carbonate groups has higher shock resistance, fire resistance, and heat resistance than PMMA. It is used as parts for lighting fixtures in automobiles or home appliances because it is easily molded.

The objective of this research is to deposit thin copper oxide films with high crystallinity on transparent plastic substrates. The authors have clarified the conditions under which crystalline copper oxides can be fabricated at low temperatures by using reactive DC magnetron sputtering with helicon plasma and have succeeded in fabricating them on PMMA and PC substrates. These copper oxides have an optical band gap comparable to that of crystalline semiconductors fabricated at higher temperatures. This method is expected to be applied to the fabrication of oxide thin films that require high crystallinity and low temperature formation.

2. Experimental and Measurement Methods

2.1. Experimental Methods

Reactive DC magnetron sputtering with helicon plasma was employed to prepare copper oxide thin films on PMMA and PC substrates. The substrate holder was held at 300 mm from the 99.9% pure copper target, which was Φ 2 inches in size. PMMA, PC, and glass (EAGLE XG: Corning Inc., Japan) with a size of 9 × 15 mm and 1 mm thickness were set in the substrate holder. The substrate surface was cleaned with Ar plasma for 10 min before film deposition. The film formation chamber of the multiprocess coating apparatus (ULVAC Inc., Japan) was reduced to 6.0 × 10−6 Pa in a vacuum pressure using a turbomolecular pump and a rotary pump. As shown in Table 1, copper oxide thin films were fabricated at ambient room temperature without heating and with heating to 45°C in an infrared lamp controller. The deposition temperature (actual temperature) of each substrate was measured with nonreversible temperature labels attached to the substrate because the ambient temperature of the substrate was slightly elevated due to the thermal radiation of the target being dependent on the DC sputtering power (10 W and 30 W), with the common RF helicon plasma power being 50 W. The O2 and Ar gas flow rates were kept at 10 and 15 sccm, respectively. The deposition rate that depended on the sputtering power was measured using a quartz-type thickness tester in the film formation chamber, and the formation time required to obtain a 200 nm thick film was calculated at the DC sputtering power of 10 W and 30 W. The required deposition times were 208 min and 73 min at 10 W and 30 W, respectively. The reason for choosing 10 W and 30 W DC sputtering powers is outlined in our previous study showing the formation of CuO, Cu2O, and both phases at different Cu sputtering rates [3]. In this experiment, the samples of conditions (A) and (B) were held at 10 W in DC sputtering power, and the samples of conditions (C) and (D) were held at 30 W under each formation condition.

2.2. Characterization Methods

The structure of the fabricated films was investigated by X-ray diffraction (XRD Smart Lab, Rigaku Corp.) with CuKα (0.154 nm) radiation at an incident angle of 0.3 degrees. X-ray photoelectron spectroscopy (XPS: Kratos Ultra2, Shimadzu Corp.) was used to obtain the molar ratio of Cu2O and CuO in a mixture film. A laser scanning microscope (OLS4500, Olympus Corp.) was employed to observe the surface morphology. A UV-Vis spectrophotometer (UV-2550, Shimadzu Corp.) device was used to determine the transmittance by light absorption from 400 nm to 800 nm in wavelength and the optical bandgap in wavelength from 300 nm to 900 nm.

3. Film Characterization and Discussion

3.1. XRD Measurements

Figure 1 shows the XRD patterns of the copper oxide films formed at the DC sputtering power of 10 W on the PMMA, PC, and glass substrates at two temperatures, (A) and (B), in an infrared lamp controller, respectively. The structure of CuO and Cu2O were determined by (JCPDS card, no. 05-0661) and (JCPDS card, no. 05-0667), respectively. The crystal structure of copper oxide films on different substrates showed a mixed crystal consisting of (110), (−111), and (111) planes of crystalline CuO and (111), (200), and (220) planes of Cu2O, as shown in Figure 1(a). The difference in the intensity of CuO (−111) for each substrate was highest for PMMA, followed by glass and PC, and the ratio of Cu2O/CuO of the mixture film was about 1/2 from the separated Cu2p spectra measured by XPS. At this temperature, it is assumed that the difference in the crystallinity was caused by the structure of the substrate itself, with no effect on the thermal resistance temperature of the substrate [17]. Under condition (B) of Figure 1, the Cu2O peaks disappeared, and strong CuO (−111) and (111) peaks were observed for all types of substrates because, according to the deposition temperatures measured using a nonreversible temperature label, the temperature of sample (B) was about 90°C higher than that of sample (A) because the temperature was raised to 45°C by an infrared lamp. It was confirmed that the crystallinity of the copper oxide thin film was improved at moderate substrate temperatures, even on plastic substrates. When calculated from the CuO (−111) spectrum in a simplified manner using Scherrer’s equation [18],the grain size at 38°C for the PMMA, glass, and PC were 8.3 nm, 6.4 nm, and 5.6 nm, respectively. In contrast, the values of the grain size at 90°C decreased to 10.3 nm for all substrates as deposition temperature increased. The grain size at 48°C was 11.7 nm on PMMA and 10.9 nm on PC and glass. Whereas, the grain size at 100°C was about 11.7 nm for the films on all substrates.

Figure 2 shows the XRD patterns of the copper oxide films formed at the DC sputtering power of 30 W on the PMMA, PC, and glass substrates at the two temperatures of room temperature (C) and 45°C (D), respectively, in an infrared lamp controller. At room temperature (C), the single-phase Cu2O with a strong peak of (111) emerged on each substrate, and as shown in Figure 1(a), the PMMA substrate had the highest crystallinity. Furthermore, as shown in Figure 2(b), the Cu2O (200) peak on the plastic substrates was increased by increasing the temperature of the infrared lamp controller to 45°C; however, the oxide films on the glass substrate underwent no change. The samples of conditions (A) and (B), shown in Table 1, obtained a CuO structure in the film’s crystallinity because (A) and (B) were fabricated at a lower sputtering rate (DC power of 10 W) than (C) and (D). Generally, in the reactive sputtering method, it has been confirmed that the sputtering power is the main factor controlling the copper oxide phase, as shown in our previous work, including the XRD and XPS results [3]. On the other hand, oxidation is also enhanced by the increase in the deposition temperature due to Cu’s easy bonding with oxygen molecules [19]. Via the XRD measurements, it was confirmed that each fabrication of CuO or Cu2O thin films on the plastic substrates could be performed by helicon plasma DC magnetron sputtering with a low heating temperature.

3.2. Surface Laser Micrographs

Figures 3(a)–3(d) show surface laser micrographs of copper oxide films formed at a DC sputtering power of 10 W and 30 W at room temperature and 45°C, respectively. Both thin films fabricated on PC and glass substrates had smooth surfaces and no noticeable cracks, indicating that they were homogeneous with good interfacial bonding between the substrate and the thin film, while for the thin films fabricated on PMMA substrates, cracks were observed on each surface from conditions (A) to (D). The different surfaces of PMMA and PC may be due to the difference in heat distortion temperature, which is thought to be responsible for the cracking. Incidentally, the heat distortion temperatures of PMMA and PC are 68–99°C and 121–132°C, respectively. Surface cracks on the PMMA substrate were hardly observed under conditions (A) (38°C) and (C) (48°C), measured by the nonreversible temperature label (Table 1), because the deposition temperatures were below the heat distortion temperature of PMMA, while the surface cracks of conditions (B) (90°C) and (D) (100°C) increased with the increasing deposition temperature, which was above the heat distortion temperature. Smooth and homogeneous CuO or Cu2O thin films could be fabricated on PC substrates by the helicon plasma DC magnetron sputtering method; however, in the case of the PMMA substrate, a lower temperature was required, as shown in Figures 1(a) and 1(b).

3.3. Optical Transmittance

The optical transmittance values of thin films fabricated on PMMA, PC, and glass substrates at room temperature and 45°C under 10 W sputtering power are shown in Figures 4(a) and 4(b), respectively. With the measurement wavelength increasing from 450 nm to 800 nm, the optical transmittance of all the substrates increased, showing the typical behavior of CuO. These films were CuO, with some Cu2O under condition (A) and CuO growing under condition (B), shown by the XRD measurements in Figure 1. Under fabrication condition (A), the transmittance decreased in the order of PMMA, PC, and glass substrates, which is consistent with the crystallinity (Figure 1(a)) measured by the XRD. Under the preparation condition (B), the transmittance of all the samples decreased due to the change in the crystal structure from a light black mixed phase (Cu2O and CuO) in (A) to a dark black single-phase CuO in (B).

Figures 5(a) and 5(b) show the typical optical transmittance of Cu2O thin films fabricated at room temperature and 45°C under 30 W input power. Both Cu2O thin films (C) and (D), with high crystallinity, showed similar behaviors; however, the transmittance of the PMMA, with many cracks, was slightly lower under condition (D). Via the optical transmittance measurements, it was found that CuO and Cu2O thin films can be fabricated on PMMA and PC substrates with almost the same optical properties as those on glass at the subdistortion temperature of PMMA, while the surface cracks of conditions (B) (90°C) and (D) (100°C) increased with the increasing deposition temperature, which was above the heat distortion temperature. Smooth and homogeneous CuO or Cu2O thin films could be fabricated on PC substrates by the helicon plasma DC magnetron sputtering method; however, in the case of the PMMA substrate, a lower temperature was required, as shown in Figures 1(a) and 1(b).

3.4. Optical Bandgap

To investigate the semiconductor properties of copper oxide thin films, the optical bandgap energy was determined using optical absorption spectra. The origin of the indirect bandgap in CuO is the energy gap between the conduction and valence bands, formed by the 3D orbitals of copper, whereas direct transitions can occur in Cu2O between oxygen 2p states, giving rise to the valence band and the conduction band formed by the copper 3D states [20]. The optical bandgap energy is illustrated by the plotted curve using Tauc’s equation . The indirect bandgap energy can be estimated by plotting versus the photon energy , while the direct bandgap energy is obtained by plotting versus and extrapolating the straight-line part of the plot. CuO is generally identified as a copper oxide with indirect bandgap energy. However, in this study, CuO was calculated by plotting , as reported in almost all papers, because it is difficult to separate the photon energy and the phonon energy of CuO from plotting . Figure 6 shows the CuO bandgap energy of samples fabricated under conditions (A) and (B) on PMMA, PC, and glass substrates, respectively. The single phase of CuO, fabricated by condition (B) at 45°C, showed a bandgap energy of 1.75 eV on a glass substrate and 1.8 eV on PC and PMMA substrates, while the mixture phase fabricated at room temperature showed increases in bandgap energy to 1.8–1.95 eV, as shown in Figure 6(a), as there was some Cu2O in CuO. In the case of Cu2O (Figures 7(a), 7(b)), the Cu2O thin films on all substrates obtained a bandgap energy of 2.1 eV because of their high crystallinity [21], similarly to as shown in Figures 2(a), 2(b). These results are in good agreement with the XRD measurements shown in Figures 1 and 2, and the bandgap energies obtained in this study were close to those of copper oxide thin films formed at a high temperature in our previous study [3].

4. Conclusion

Copper oxides of CuO and Cu2O phases have been fabricated on plastic substrates using the helicon plasma DC magnetron sputtering method under various conditions. In film fabrication, the important parameters are the DC input power and the deposition temperature, including the radiation temperature related to the DC input power. In particular, the deposition temperatures of substrates were kept low by long-throw sputtering using helicon plasma without affecting the plastic substrates. PMMA, PC, and glass substrates with high transparency were used to investigate the structural, surface, and optical properties. As a result, CuO and Cu2O thin films were successfully formed with high crystallinity at low temperatures on plastic and glass substrates. The thin films on PC and glass substrates had homogeneous and smooth surfaces without any cracks, as observed via laser micrographs, while the cracks in the thin films that formed on a PMMA substrate increased with the increases in deposition temperature because of the low heat distortion temperature. The optical transmittance of the CuO and Cu2O thin films showed obvious differences. The optical transmittance values of the thin films fabricated at the lowest deposition temperatures depended on the crystallinity of the CuO and Cu2O mixtures, while the thin films with high crystallinity were not affected by the substrate types. Additionally, the optical bandgap energies were influenced by the crystallinity of CuO and Cu2O, and the bandgap energy value was nearly equal to that in other reference papers. This study showed novel results in the forming of CuO and Cu2O thin films on transparent plastic substrates at a low deposition temperature with high crystal structure and a bandgap similar to the band gap of the films fabricated at a high temperature. These thin films can be used in various optical applications that require transparency or flexibility from a substrate.

Data Availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

The authors thank MEXT, Japan, and Wasit University, Iraq, for their support.