We have demonstrated that the bandgap of Cu(In1-xGax)Se2(CIGS) absorber layers was readily controlled by using a one-step sputtering process. CIGS thin-film sample libraries with different Ga/(In + Ga) ratios were synthesized on soda-lime glass at 550°C using a combinatorial magnetron sputtering system employing CuInSe2(CIS) and CuGaSe2(CGS) targets. Energy-dispersive X-ray fluorescence spectrometry (EDS-XRF) confirmed that the CIGS films had different Ga/(In + Ga) ratios, which were varied by the sample configuration on the substrate and ranged from 0.2 to 0.9. X-ray diffraction and Raman spectroscopy revealed that the CIGS films had a pure chalcopyrite phase without any secondary phase such as Cu-Se or ordered vacancy compound (OVC), respectively. Furthermore, we found that the optical bandgap energies of the CIGS films determined by transmittance measurements ranged from 1.07 eV to 1.53 eV as the Ga/(In + Ga) ratio increased from 0.2 to 0.9, demonstrating that the one-step sputtering process using CIS and CGS targets is another simple route to control the bandgap energy of the CIGS absorber layer.
© 2016 Optical Society of America
Cu(In1-xGax)Se2(CIGS) chalcopyrite compounds have been studied extensively for thin-film solar cells due to their high absorption coefficient (1⨯105 cm−1) and controllable bandgap energy (1.04~1.67 eV) [1, 2, 32–34]. CIGS solar cells made by co-evaporation have shown record cell efficiencies exceeding ~20% in the laboratory scale [3–7]. To get such higher cell efficiency, the characteristics of CIGS absorber layers—including the structural, electrical, and optical properties—should be carefully optimized. Above all, bandgap engineering of CIGS absorber layers is very effective in improving cell performance, because the useful solar spectrum is determined by the absorption edge of the absorber layer. In addition, the technique of bandgap engineering can make bandgap grading in an absorber layer, which is necessary to get a higher efficiency by reducing the recombination losses and amplifying the carrier collection in the cell . Indeed, co-evaporation is one of the preferred methods to control the bandgap of the CIGS absorber layer, demonstrating why the record efficiency was achieved using this method [9,10].
On the other hand, although the sputtering process has several advantages such as large-area deposition and high deposition rates, the precise control of the Ga/(In + Ga) ratio to engineer the bandgap of the CIGS absorber layer is relatively difficult [11–16]. In particular, bandgap engineering of CIGS absorber layers by the conventional sputtering process, which employs a post-selenization process, requires great sophistication to optimize the entire process [17,18]. Because bandgap engineering of the CIGS absorber layer is necessary for high conversion efficiency, a simpler and cost-effective process should be developed to adjust the Ga/(In + Ga) ratio to control the bandgap.
In this study, we suggest a one-step sputtering process in which CuInSe2 (CIS) and CuGaSe2 (CGS) targets are used to fabricate CIGS films with an adjusted bandgap. By using the one-step sputtering process, we can systematically and easily control the bandgap of the CIGS absorber layer, showing the method to be a relatively simple technique to create bandgap grading.
The CIGS absorber layers were deposited on soda-lime glass (SLG) substrates using a combinatorial magnetron sputtering system employing CIS and CGS targets with a 1:1:2 stoichiometry, respectively. The targets employed had a 4-inch diameter and were purchased from High Purity Chemical Co. Japan. Deposition of CIGS absorber layers using one-step sputtering, in which CIGS absorber layers can be directly synthesized from compound chalcopyrite targets without an additional selenization process, has been well established as reported elsewhere [19–22]. In our case, CIGS absorber layers having a continuous gradient in Ga/(In + Ga) ratios have been deposited during a single batch. Using this technique, we made CIGS sample libraries having different Ga/(In + Ga) ratios on the SLG substrate, as shown in Fig. 1. The Ga/(In + Ga) ratio in the CIGS absorber layers were varied depending on the location on the substrate. In this study, bare SLG substrates instead of Mo-coated SLG were used for characterizing the bandgap directly from transmittance measurements. The deposition conditions previously optimized were as follows: substrate temperature, 550°C; ambient gas, Ar; gas flow rate, 30 sccm; process pressure, 0.43 Pa; RF power, 300 W; and distance from target to substrate, 150 mm. Five kinds of CIGS film samples with a size of 12 x 12 mm2 have been deposited on the substrates with a size of 180 x 50 mm2 using a patterned shadow mask.
The crystalline structures of CIGS absorber layers were examined by an X-ray diffractometer (XRD, Panalytical X’pert MPD). Surface morphology and cross-section images were analyzed by using a field-emission scanning electron microscope (FE-SEM, Quanta 200). Chemical compositions were determined by energy-dispersive X-ray fluorescence spectrometry (EDX-XRF, SHIMADZU). The phase analysis was performed by Raman spectroscopy having a He-Ne laser with 632-nm wavelength (JOVIN YVON SAS). Optical properties of films were measured using a UV-VIS-IR spectrometer (Varian Cary 5000).
3. Results and discussion
Table 1 shows the chemical composition of the CIGS samples determined by EDS-XRF, indicating that each sample has a significantly different Ga/(In + Ga) ratio while maintaining Cu/(In + Ga) and Se/(Cu + In + Ga) ratios. The Ga/(In + Ga) ratios of the samples are changed from 0.2~0.9 as a function of the location on the substrate. These results can be explained by the fact that the flux of sputtered particles from the each target have an angular dependence. Therefore, the CIGS sample libraries deposited on the SLG substrate have a continuous gradient in Ga/(In + Ga) ratios depending on the lateral distance from the target normal. These gradient in Ga/(In + Ga) ratios have been formed spontaneously under the configuration of the combinatorial sputtering system. Additionally, we have checked the ununiformity of the Ga/(In + Ga) ratios in each samples themselves by repeated measurements, however, the deviations of Ga/(In + Ga) ratios in each samples themselves were negligible.
3.2 SEM images
Figure 2(a) shows that the thickness of CIGS thin films range from 0.8 to 1.0 µm, which appear to be dense and have excellent adhesion to the SLG substrate for all samples. However, the surface morphology is remarkably different as a function of Ga/(In + Ga) ratio even though all the CIGS samples are deposited at the constant substrate temperature of 550°C. Previously, we checked the surface morphologies of ternary thin films, which have the same growth conditions as the co-sputtered CIGS thin films, as shown in Fig. 2(b). The CGS thin films showed the densely packed grains without any void;but had small grains in the surface morphology. On the contrary, the CIS thin films showed the considerable improvement in grain sizes above 1µm scale. Despite the big differences in the morphological properties, no structural and optical changes was found in each film. The CIGS samples (A-D) co-deposited by using CIS and CGS targets, which have a Ga-rich characteristic, show a smoother surface and compact microstructure, whereas the CIGS sample (E) with Ga/(In + Ga) ratio of 0.2 exhibits irregular and facet-shaped structure. In regard to effects on Ga contents, it was reported that the CIGS thin films has smaller grains at the higher Ga content compared to the films grown at the low Ga content [23, 30]. This shows that the CIGS morphologies varied with Ga content in the CIGS thin films, showing that the CIGS samples (A-D) with high Ga content has the similar morphology like a CGS thin film.
3.3 XRD patterns
The Fig. 3(a) shows the XRD patterns of CIGS thin films as a function of Ga/(In + Ga) ratio deposited at a substrate temperature of 550°C. All the CIGS thin films reveal dominant (112) peaks, including chalcopyrite characteristic peaks of (220), (204), and (312). Also, no secondary peaks indicating Cu-Se or OVC (ordered vacancy compound) phases are observed in the CIGS thin films despite using two different kinds of sputtering targets (CIS and CGS) at the same time. The slight peak split of CGS film is confirmed to be the distortion of lattice constant in the tetragonal CIGS structure due to high Ga content in CIGS thin films but not secondary phases . In addition, the dominant (112) peaks are almost linearly shifted to high diffraction angle with increasing Ga/(In + Ga) ratio, as shown in Fig. 3(b). This is attributed to the decrease of the lattice parameter in the chalcopyrite CIGS structure by substituting In atoms with smaller Ga atoms. According to Scherrer equation (τ = Kλ/βcosθ), the crystallite size can be calculated by full-width at half-maximum (FWHM) values. The crystallite size of the CIGS thin films for (112) plane is ranged from 12 to 23nm and is slightly decreased at the Ga contents of 0.5 (sample D) and 0.7(sample C). This slight deviation can be explained by the fact that fluxes of particles are reduced at the center of substrate region. In other words, the main plasmas are located near the sputtering target normal, as mentioned in Fig. 1. The sample C and D were located at the place where fluxes of particles sputtered from both targets were the lowest due to the angular dependence of sputtering fluxes. For that reason, it is considered that the grain growth is slightly affected by the sputtered particles with low fluxes at the center of substrate region.
3.4 Raman spectra
Raman spectroscopy was also used to check whether or not the CIGS thin films with different composition ratios of Ga/(In + Ga) are only chalcopyrite single phase. Generally, the frequency of the A1 mode in Cu(In1-xGax)Se2 films shifts linearly from 173 cm−1 to 186 cm−1 with varying x from 0 to 1. According to the report of Roy et al. , the Raman shift caused by Ga content in CIGS thin films can be estimated by F = 173 + 12.92x, where F is the Raman shift value of A1 mode and x is the composition ratio of Ga/(In + Ga). This formula means that Raman frequency increases by substituting In sites with Ga atoms which have a lower mass than In atoms, resulting in a larger bond length formed in the lattice site . In our case, we confirmed that the A1 mode frequency of CIGS films shifts from 173.4 cm−1 to 186.1 cm−1 with increasing Ga/(In + Ga) ratio from 0 to 1, as shown in Fig. 4(a), respectively. To confirm this relationship, composition ratio and A1 scattering peaks are depicted in Table 2. Raman shift values of A1 mode peaks are generally well matched with Ga/(In + Ga) ratios, but show a small deviation in sample B and C. Regarding the compositional behavior in physical property, accurate composition values should be accompanied but the Ga/(In + Ga) ratios determined by EDS-XRF measurements are not perfectly adjusted with above equation due to uncertainty of composition analysis, leading similar scattering peak positions in spite of different sample positions. However, it is clearly confirmed that A1 mode peak positions are systematically distributed as a function of Ga/(In + Ga) ratio, which is not fully consistent with composition ratio. In addition, the FWHM values of all the CIGS thin films range from 9.2 to 12.5 cm−1, demonstrating the high quality of CIGS thin films. Furthermore, B2 and E1 modes of the CIGS thin films are also observed, which correspond to the vibration mode of all CIGS atoms. Consequently, one-step sputtered CIGS thin films show only chalcopyrite characteristic peaks (A1, B2, E1 modes) regardless of Ga/(In + Ga) ratio in the CIGS films without any secondary phase.
3.5 Optical bandgaps
The optical bandgap energy (Eg) is studied by using transmittance, reflectance spectra from the CIGS absorber layer on a SLG substrate. The optical coefficient (α) can be deduced from the following equation :28]:Figure 5 indicates the plot of (αhv)2 versus hv for CIGS thin films, and the inset shows the optical bandgap energies with different Ga/(In + Ga) ratios. Generally, the CIS and CGS thin films having stoichiometric composition exhibit the bandgap energy of 1.02 and 1.62 eV, respectively . In our case, CIS and CGS films deposited at the substrate temperature of 550°C show the bandgaps of 0.98 and 1.62 eV, respectively. By using these stoichiometric single targets, we confirmed that the CIGS thin films exhibit the optical bandgaps of 1.07, 1.24, 1.35, 1.47, and 1.53 for the Ga content of 0.2, 0.5, 0.7, 0.8, and 0.9, respectively. In the previous study, it was reported that the bandgaps of (CuIn1-xGaxSe2) thin films grown by co-evaporation technique are followed by the relationship between x (Ga content) and Eg(x) = 0.998 + 0.291x + 0.43x2 eV . Figure 5(b) shows the variation of bandgaps as a function of Ga content in the reference (red circle) and co-sputtered CIGS thin films (black square), respectively. The optical bandgaps of co-sputtered CIGS thin films increase with the Ga content and have a good agreement with the expected bandgaps at each the Ga content. Nevertheless, a slight deviation in the bandgaps was observed as the Ga content exceeding 0.7, which can be caused by the inaccuracy in the determination of compositions between the two studies. However, it should be noted that the bandgaps correspond approximately with the Ga content in the CIGS thin films, which have sharp absorption edges with fewer defects and good crystallinity for all the samples.
In summary, we have introduced a systematic approach to obtain CIGS films with different (In,Ga) ratios by one-step sputtering without a selenization process using a combinatorial magnetron sputtering system from a CIS/CGS stoichiometric single target. The CIGS thin films show that the composition ratio of Ga/(In + Ga) consistently changed from 0.2 to 0.9 depending on the distance from the substrate to each sputtering target (CIS and CGS). XRD patterns reveal that diffraction peak positions are linearly shifted to high angle for both (112) and (220) planes with increasing Ga/(In + Ga) ratio. Raman spectra also exhibit similar properties to the XRD patterns, demonstrating high-quality CIGS thin films with A1 mode scattering peak in each composition ratio. Moreover, the optical property shows that bandgap energies are also consistent with the composition ratio of Ga/(In + Ga), while maintaining the standard bandgap energy.
Korea Institute of Industrial Technology (KITECH) (EO160024); National Renewable Energy Laboratory (NREL) (DE-AC36-08GO28308).
References and links
1. M. G. Park, S. J. Ahn, J. H. Yun, J. Gwak, A. Cho, S. K. Ahn, K. Shin, D. Nam, H. Cheong, and K. Yoon, “Characteristics of Cu(In, Ga)Se2 (CIGS) thin films deposited by a direct solution coating process,” J. Alloys Compd. 513, 68–74 (2012). [CrossRef]
2. A. J. Han, Y. Zhang, W. Song, B. Li, W. Liu, and Y. Sun, “Structure, morphology and properties of thinned Cu(In,Ga)Se2 films and solar cells,” Semicond. Sci. Technol. 27(3), 035022 (2012). [CrossRef]
3. H. Zachmann, S. Heinker, A. Braun, A. V. Mudryi, V. F. Gremenok, A. V. Ivaniukovich, and M. V. Yakushev, “Characterization of Cu(In,Ga)Se2-based thin film solar cells on polyimide,” Thin Solid Films 517(7), 2209–2212 (2009). [CrossRef]
4. A. Hultqvist, C. Platzer-Bjorkman, J. Pettersson, T. Törndahl, and M. Edoff, “CuGaSe2 solar cells using atomic layer deposited Zn(O,S) and (Zn,Mg)O buffer layers,” Thin Solid Films 517(7), 2305–2308 (2009). [CrossRef]
5. L. Zhang, Q. He, W. L. Jiang, F.-F. Liu, C.-J. Li, and Y. Sun, “Effects of substrate temperature on the structural electrical properties of Cu(In,Ga)Se2 thin films,” Sol. Energy Mater. Sol. Cells 93(1), 114–118 (2009). [CrossRef]
6. D. G. Moon, S. J. Ahn, J. H. Yun, A. Cho, J. Gwak, S. K. Ahn, K. Shin, K. Yoon, H.-D. Lee, H. Pak, and S. Kwon, “Ex-situ and in-situ analyses on reation mechanism of CuInSe2(CIS) formed by selenization of sputter deposited CuIn precursor with Se vaper,” Sol. Energy Mater. Sol. Cells 95(10), 2786–2794 (2011). [CrossRef]
7. I. H. Choi, “Raman spectroscopy of CuIn1-xGaxSe2 for in-situ monitoring of the composition ration,” Thin Solid Films 519(13), 4390–4393 (2011). [CrossRef]
8. N. H. Rafat and S. E.-D. Habib, “The limiting efficiency of band gap graded solar cells,” Sol. Energy Mater. Sol. Cells 55(4), 341–361 (1998). [CrossRef]
9. T. Drobiazg, L. Arzel, A. Donmez, P. Zabierowski, and N. Barreau, “Influence of indium/gallium gradients on the Cu(In,Ga)Se2 devices deposited by the co-evaporation without recrystallization,” Thin Solid Films 582, 47–50 (2015). [CrossRef]
10. T. Klinkert, M. Jubault, F. Donsanti, D. Lincot, and J.-F. Guillemoles, “Differential in-depth characterization of co-evaporated Cu(In,Ga)Se2 thin films for solar cell applications,” Thin Solid Films 558, 47–53 (2014). [CrossRef]
11. G. S. Chen, J. C. Yang, Y. C. Chan, L. C. Yang, and W. Huang, “Another route to fabricate single-phase chalcogenides by post-selenization of Cu–In–Ga precursors sputter deposited from a single ternary target,” Sol. Energy Mater. Sol. Cells 93(8), 1351–1355 (2009). [CrossRef]
12. S. U. Park, R. Sharma, and C. R. Lee, “A study on composition, structure and optical properties of copper-poor CIGS thin film deposited by sequential sputtering of CuGa/In and In/(CuGaþIn) precursors,” J. Cryst. Growth 359, 1–10 (2012). [CrossRef]
13. Zs. Baji, Z. Labadi, G. Molnar, B. Pécz, A. L. Tóth, J. Tóth, A. Csik, and I. Bársony, “Post-selenization of stacked precursor layers for CIGS,” Vacuum 92, 44–51 (2013). [CrossRef]
14. H. K. Song, S. G. Kim, and H. J. Joon, “Preparation of CuIn1-xGaxSe2 thin films by sputtering and selenization process,” Sol. Energy Mater. Sol. Cells 75(1-2), 145–153 (2003). [CrossRef]
15. C. Y. Su, W. H. Ho, H. C. Lin, C.-Y. Nieh, and S.-C. Liang, “The effects of the morphology on the CIGS thin films prepared by CuInGa single precursor,” Sol. Energy Mater. Sol. Cells 95(1), 261–263 (2011). [CrossRef]
16. S. J. Park and J. H. Kim, “Structural analysis of Cu(In,Ga)Se2 films fabricated by using sputtering and post-selenization,” Curr. Appl. Phys. 13(6), 1046–1049 (2013). [CrossRef]
17. K. Kushiya, M. Tachiyuki, T. Kase, I. Sugiyama, Y. Nagoya, D. Okumura, M. Sato, O. Yamase, and H. Takeshita, “Fabrication of graded band-gap Cu(InGa)Se2 thin-film mini-modules with a Zn(O,S,OH)x buffer layer,” Sol. Energy Mater. Sol. Cells 49(1-4), 277–283 (1997). [CrossRef]
18. Y. Nagoya, K. Kushiya, M. Tachiyuki, and O. Yamase, “Role of incorporated sulfur into the surface of Cu(InGa)Se2 thin-film absorber,” Sol. Energy Mater. Sol. Cells 67(1-4), 247–253 (2001). [CrossRef]
19. C. H. Chen, W. C. Shih, C. Y. Chien, C.-H. Hsu, Y.-H. Wu, and C.-H. Lai, “A promising sputtering route for one-step fabrication of chalcopyrite phase Cu(In,Ga)Se2 absorbers without extra Se supply,” Sol. Energy Mater. Sol. Cells 103, 25–29 (2012). [CrossRef]
20. J. A. Frantz, R. Y. Bekele, V. Q. Nguyen, J. S. Sanghera, A. Bruce, S. V. Frolov, M. Cyrus, and I. D. Aggarwal, “Cu(In,Ga)Se2 thin films and devices sputtered from a single target without additional selenization,” Thin Solid Films 519(22), 7763–7765 (2011). [CrossRef]
21. J. H. Shi, Z. Q. Li, D. W. Zhang, Q. Q. Liu, Z. Sun, and S. M. Huang, “Fabrication of Cu(In, Ga)Se2 thin films by sputtering from a single quaternary chalcogenide target,” Prog. Photovolt. Res. Appl. 19(2), 160–164 (2011). [CrossRef]
22. A. J. Zhou, D. Mei, X. G. Kong, X. H. Xu, L. D. Feng, X. Y. Dai, T. Gao, and J. Z. Li, “One-step synthesis of Cu(In,Ga)Se2 absorber layers by magnetron sputtering from a single quaternary target,” Thin Solid Films 520(19), 6068–6074 (2012). [CrossRef]
23. S. H. Jung, S. J. Ahn, J. H. Yun, J. Gwak, D. Kim, and K. Yoon, “Effects of Ga contents on properties of CIGS thin films and solar cells fabricated by co-evaporation technique,” Curr. Appl. Phys. 10(4), 990–996 (2010). [CrossRef]
24. M. Steichen, J. Larsen, and P. J. Dale, “Preparation of CuGaSe2 absorber layers for thin film solar cells by annealing of efficiently electrodeposited Cu–Ga precursor layers from ionic liquids,” Thin Solid Films 519(21), 7254–7258 (2011). [CrossRef]
25. S. Roy, P. Guha, S. N. Kundu, H. Hanzawa, S. Chaudhuri, and A. K. Pal, “Characterization of Cu(In,Ga)Se2 films by Raman scattering,” Mater. Chem. Phys. 73(1), 24–30 (2002). [CrossRef]
26. H. Tanino, H. Deai, and H. Nakanishi, “Raman Spectra of CuGaxIn1-xSe2,” Jpn. J. Appl. Phys. 32(S3), 436–438 (1993). [CrossRef]
27. S. K. Lee, J. K. Sim, and C. R. Lee, “Se interlayer in CIGS absorption layer for solar cell devices,” J. Alloys Compd. 633, 31–36 (2015). [CrossRef]
28. S. Kang, R. Sharma, and C. R. Lee, “Band gap engineering of tandem structured CIGS compound absorption layer fabricated by sputtering and selenization,” J. Alloys Compd. 563, 207–215 (2013). [CrossRef]
29. G. H. Bauer, S. Tardon, and S. Vignoli, “Quasi-Fermi level splitting and identification of recombination losses from room temperature luminescence in Cu(In1-xGax)Se2 thin films versus optical band gap,” Thin Solid Films 480–481, 410–414 (2005). [CrossRef]
30. K. Ramanathan, G. Teeter, and R. Noufi, “Properties of high-efficiency CuInGaSe2 thin film solar cells,” Thin Solid Films 480–481, 499–502 (2005). [CrossRef]
31. F. B. Dejene and V. Alberts, “Structural and optical properties of homogeneous Cu(In,Ga)Se2 thin films prepared by thermal reaction of InSe/Cu/GaSe alloys with elemental Se vapour,” J. Phys. D Appl. Phys. 38(1), 22–25 (2005). [CrossRef]
32. C. C. Chen, X. Qi, W. C. Chang, M.-G. Tsai, I.-G. Chen, C.-Y. Lin, P.-H. Wu, and K.-P. Chang, “The effects of pulse repetition rate on the structural, optical, and electrical properties of CIGS films grown by pulsed laser deposition,” Appl. Surf. Sci. 351, 772–778 (2015). [CrossRef]
33. M. Saifullah, J. H. Moon, and J. H. Yun, “Effects of Cu content on the photovoltaic properties of wide bandgap CIGS thin-film solar cells prepared by single-stage process,” Curr. Appl. Phys. 16(11), 1517–1522 (2016). [CrossRef]
34. H. X. Zhang and R. J. Hong, “CIGS absorbing layers prepared by RF magnetron sputtering from a single quaternary target,” Ceram. Int. 42(13), 14543–14547 (2016). [CrossRef]