Cuprous oxide (Cu2O) films synthesis by radical oxidation with nitrogen (N2) plasma treatment and different RF power at low temperature (500°C) are studied in this paper. X-ray diffraction measurements show that synthesized Cu2O thin films grow on c-sapphire substrate with preferred (111) orientation. With nitrogen (N2) plasma treatment, the optical bandgap energy is increased from 1.69 to 2.42 eV, when N2 plasma treatment time is increased from 0 min to 40 min. Although the hole density is increased from 1014 to 1015 cm−3 and the resistivity is decreased from 1879 to 780Ωcm after N2 plasma treatment, the performance of Cu2O films is poorer compared to that of Cu2O using RF power of 0. The fabricated ZnO/Cu2O solar cells based on Cu2O films with RF power of 0 W show a good rectifying behavior with a efficiency of 0.02%, an open-circuit voltage of 0.1 V, and a fill factor of 24%.
© 2013 OSA
Cuprous oxide (Cu2O) is metal oxide semiconductor material with a cubic crystallinity structure, which has attracted much attention since 1920 due to its good room temperature mobility  and fairly high minority carrier diffusion length . Moreover, Cu2O is a p-type semiconductor material with a direct band gap energy of 2.1 eV, which has been regarded as a potential application in solar cells, because of its high absorption coefficient in the visible region . Especially, for the case of ZnO/Cu2O solar cell heterojunction [4–6], both Cu2O and ZnO are abundant, nontoxicity, and low cost. Although ZnO/Cu2O solar cell has such advantages, the reported highest conversion efficiency still has not yet exceeded 4%, which is much lower than the theoretical limit power conversion efficiency of 20%. The main reason for the low conversion efficiency is the poor quality of Cu2O ðlms, i.e., thin CuO layer forms easily on the surface during the synthesis process. For ZnO/Cu2O solar cell fabrication in these papers [5,6],usually the CuO layer is removed by wet etching method. In addition, they need to take out the Cu2O films samples from the chamber for growing ZnO films using a different growth system. We think that this is another reason for the low conversion efficiency, which results from the poor quality of the interface layer between Cu2O and ZnO fims. Thus, the synthesis of high quality of single Cu2O film without wet etching is important. Cu2O films have been fabricated using different methods such as, thermal oxidation [7,8], chemical vapor deposition , electrodeposition ,and reactive sputtering . However, so far, there are few reports of using radical oxidation for Cu2O synthesis and nitrogen doping effect in Cu2O films [12,13].In most of these reports, the thickness of original Cu film is usually between 50~130 nm, which induces the thickness of Cu2O films less than 300 nm. For the thick Cu films oxidation, it is still difficult to obtain high pure Cu2O without traces of CuO and Cu, resulting in low conversion efficiency in solar cells. Therefore, we attempt to fabricate single Cu2O films by radical oxidation of Cu films at low temperature. Moreover, to overcome the above-mentioned issues, we grow the ZnO on Cu2O films directly in the same growth system without opening the chamber cover.
In this report, single Cu2O films are demonstrated by radical oxidation of Cu films with N2 plasma surface treatment and different RF power of oxygen (O2) at low temperature of 500°C. The effects of N2 plasma treatment and different RF power for Cu2O films on electrical and optical prosperities are studied. It is found that the N2 doping can efficiently widen the bandgap of Cu2O films. In addition, the crystal structure and morphology of the Cu2O films are investigated by x-ray diffraction (XRD) and scanning electron microscope (SEM). The rectification properties of the fabricated ZnO/Cu2O heterojunction solar cells are also investigated.
2. Experimental details
Synthesis of Cu2O films was carried out by radical oxidation of high purity Cu films (99.9%). Cu ðlms of 500 nm thicknesses with 6 × 6 mm2 size were deposited on c-sapphire substrate by a thermal evaporation under high vacuum of 10−3 Pa. The total experimental process consisted of four steps. The first step was annealing the Cu films to 500°C with a rate of 33°C/min in the hydrogen (H2) plasma atmosphere (0.04 Torr). In the annealing process, the flow rate and RF power of H2 were 20 sccm and 60 W, respectively. The second step was oxidation process: Cu ðlms were exposed to O2 plasma for 90 min at 500°C in the remote –plasma-enhanced metalorganic chemical vapor deposition (RPE-CVD) system, in which the O2 plasma power was varied from 0 to 30 W and the flow rate was kept 50 sccm, respectively, with a pressure of 0.11 Torr. The oxidation time of 90 min was necessary to promote the Cu films fully oxidized. For the RF = 0 W, the surface color of Cu2O always kept red. However, during the oxidation process with RF > 0 W, we found that Cu2O was formed first and after a sufficiently long oxidation time, black color of thin CuO layer was formed. To remove the thin CuO layer and investigate the N2 doping effect in the Cu2O films, we carried out the following step. The third step was N2 plasma treatment process: the surface of films was treated with different time using N2 plasma (60 W, 50 sccm). After N2 plasma treatment, ZnO layers were deposited directly on the Cu2O by RPE-CVD system at 500°C. The flow rate of the O2 reaction gas and Zn were fixed at 20 sccm and 8 μmol/min. The flow rate of the N2 pushing gas was fixed at 20 sccm. The ZnO growth time was 10 min. Finally, the N2 gas with a flow rate of 150 sccm was introduced in the chamber and the substrate temperature was gradually decreased to room temperature.
The crystallinity of Cu2O films was investigated using x-ray diffraction (XRD) measurements. The optical properties and electrical properties of the films were characterized by optical transmittance and Hall effect measurement, respectively. The transmittance spectra were recorded on an UV-visible-near infrared scanning spectrometer over a wavelength range of 2000-200 nm. The optical band gap energy was determined by linear extrapolation in a Tauc plot from the transmittance spectra. The surface morphology of the films was studied using a scanning electron microscope (SEM).
3. Results and discussion
3.1. Radical oxidation with N2 plasma treatment
We fabricated Cu2O sample using two different methods. One method is using radical oxidation (in Section 3.1), the other one is oxygen oxidation (in Section 3.2). Because the black CuO was easily generated during the oxidation process using the radical oxidation method, we used the N plasma treatment to remove this very thin CuO. Compared with oxygen gas, radical oxidation is expected to have a high energy. So for the case of radical oxidation, excited atomic oxygen and atomic ions which easily react with the surface of Cu film.
The photograph of the Cu2O films synthesized at 500°C with O2 RF power of 30 W and different N2 plasma treatment time ranging from 0 to 40 min is shown in Fig. 1(a). The color of the Cu2O samples without N2 plasma treatment is black, as the thin black CuO formed on the surface after oxidation. However, with the N2 plasma treatment time increase, the color of the samples is changed from light blue to red, and the appearance of these samples is changed from opaque into translucent, which indicates that the transformation of CuO into Cu2O occurred. Figure 1(b) shows the XRD results of Cu2O film samples with different N2 plasma treatment time. There is no Cu diffraction peak in all the samples. Two different CuO diffraction peaks with (200) and (−111) planes can be observed in the sample without N2 plasma treatment. When the N2 plasma treatment time increase from 0 min to 20 min, Cu2O (111) peak intensity becomes stronger, since the CuO peak intensitybecomes weak and finally disappeared. However, as the N2 plasma treatment time further increased, the Cu2O (111) peak intensity has a trend to decrease.
Figure 2(a) shows the transmittance results. It can be seen clearly that there is a distinct difference between the optical absorption edges for all the Cu2O films. The cut-off wavelength of optical transmittance is change from 700 nm to 510 nm, when the N2 plasma treatment time increases from 0 minute to 40 minutes. The Cu2O film without N2 plasma treatment exhibits sharper absorption edge than other Cu2O films with N2 plasma treatment. In addition, the transmittance has a reduction with the N2 plasma treatment time increase in the infrared region, which may be attributed to scattering on the film surface. It is found that, as the N2 plasma treatment time increase, high roughness appears on the film surface, which inducing scattering losses increase . To confirm this absorption edge changes, the optical bandgap energies of the Cu2O films were calculated. The optical bandgap energies were determined from a plot of (αhν)2 as a function of photon energy (hν) is shown in Fig. 2(b). A distinct difference bandgap of the samples can be clearly seen. For the samples without N2 plasma treatment, the bandgap is about 1.7 eV, which is a little larger than the CuO film with a bandgap of 1.5 eV ; this is due to the mixture of CuO and Cu2O appearance. It is found that the optical bandgap of the Cu2O films with N2 plasma treatment time of 40 minutes is 2.42 eV, which is a little larger compared to the published result. This phenomenon was also reported in the N2 doping into Cu2O films by using reactive sputtering method . In the , Nakano et al. demonstrated that with increasing N-doping concentration up to 3%, the optical band gap energy was enlarged from 2.1 to 2.5 eV. They attributed this phenomenon to the rich N content in the ðlms, which is probably associated with the structural change in Cu2O induced by N doping
To investigate the electrical properties of the Cu2O films, we made a Hall effect measurement with Van der Pauw configuration at room temperature. The dependence of the carrier concentration, carrier mobility, and resistivity on the N2 plasma treatment time are shown in Fig. 3(a) and Fig. 3(b), respectively. All the Cu2O films present a p-type semiconductor feature, which reveals the fact that the obtained Cu2O films are all p-type semiconductors. As can be seen from these figures, the mobility decreases and carrier concentration increases with increasing the N2 plasma treatment time from 0 minute to 40 minutes. In the case of Cu2O films treated by N plasma for 40 minutes, the carrier concentration is ~1015, which has one order of magnitude increase, compared to other Cu2O films. This result indicates that the nitrogen contributes to increasing the hole density, since the nitrogen acts as an acceptor in the doping process. It is reasonable that the nitrogen in Cu2O can be incorporated in the oxygen lattice site, and this may be the reason for the nitrogen acting as an acceptor. At the same time, as the carrier concentration increases, the mobility has a significant decrease from 28 to several. The mobility of crystalline thin film is related with carrier scattering mechanisms. The decrease in mobility with increasing N2 plasma treatment time may be due to carrier scattering, which is increased by ionized acceptors. In Fig. 3(b), the resistivity decreases with increasing the N2 plasma treatment time, and the Cu2O film (40 min N2 plasma treatment) exhibits the lowest resistivity of 780 Ωcm. This decrease of resistivity can be mainly attributed to an increase of additional carriers resulting from N-doping atoms are easily ionized. These results indicate that the nitrogen is very effective in controlling the electrical properties of Cu2O films.
3.2 Radical oxidation with different RF power
In order to obtain a good understanding of the radical oxidation behavior of Cu films, different RF power of O2 was performed. During the oxidationprocess with RF power of 0 W, the surface color of the films was always kept red. However, for the oxidation process with different RF power of 10, 20, 30 W, the red color was firstly appeared on the surface, and it was changed into black color a moment later. When the RF is 0, it means that there is no plasma generated. We can directly change the RF power of plasma by control the different voltage. Figure 4 shows the XRD patterns of the measured samples at various RF powers of 0, 10, 20, 30 W. The XRD diffraction pattern shows that a single diffraction peak phase of Cu2O was obtained for the RF power of 0 W. Although Cu2O diffraction peaks were also observed for the RF power of 10, 20 and 30W, black CuO diffraction peaks appeared at the same time. Inaddition, the diffraction peak of Cu2O at RF = 0 is stronger compared to that of the other cases.
Figure 5(a) shows the carrier concentration and Hall mobility of the films as function of O2 RF power. As it can be seen, the hole density decreases nearly linearly with the RF power increases. Highest density of more than 1015 cm−3 was obtained for the RF power of 0 W. Correspondingly, the Hall mobility decreased with increasing the RF power. As shown in Fig. 5(b), the resistivity is reduced to less than 120 Ohm-cm when the RF power is 0 W. This result is almost one order of magnitude reduced, compared to RF power with 10, 20, 30 W. One of the possible reasons for the higher resistivity of the Cu2O films with RF >0 maybe the reduction in the presence of ionized copper vacancies as well as the densifications of the films due to the bombardment by the oxygen plasma. It is concluded that the O2 RF power magnitude is also effectively in controlling the electrical properties of the Cu2O films. Comparing the radical oxidation with N2 plasma treatment and RF power of 0 W, it is found that the performance of Cu2O is greatly improved when the RF power is 0 W.
3.3. PV characteristics of ZnO/Cu2O heterojunction
Two different kinds of the ZnO/Cu2O heterojunction solar cells were fabricated. One is based on Cu2O with N2 plasma treatment time of 40 min, the other one is based on Cu2O with RF power of 0 W. ZnO layers were directly deposited on Cu2O films for 10 min by RPE-CVD system without opening the chamber cover. The thickness of the deposited ZnO for both solar cells was 200 nm. Figure 6 shows the current density (J-V) characteristics of the ZnO/Cu2O heterojunction solar cells measured in the dark and underillumination intensity of 36 mW/cm2 at room temperature. As can be seen from Fig. 6(b), the J-V characteristics of the ZnO/Cu2O heterojunction solar cells fabricated by RF power of 0 W exhibit a significant rectify behavior, which indicates that a p-n junction is formed between the interfere layer of ZnO and Cu2O. The reproducibility was also confirmed for this solar cell. However, as shown in Fig. 6(a), only an ohmic behavior was obtained for the ZnO/Cu2O heterojunction solar cells with N2 plasma treatment time of 40 min. The reasons for the difference performance of these two kinds of solar cells will be explained by combing the morphology analysis. Although the current of the heterojunction solar cell was very small under the dark condition, the optical illumination generated a relatively large current. The measured short-circuit current density (Jsc), open-circuit voltage (Voc), and fill factor (FF) of the solar cell are 0.03 mA/cm2, 100 mV, and 24%, respectively, which can be obtained from the Fig. 7. The photovoltaic energy conversion (0.02%) is not so high compared to the published papers .The main reason for this low efficiency is because of the ZnO growth on Cu2O at high temperature of 500°C, which may induce the performance of Cu2O poor during the ZnO process. Although the efficiency is not so high, we develop a new method for ZnO/Cu2O heterojunction solar cells fabrication, i.e., ZnO directly grow on Cu2O without wet etching.
The morphology of the Cu2O ðlms was analyzed by the SEM. Figure 8 shows the SEM micrographs of Cu2O ðlms synthesized by radical oxidation with N2 plasma treatment time of 40 min and RF power of 0 min. It can be seen clearly that N2 plasma treatment has an obvious effect on the surface morphology. As shown in Fig. 8(a), the surface morphology of the Cu2O ðlmsis irregular with roughness when the N2 plasma treatment is introduced. This is the reason why there is no p-n junction formation for the solar cell with N2 plasma treatment. However, it can be seen from the Fig. 8(b), the surface is compact and smooth, no roughness is observed. The difference in film morphology for these films can be associated with the activation energy of N2 plasma for crystal structure formation. The rough surface with N2 plasma treatment may be a result of surface diffusion of the N atom as an acceptor that eventually leads into Cu2O films. The morphology of the Cu2O ðlms in this paper shows the similar results with the previous reports on the nanocrystalline Cu2O films prepared by using the activated reactive evaporation technique . In a word, if ZnO is deposited on Cu2O at room temperature, we still think that higher efficiency of solar cell with RF power of 0 W and short N2 plasma treatment time can be expected
Synthesis of Cu2O films by radical oxidation of Cu films at low temperature (500°C) was investigated. Single Cu2O phase ðlms with high quality (111) orientation can be obtained by N2 plasma treatment and RF power of 0 W. The experimental results indicate that different N2 plasma treatment time has a significant influence on the optical transmittance, optical bandgap and electrical properties of the films. By varying the N2 plasma treatment time from 0 min to 40min, the transmittance has a reduction in the infrared region and the bandgap has a blues shift from 1.69 to 2.42 eV. The resistivity of the Cu2O films is reduced to780Ωafter 40 min N2 plasma treatment. However, the performance of Cu2O films with RF power of 0 W is better than the Cu2O films with N2 plasma treatment. The ZnO/Cu2O solar cells based on Cu2O films with RF power of 0 show a good rectify behavior with a efficiency of 0.02%, an open-circuit voltage of 0.1 V, and a fill factor of 24%. Although the efficiency is not so high, we develop a new method for ZnO/Cu2O heterojunction solar cells fabrication, i.e., ZnO directly grow on Cu2O without wet etching.
References and links
1. B. S. Li, K. Akimoto, and A. Shen, “Growth of Cu2O thin films with high hole mobility by introducing a low-temperature buffer layer,” J. Cryst. Growth 311(4), 1102–1105 (2009). [CrossRef]
2. S. S. Jeong, A. Mittiga, E. Salza, A. Masci, and S. Passerini, “Electrodeposited ZnO/Cu2O heterojunction solar cells,” Electrochim. Acta 53(5), 2226–2231 (2008). [CrossRef]
3. I. Grozdanov, “Electroless chemical deposition technique for Cu2O thin films,” Mater. Lett. 19(5-6), 281–285 (1994). [CrossRef]
4. P. Wang, X. H. Zhao, and B. J. Li, “ZnO-coated CuO nanowire arrays: fabrications, optoelectronic properties, and photovoltaic applications,” Opt. Express 19(12), 11271–11279 (2011). [CrossRef] [PubMed]
5. A. Mittiga, E. Salza, F. Sarto, M. Tucci, and R. Vasanthi, “Heterojunction solar cell with 2% efficiency based on a Cu2O substrate,” Appl. Phys. Lett. 88(16), 163502 (2006). [CrossRef]
6. T. Minami, T. Miyata, K. Ihara, Y. Minamino, and S. Tsukada, “Effect of ZnO film deposition methods on the photovoltaic properties of ZnO–Cu2O heterojunctiondevices,” Thin Solid Films 494(1–2), 47–52 (2006). [CrossRef]
7. A. O. Musa, T. Akomolafe, and M. Carter, “Production of cuprousoxide, asolarcell material, by thermal oxidation and a study of its physical and electrical properties,” Sol. Energy Mater. Sol. Cells 51(3–4), 305–316 (1998). [CrossRef]
8. V. Figueiredo, E. Elangovan, G. Goncalves, P. Barquinha, L. Pereira, N. Franco, E. Alves, R. Martins, and E. Fortunato, “Effect of post-annealing on the properties of copper oxide thin films obtained from the oxidation of evaporated metallic copper,” Appl. Surf. Sci. 254(13), 3949–3954 (2008). [CrossRef]
9. S. H. Jeong and E. S. Aydil, “Heteroepitaxial growth of Cu2O thin film on ZnO by metal organic chemical vapor deposition,” J. Cryst. Growth 311(17), 4188–4192 (2009). [CrossRef]
10. S. Kang, S. Hong, J. Choi, J. Kim, I. Hwang, I. Byun, Y. Kim, W. Kim, and B. Park, “Layer-to-island growth of electrodeposited Cu2O films and filamentary switching in single-channeled grain boundaries,” J. Appl. Phys. 107(5), 053704 (2010). [CrossRef]
12. C. C. Ooi and G. K. L. Goh, “Formation of cuprous oxide films via oxygen plasma,” Thin Solid Films 518(24), e98–e100 (2010). [CrossRef]
13. H. J. Li, C. Y. Pu, C. Y. Ma, S. Li, W. J. Dong, S. Y. Bao, and Q. Y. Zhang, “Growth behavior and optical properties of N doped Cu2O films,” Thin Solid Films 520(1), 212–216 (2011). [CrossRef]
14. D. Mardare and G. I. Rusu, “The influence of heat treatment on the optical properties of titanium oxide thin films,” Mater. Lett. 56(3), 210–214 (2002). [CrossRef]
15. W. Seiler, E. Millon, J. Perrière, R. Benzerga, and C. Boulmer-Leborgne, “Epitaxial growth of copper oxide films by reactive cross-beam pulsed-laser deposition,” J. Cryst. Growth 311(12), 3352–3358 (2009). [CrossRef]
16. Y. Nakano, S. Saeki, and T. Morikawa, “Optical bandgap widening of p-type Cu2O films by nitrogen doping,” Appl. Phys. Lett. 94(2), 022111 (2009). [CrossRef]
17. S. H. Jeong, S. H. Song, K. Nagaich, S. A. Campbell, and E. S. Aydil, “An analysis of temperature dependent current-voltage characteristics of Cu2O/ZnO heterojunction solar cells,” Thin Solid Films 519(19), 6613–6619 (2011). [CrossRef]
18. B. Balamurugan and B. R. Mehta, “Optical and structural properties of nanocrystalline copper oxide thin films prepared by activated reactive evaporation,” Thin Solid Films 396(1–2), 90–96 (2001). [CrossRef]