- Email: [email protected]
Characterization of ZnO:Ga transparent contact electrodes for microcrystalline silicon thin film solar cells
ARTICLE IN PRESS Solar Energy Materials & Solar Cells 94 (2010) 397–401
Contents lists available at ScienceDirect
Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat
Review
Characterization of ZnO:Ga transparent contact electrodes for microcrystalline silicon thin film solar cells Kuang-Chieh Lai a, Chien-Chih Liu b, Chun-hsiung Lu c, Chih-Hung Yeh c, Mau-Phon Houng a,n a
Institute of Microelectronics, Department of Electrical Engineering, National Cheng Kung University, No.1, Dasyue Rd., East District, Tainan City 701, Taiwan Department of Electrical Engineering, Nan Jeon Institute of Technology, Tainan County 737, Taiwan c NexPower Technology Corporation, Taichung County 421, Taiwan b
a r t i c l e in fo
abstract
Article history: Received 14 August 2009 Received in revised form 4 December 2009 Accepted 4 December 2009 Available online 29 December 2009
Gallium-doped zinc oxide (ZnO:Ga) thin films are of interest to the semiconductor industry as transparent conductive surfaces and as transparent contact electrode layers for applications such as microcrystalline silicon (mc-Si) thin film solar cells. Physical vapor deposition (PVD) via sputtering is commonly used to produce thin films such as ZnO:Ga, but film quality and characteristics depend significantly on the PVD processing parameters. For use as contact electrode layers in mc-Si thin film solar cells, this study investigates some of the important changes of ZnO:Ga thin films that result from varying DC magnetron PVD sputtering parameters, specifically the working power (500, 1200, and 1900 w), process gas (Ar, Ar/O2 =50/0.2 sccm) and working pressure (0.74 and 1.06 Pa). Process temperature is held at 200 1C because thin film solar cells are damaged above 200 1C. Adding O2 to the Ar carrier gas improved transmittance but the resistivity suffered. However, high-sputtering power solved the resistivity problem. Additionally, the effects of the produced ZnO:Ga material when applied as multi-layer front and back layer electrodes to mc-Si thin film solar cells is evaluated in terms of opencircuit voltage (DVOC), short-circuit current density (DJSC), fill factor (DFF) and efficiency (DZ) of the cells. & 2009 Elsevier B.V. All rights reserved.
Keywords: Gallium-doped zinc oxide Transparent conductive oxide Solar cells Microcrystalline silicon
Contents 1. 2. 3. 4.
Introduction . . . . . . . . . Experimental . . . . . . . . Results and discussion. Conclusions . . . . . . . . . Acknowledgement . . . . References . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
1. Introduction Transparent conductors have become a basic material of modern society. They are found primarily in optoelectronic applications such as light emitting diodes (LED), solid-state lighting, flat panel liquid crystal displays (LCD) and solar cells. With mounting consumer pressure for larger video displays, energy-efficient lighting and alternative energy sources, the worldwide demand for transparent conductive films is increasing rapidly.
n
Corresponding author. Tel.: + 886 6 275 7575 62342; fax: + 886 6 234 5482. E-mail address: [email protected] (M.-P. Houng).
0927-0248/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.solmat.2009.12.002
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
397 398 398 400 400 401
Transparent conductive materials are simultaneously optically transparent and electrically conductive. In 1907 Baedeker discovered that thin CdO films possess such characteristics [1]. Further work revealed that concurrent transparency and conductivity are not uncommon in metal oxides. In consequence, the majority of contemporary transparent conductors are based on metal oxides. The electrical conductivity and optical transparency of these materials depend on the nature, number and atomic arrangements of metal cations in the oxide structure and on the presence of intrinsic or intentionally introduced defects. Conductivity is due to doping either by oxygen vacancies or by extrinsic dopants. In the absence of doping, these oxides become good insulators. Al-doped zinc oxide (ZnO:Al, AZO), tin-doped indium oxide (In2O3:Sn, ITO) and antimony- or fluorine-doped tin
ARTICLE IN PRESS 398
K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
oxide (SnO2:Sb, SnO2:F, ATO and FTO) are at present the most utilized transparent conductive oxides (TCO) TCO thin films in modern technology [2]. In fact, the industry standard TCO is ITO, with low resistivity ( 10 4 O cm) and good transmittance (480%). The primary disadvantage of ITO is the high price of indium, which is a little less rare than the metal silver. Largely as a consequence of increased worldwide LCD production, demand for indium has risen rapidly, with an approximately 10-fold increase in price over the last decade (US$94/kg in 2002, US$700–$1000/kg in 2005– 2007) [3]. Increasingly high indium prices are helping to drive the search for alternative TCO materials. In addition to economic issues, ITO suffers from limited optical tunability and poor mechanical flexibility, e.g. it cracks when used in the current generation of touch screen displays and is likely to do so in nextgeneration rollable displays. In consequence of price and performance issues, a number of promising alternative materials are being investigated. Transparent conductors made from carbon-based graphene nanotubes are attracting significant attention. For example, Hu et al. recently reported that transparent nanotube electrodes and transparent graphene electrodes outperform contemporary alternatives in several key categories [4]. Transparent conductive polymers (TCP) are less conductive than TCO and absorb significant visible and IR spectrum, but TCP can be made into flexible films well suited to flexible electronic devices. Nevertheless, the present need of the industry is an inexpensive substitute for indium in the already successful metallic oxides. Moreover, a successful alternative TCO film material must be compatible with high-throughput largearea processing. Further, the material must be well-characterized for various processing methods because small processing changes can result in large changes in product quality. It is also desired that processing temperatures be minimized, since higher temperatures cause degradation of the silicon substrates used in thin film solar cells. ZnO, commonly found in baby powder and sun tan oil, is an inexpensive, environmentally friendly material that is wellcharacterized for a wide variety of industrial applications. Compared with SnO2- or In2O3-based films, ZnO-based TCO films have superior electro-optical properties and are more durable when subjected to hydrogen plasma treatment [5–7]. Al and Ga dopants enhance free-carrier generation in ZnO films [8–11]. Thus, AZO is one of the alternative TCO materials under current study. Similar to AZO, gallium-doped zinc oxide (ZnO:Ga) films have been receiving attention due to low material costs in conjunction with relatively low deposition temperatures during fabrication [12,13]. Hydrogenated microcrystalline silicon (mc-Si) thin film is considered as one of the most promising materials for siliconbased thin film solar cells because of superior performance stability and relatively higher conversion efficiency relative to amorphous silicon solar cells. Therefore, the question of how best and most inexpensively to fabricate high efficiency mc-Si solar cells has becomes an important issue [14,15]. A key aspect of this issue is optimum design and fabrication of TCO electrodes for high efficiency single mc-Si solar cells. In consequence, this article considers ZnO:Ga film as transparent contact electrode material on mc-Si thin film solar cells. In particular, this study characterizes some of the important changes to a ZnO:Ga thin film that result from varying DC magnetron PVD sputtering parameters, specifically the working power (500, 1200, and 1900 w), process gas (Ar, Ar/O2 = 50/0.2 sccm) and working pressure (0.74 and 1.06 Pa). Process temperature is held at 200 1C because thin film solar cells are damaged above 200 1C. The paper is organized as follows: after an initial introduction, we present the experimental setup and discuss the conditions
that are varied during the DC magnetron sputtering fabrication process. Experimental results are then presented. The effects of the varied parameters on the electrical, optical and structural properties of ZnO:Ga films are analyzed and discussed. Additionally, this study also performs experimental evaluation of SnO2:F/ ZnO:Ga bi-layer front contact electrodes and ZnO:Ga/Ag back reflector contact electrodes with regard to the performance of the fabricated mc-Si thin film solar cells. Finally, the concluding section summarizes the present work and offers a brief view of our planned future study.
2. Experimental ZnO:Ga films were deposited on glass substrates by commercial DC magnetron sputtering system using a ceramic oxide target (0.6 wt% Ga2O3). The glass substrates were ultrasonically cleaned using detergent and deionized water, then dried by blown nitrogen. A series of ZnO:Ga films with a thickness of 100710 nm were prepared at 200 1C. Film thickness was measured by a spectroscopic ellipsometer (GES5E CCD Type, Sopra, France). During film deposition, two system gas pressures were used, 0.74 and 1.06 Pa. To investigate the effects of oxygen species on the film properties, the ambient gas of the PVD system was varied between two alternate states, one condition being pure Ar, the other being a mixture of Ar+ O2 (Ar/O2 = 50/0.2 sccm). Different deposition rates of ZnO:Ga films were created by controlling the deposition power at 500, 1200, and 1900 W. mc-Si thin film solar cells were fabricated by plasma enhanced chemical vapor deposition (PECVD) using a conventional 27.12 MHz plasma CVD system. The pin layers of the mc-Si thin film solar cells were fabricated by PECVD, but the TCO layer was fabricated by sputtering PVD. SiH4, H2, B2H6, CH4, and PH3 were used as gas sources. Commercially prepared textured glass/SnO2:F was used as the substrate for a ZnO:Ga layer that was deposited as described above. Solar cells were then grown on the SnO2:F / ZnO:Ga coated glass substrate with a p–i–n structure as: glass/ TCO/p–mc-Si:H/i–mc-Si:H/n–mc-Si:H/TCO/Ag. Under illumination, the I–V properties and the electrical properties of the produced solar cells were measured by an AM 1.5 G double beam Solar simulator (YSS-50A, Yamashita Denso) and characterized by fourpoint probe and Hall effect measurement in the van der Pauw configuration (HL 550PC, Bio-rad) at room temperature. X-ray diffraction (XRD) patterns were obtained by 40 kV-20 mA CuKa radiation (UltimaIV, Rigaku). Optical transmission of the films was measured using a UV–visible spectrophotometer (UV-4100, Hitachi).
3. Results and discussion Resistivity and deposition rate as a function of deposition power for ZnO:Ga thin films deposited at 200 1C can be seen for Ar-only ambient in Fig. 1 (a) and Ar+O2 ambient in Fig. 1 (b). Clearly, the deposition rate increases monotonically and ˚ identically from 4 to 18 A/s as the deposition power increases from 500 to 1900 W in both figures, which is to say that the tested gas pressures and gas mixtures have no significant effect on the deposition rate. Specifically, the deposition power directly refers to the number of atoms sputtered from the target. For higher sputtering power, the sputtered species get a higher energy that contributes to the film growth. These higher energy particles have high-surface mobility and therefore, a higher growing process at the surface takes place [16–18]. With regard to resistivity, the films fabricated in Ar ambient at 1.06 Pa show little resistivity variance with increasing sputtering power, remaining almost
ARTICLE IN PRESS K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
399
Fig. 1. Resistivity and deposition rate as a function of the deposition power for ZnO:Ga thin films deposited under: (a) Ar and (b) Ar + O2 gas ambient at various gas pressures. Fig. 2. X-ray diffraction patterns of ZnO:Ga films for: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar + O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
constant at 3.5 10 3 O cm. Under 0.74 Pa Ar ambient, however, there seems to be a significant but small increase in resistivity, changing from 3 10 3 to 4.5 10 3 O cm with increasing sputtering power. The reason for increasing resistance is unclear, but we speculate morphological differences related to the number and type of defects. Fig. 1 (a) shows that the lowest resistances occur at slowest deposition rates, i.e. lowest sputtering powers. In general, slow deposition and crystallization processes favor better crystal quality, i.e. fewer defects. Resistance is even lower at the lower 0.74 gas pressure, under which conditions there would be less crowding and ‘‘premature’’ bonding with the growing ZnO:Ga surface. Thus it seems reasonable that, under pure Ar carrier, the observed changes in resistance are related to the morphology of the deposited film, with faster deposition resulting in reduced crystal quality. In contrast, the films fabricated under Ar+ O2 ambient at both tested pressures show dramatic and very similar decrease of resistivity with increasing sputtering power. At the lowest sputtering power of 500 W, the resistivity curves for the films produced under both gas pressures show an average resistivity of approximately 2 10 2 O cm. At 1200 W, however, the resistivity for both curves has decreased dramatically to around 0.45 10 2 O cm, i.e. about one-fourth of the resistivity obtained at the lowest deposition power. We attribute the higher film resistance at the lowest deposition power to vacancy filling by the oxygen species contained in the Ar+ O2 gas ambient, thus reducing the number of oxygen vacancies in the samples. Conversely, increasing the deposition rate by increasing the deposition power naturally and inevitably reduces the probability of vacancy filling by the oxygen species, which in turn causes the samples to have resistivity nearly equal to that of the samples fabricated in Ar-only ambient (compare Fig. 1 (a) and (b) at 1200 W). Fig. 2 illustrates the XRD patterns for the ZnO:Ga films deposited under various atmospheres. In the case of the Ar+ O2 gas mixture, the intensity of the (0 0 2) peak is clearly enhanced. Obviously, the crystallization of a ZnO:Ga thin film produced by DC magnetron sputtering is significantly improved under Ar+ O2 gas ambiance. Table 1 shows the resistivity, mobility and carrier concentrations of the ZnO:Ga films fabricated under different gas atmospheres. When the oxygen species is included in the gas mixture, the film exhibits a resistivity of 5.2 10 3 O cm, a carrier concentration of 1.17 1020 cm 3 and a mobility of 10.3 cm2/V s. The observed improvement over the withoutoxygen deposition conditions seems due to improved ZnO:Ga
Table 1 Resistivity, Hall mobility and carrier concentrations of ZnO:Ga films fabricated under Ar and Ar+ O2. Gas ambient
Resistivity (O cm)
Ar Ar +O2
3.6 10 5.2 10
3 3
Hall mobility (cm2/V s)
Carrier concentration. (cm 3)
8.42 10.3
2.06 1020 1.17 1020
Fig. 3. Transparency versus wavelength for: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar+ O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
crystallization, as indicated by the XRD results and supported by the SEM data presented below. Comparison of transmittance of the ZnO:Ga films prepared under different gas sources as a function of wavelength is shown in Fig. 3. Curve (a) represents Ar-only at low-deposition power, while curve (b) represents Ar +O2 at high-deposition power. Although both samples exhibit the same transmittance versus wavelength trend, the ZnO:Ga film obtained by DC magnetron sputtering under Ar+ O2 gas ambient has significantly higher
ARTICLE IN PRESS 400
K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
Fig. 4. SEM micrographs of ZnO:Ga thin films prepared under: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar + O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
Table 2 I–V properties of p–i–n mc-Si solar cells produced under: (A) Ar and (B) Ar + O2. Cell ID
Front electrode
Back electrode
DVOC
DJSC
DFF
DZ
Cell Cell Cell Cell
SnO2:F SnO2:F/ZnO:Ga_(A) SnO2:F/ZnO:Ga_(A) SnO2:F/ZnO:Ga_(B)
ZnO:Ga_(A)/Ag ZnO:Ga_(A)/Ag ZnO:Ga_(B)/Ag ZnO:Ga_(B)/Ag
1 1 1.02 1.02
1 1.17 1.19 1.23
1 1.08 1.08 1.08
1 1.26 1.31 1.36
A B C D
transmittance. Notably, Ar-only is less transparent than Ar+O2 under all fabrication conditions. Moreover, changing the fabrication sputtering power and/or gas pressure for both gas mixtures shows no obvious change in transparency. The observed changes in both resistance and transparency are reflected in the SEM data shown in Fig. 4 (a) (Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness) and 4(b) (Ar+ O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness). The smaller grains seen in Fig. 4 (a) commonly indicate worse crystal quality. The larger grains and more consistent grain structure seen in Fig. 4 (b) typically indicate better crystal quality and improved transparency. To study the influence of ZnO:Ga film as transparent front and back contact electrodes on mc-Si thin film solar cells, we fabricated single junction solar cells with the structure Glass/ SnO2:F/ZnO/p–mc-Si:H/i–mc-Si:H/n–mc-Si:H/ZnO/Ag with various TCO electrodes. The open-circuit voltage (DVOC), short-circuit current density (DJSC), fill factor (DFF), and efficiency (DZ) of these cells are depicted in Table 2. It may be seen that both JSC and fill factor are lowest in the case of the SnO2:F-coated substrate. For a SnO2:F/ZnO:Ga substrate, in all cases the JSC and fill factor are higher than for SnO2 alone. The cell prepared on the SnO2: F-coated glass substrate (Cell A) shows low values for efficiency, short-circuit current density (JSC), open-circuit voltage (VOC) and fill factor. This is due to degradation of the SnO2:F film under the hydrogen plasma, which is an essential processing condition for fabrication of the p–mc-Si:H layer. Comparison of cells B (Ar only) and D (Ar+O2) shows superior efficiency in cell D, which is attributed mainly to the higher optical transmission of the ZnO:Ga contact electrodes, transmission having been enhanced by preparation under Ar+ O2 ambient. Cells B (Ar only) and D (Ar+O2) have the identical cell configuration, but cell D shows a short-circuit current density of 1.23, significantly better than the 1.17 of cell B. Further, efficiency of cell D is 1.36, significantly higher than the 1.26 of cell B. These results clearly show that use of Ar+O2 ambient during DC magnetron sputtering preparation of ZnO:Ga thin films can improve the performance of mc-Si thin film solar cells. The improved performance via Ar+O2 ambient can be attributed to two effects, namely enhanced transparency of the ZnO:Ga TCO electrodes (both front and back) and also reduced resistivity of the
ZnO:Ga TCO material. However, the use of Ar+O2 at our lowest deposition rate results in high film resistance. It is only under our two higher sputtering powers (and consequentially higher deposition rates) that the film resistance drops to acceptable levels.
4. Conclusions This paper has investigated the relative transparency and conductivity of ZnO:Ga thin films deposited by DC magnetron sputtering under various working powers (500, 1200, and 1900 W), process gas conditions (Ar, Ar/O2 =50/0.2 sccm) and working pressures (0.74 Pa, 1.06 Pa). Process temperature in all cases was held at 200 1C. Working pressure was found to have no effect on deposition rate. For Ar-only ambient at low-deposition power, higher working pressure resulted in slightly higher film resistance. At middle deposition power, however, this effect disappeared and, at high-deposition power, higher working pressure reversed this effect. For all deposition conditions, addition of O2 to the Ar carrier gas improved the optical transmittance of the films because of better crystalline structure. Higher transmittance occurred in the (0 0 2) orientation along the c-axis. Unfortunately, use of Ar+O2 at low-deposition power resulted in sharply increased resistivity of the deposited film. The high resistivity disappeared, however, when deposition was performed at middle or high-sputtering power. The sputtering conditions which produced the lowest resistivity and highest transmittance were used under both Ar-only (500 W, 0.74 Pa) and Ar+O2 (1900 W, 0.74 Pa) ambient to produce SnO2:F/ ZnO:Ga bi-layer front contact electrodes and ZnO:Ga/Ag back reflector contact electrodes for mc-Si thin film solar cells. The performance of the resulting solar cells were compared for ZnO:Ga TCO fabrication under Ar-only versus Ar+O2. Efficiency was lower for SnO2-only mc-Si thin film solar cells since the SnO2 TCO degrades under hydrogen plasma. The JSC and fill factor were higher when the mc-Si thin film solar cells were fabricated using SnO2/ZnO:Ga because the ZnO:Ga layer protected the SnO2 from plasma damage. Of the SnO2/ZnO:Ga contact electrodes, the cell fabricated under Ar+O2 showed better JSC compared to fabrication under Ar-only. This latter effect is attributed to the greater transparency of the ZnO:Ga produced under Ar+O2, which allows more light into the front of the cell and reduced light absorption in the back contact layer. It is expected that the results of this study will contribute meaningfully to the development of the next generation of transparent conducting materials.
Acknowledgement This work was supported by the National Science Council of Taiwan under Contract nos. NSC-97-2221-E-006-239-MY2.
ARTICLE IN PRESS K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
References [1] K. Baedeker, Uber die elektrische Leitfahigkeit und die thermoelektrische Kraft einiger Schwermetallverbindungen, Ann. Phys. 22 (1907) 749–766. [2] G. Haacke, Transparent conducting oxides, Ann. Rev. Mater. Sci. 7 (1977) 73–93. [3] S.J. Hsieh, C.C. Chen, W.C. Say, Process for recovery of indium from ITO scraps and metallurgic microstructures, Mater. Sci. Eng. B 158 (2009) 82–87. [4] L. Hu, D.S. Hecht, G. Gruner, Infrared transparent carbon nanotube thin films, Appl. Phys. Lett. 94 (2009) 081103-1–081103-3. [5] M. Kubon, E. Boehmer, F. Siebke, B. Rech, C. Beneking, H. Wagner, Solution of the ZnO/p contact problem in a-Si:H solar cells, Sol. Energy Mater. Sol. Cells 41/42 (1996) 485–492. [6] R. Das, T. Jana, S. Ray, Degradation studies of transparent conducting oxide: a substrate for microcrystalline silicon thin film solar cells, Sol. Energy Mater. Sol. Cells 86 (2005) 207–216. [7] J.C. Lee, V. Dutta, J. Yoo, J. Yi, J. Song, K.H. Yoon, Superstrate p–i–n a-Si:H solar cells on textured ZnO:Al front transparent conduction oxide, Superlattices Microstruct. 42 (2007) 369–374. [8] P.K. Song, M. Watanabe, M. Kon, A. Mitsui, Y. Shigesato, Electrical and optical properties of gallium-doped zinc oxide films deposited by dc magnetron sputtering, Thin Solid Films 411 (2002) 82–86. [9] V. Assuncao, E. Fortunato, A. Marques, H. Aguas, I. Ferreira, M.E.V. Costa, R. Martins, Influence of the deposition pressure on the properties of transparent and conductive ZnO:Ga thin-film produced by r.f. sputtering at room temperature, Thin Solid Films 427 (2003) 401–405. [10] A. Tiburcio-Silver, A. Sanchez-Juarez, A. Avila-Garcia, Properties of galliumdoped ZnO deposited onto glass by spray pyrolysis, Sol. Energy Mater. Sol. Cells 55 (1998) 3–10.
401
[11] V. Khranovskyy, U. Grossner, V. Lazorenko, G. Lashkarev, B.G. Svensson, R. Yakimova, PEMOCVD of ZnO thin films, doped by Ga and some of their properties, Superlattices Microstruct. 39 (2006) 275–281. [12] E. Fortunato, V. Assuncao, A. Goncalves, A. Marques, H. Aguas, L. Pereira, I. Ferreira, P. Vilarinho, R. Martins, High quality conductive gallium-doped zinc oxide films deposited at room temperature, Thin Solid Films 451–452 (2004) 443–447. [13] T. Yamada, A. Miyake, S. Kishimoto, H. Makino, N. Yamamoto, T. Yamamoto, Effects of substrate temperature on crystallinity and electrical properties of Ga-doped ZnO films prepared on glass substrate by ion-plating method using DC arc discharge, Surf. Coat. Technol. 202 (2007) 973–976. [14] X.D. Zhang, Y. Zhao, Y.T. Gao, F. Zhu, C.C. Wei, X.L. Chen, J. Sun, G.F. Hou, X.H. Geng, S.Z. Xiong, Influence of front electrode and back reflector electrode on the performances of microcrystalline silicon solar cells, J. Non-Cryst. Solids 352 (2006) 1863–1867. [15] E. Fortunato, D. Ginley, H. Hosono, D.C. Paine, Transparent conducting oxides for photovoltaics, MRS Bull. 32 (2007) 242–247. [16] Q.B. Ma, Z.Z. Ye, H.P. He, J.R. Wang, L.P. Zhu, B.H. Zhao, Substrate temperature dependence of the properties of Ga-doped ZnO films deposited by DC reactive magnetron sputtering, Vacuum 82 (2008) 9–14. [17] H. Gsomez, M. de la L. Olvera, Ga-doped ZnO thin films: effect of deposition temperature, dopant concentration, and vacuum-thermal treatment on the electrical, optical, structural and morphological properties, Mater. Sci. Eng. B 134 (2006) 20–26. [18] X. Yu, J. Ma, F. Ji, Y. Wang, X. Zhang, C. Cheng, H. Ma, Effects of sputtering power on the properties of ZnO:Ga films deposited by r.f. magnetron-sputtering at low temperature, J. Cryst. Growth 274 (2005) 474–479.
Contents lists available at ScienceDirect
Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat
Review
Characterization of ZnO:Ga transparent contact electrodes for microcrystalline silicon thin film solar cells Kuang-Chieh Lai a, Chien-Chih Liu b, Chun-hsiung Lu c, Chih-Hung Yeh c, Mau-Phon Houng a,n a
Institute of Microelectronics, Department of Electrical Engineering, National Cheng Kung University, No.1, Dasyue Rd., East District, Tainan City 701, Taiwan Department of Electrical Engineering, Nan Jeon Institute of Technology, Tainan County 737, Taiwan c NexPower Technology Corporation, Taichung County 421, Taiwan b
a r t i c l e in fo
abstract
Article history: Received 14 August 2009 Received in revised form 4 December 2009 Accepted 4 December 2009 Available online 29 December 2009
Gallium-doped zinc oxide (ZnO:Ga) thin films are of interest to the semiconductor industry as transparent conductive surfaces and as transparent contact electrode layers for applications such as microcrystalline silicon (mc-Si) thin film solar cells. Physical vapor deposition (PVD) via sputtering is commonly used to produce thin films such as ZnO:Ga, but film quality and characteristics depend significantly on the PVD processing parameters. For use as contact electrode layers in mc-Si thin film solar cells, this study investigates some of the important changes of ZnO:Ga thin films that result from varying DC magnetron PVD sputtering parameters, specifically the working power (500, 1200, and 1900 w), process gas (Ar, Ar/O2 =50/0.2 sccm) and working pressure (0.74 and 1.06 Pa). Process temperature is held at 200 1C because thin film solar cells are damaged above 200 1C. Adding O2 to the Ar carrier gas improved transmittance but the resistivity suffered. However, high-sputtering power solved the resistivity problem. Additionally, the effects of the produced ZnO:Ga material when applied as multi-layer front and back layer electrodes to mc-Si thin film solar cells is evaluated in terms of opencircuit voltage (DVOC), short-circuit current density (DJSC), fill factor (DFF) and efficiency (DZ) of the cells. & 2009 Elsevier B.V. All rights reserved.
Keywords: Gallium-doped zinc oxide Transparent conductive oxide Solar cells Microcrystalline silicon
Contents 1. 2. 3. 4.
Introduction . . . . . . . . . Experimental . . . . . . . . Results and discussion. Conclusions . . . . . . . . . Acknowledgement . . . . References . . . . . . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
1. Introduction Transparent conductors have become a basic material of modern society. They are found primarily in optoelectronic applications such as light emitting diodes (LED), solid-state lighting, flat panel liquid crystal displays (LCD) and solar cells. With mounting consumer pressure for larger video displays, energy-efficient lighting and alternative energy sources, the worldwide demand for transparent conductive films is increasing rapidly.
n
Corresponding author. Tel.: + 886 6 275 7575 62342; fax: + 886 6 234 5482. E-mail address: [email protected] (M.-P. Houng).
0927-0248/$ - see front matter & 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.solmat.2009.12.002
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
. . . . . .
397 398 398 400 400 401
Transparent conductive materials are simultaneously optically transparent and electrically conductive. In 1907 Baedeker discovered that thin CdO films possess such characteristics [1]. Further work revealed that concurrent transparency and conductivity are not uncommon in metal oxides. In consequence, the majority of contemporary transparent conductors are based on metal oxides. The electrical conductivity and optical transparency of these materials depend on the nature, number and atomic arrangements of metal cations in the oxide structure and on the presence of intrinsic or intentionally introduced defects. Conductivity is due to doping either by oxygen vacancies or by extrinsic dopants. In the absence of doping, these oxides become good insulators. Al-doped zinc oxide (ZnO:Al, AZO), tin-doped indium oxide (In2O3:Sn, ITO) and antimony- or fluorine-doped tin
ARTICLE IN PRESS 398
K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
oxide (SnO2:Sb, SnO2:F, ATO and FTO) are at present the most utilized transparent conductive oxides (TCO) TCO thin films in modern technology [2]. In fact, the industry standard TCO is ITO, with low resistivity ( 10 4 O cm) and good transmittance (480%). The primary disadvantage of ITO is the high price of indium, which is a little less rare than the metal silver. Largely as a consequence of increased worldwide LCD production, demand for indium has risen rapidly, with an approximately 10-fold increase in price over the last decade (US$94/kg in 2002, US$700–$1000/kg in 2005– 2007) [3]. Increasingly high indium prices are helping to drive the search for alternative TCO materials. In addition to economic issues, ITO suffers from limited optical tunability and poor mechanical flexibility, e.g. it cracks when used in the current generation of touch screen displays and is likely to do so in nextgeneration rollable displays. In consequence of price and performance issues, a number of promising alternative materials are being investigated. Transparent conductors made from carbon-based graphene nanotubes are attracting significant attention. For example, Hu et al. recently reported that transparent nanotube electrodes and transparent graphene electrodes outperform contemporary alternatives in several key categories [4]. Transparent conductive polymers (TCP) are less conductive than TCO and absorb significant visible and IR spectrum, but TCP can be made into flexible films well suited to flexible electronic devices. Nevertheless, the present need of the industry is an inexpensive substitute for indium in the already successful metallic oxides. Moreover, a successful alternative TCO film material must be compatible with high-throughput largearea processing. Further, the material must be well-characterized for various processing methods because small processing changes can result in large changes in product quality. It is also desired that processing temperatures be minimized, since higher temperatures cause degradation of the silicon substrates used in thin film solar cells. ZnO, commonly found in baby powder and sun tan oil, is an inexpensive, environmentally friendly material that is wellcharacterized for a wide variety of industrial applications. Compared with SnO2- or In2O3-based films, ZnO-based TCO films have superior electro-optical properties and are more durable when subjected to hydrogen plasma treatment [5–7]. Al and Ga dopants enhance free-carrier generation in ZnO films [8–11]. Thus, AZO is one of the alternative TCO materials under current study. Similar to AZO, gallium-doped zinc oxide (ZnO:Ga) films have been receiving attention due to low material costs in conjunction with relatively low deposition temperatures during fabrication [12,13]. Hydrogenated microcrystalline silicon (mc-Si) thin film is considered as one of the most promising materials for siliconbased thin film solar cells because of superior performance stability and relatively higher conversion efficiency relative to amorphous silicon solar cells. Therefore, the question of how best and most inexpensively to fabricate high efficiency mc-Si solar cells has becomes an important issue [14,15]. A key aspect of this issue is optimum design and fabrication of TCO electrodes for high efficiency single mc-Si solar cells. In consequence, this article considers ZnO:Ga film as transparent contact electrode material on mc-Si thin film solar cells. In particular, this study characterizes some of the important changes to a ZnO:Ga thin film that result from varying DC magnetron PVD sputtering parameters, specifically the working power (500, 1200, and 1900 w), process gas (Ar, Ar/O2 = 50/0.2 sccm) and working pressure (0.74 and 1.06 Pa). Process temperature is held at 200 1C because thin film solar cells are damaged above 200 1C. The paper is organized as follows: after an initial introduction, we present the experimental setup and discuss the conditions
that are varied during the DC magnetron sputtering fabrication process. Experimental results are then presented. The effects of the varied parameters on the electrical, optical and structural properties of ZnO:Ga films are analyzed and discussed. Additionally, this study also performs experimental evaluation of SnO2:F/ ZnO:Ga bi-layer front contact electrodes and ZnO:Ga/Ag back reflector contact electrodes with regard to the performance of the fabricated mc-Si thin film solar cells. Finally, the concluding section summarizes the present work and offers a brief view of our planned future study.
2. Experimental ZnO:Ga films were deposited on glass substrates by commercial DC magnetron sputtering system using a ceramic oxide target (0.6 wt% Ga2O3). The glass substrates were ultrasonically cleaned using detergent and deionized water, then dried by blown nitrogen. A series of ZnO:Ga films with a thickness of 100710 nm were prepared at 200 1C. Film thickness was measured by a spectroscopic ellipsometer (GES5E CCD Type, Sopra, France). During film deposition, two system gas pressures were used, 0.74 and 1.06 Pa. To investigate the effects of oxygen species on the film properties, the ambient gas of the PVD system was varied between two alternate states, one condition being pure Ar, the other being a mixture of Ar+ O2 (Ar/O2 = 50/0.2 sccm). Different deposition rates of ZnO:Ga films were created by controlling the deposition power at 500, 1200, and 1900 W. mc-Si thin film solar cells were fabricated by plasma enhanced chemical vapor deposition (PECVD) using a conventional 27.12 MHz plasma CVD system. The pin layers of the mc-Si thin film solar cells were fabricated by PECVD, but the TCO layer was fabricated by sputtering PVD. SiH4, H2, B2H6, CH4, and PH3 were used as gas sources. Commercially prepared textured glass/SnO2:F was used as the substrate for a ZnO:Ga layer that was deposited as described above. Solar cells were then grown on the SnO2:F / ZnO:Ga coated glass substrate with a p–i–n structure as: glass/ TCO/p–mc-Si:H/i–mc-Si:H/n–mc-Si:H/TCO/Ag. Under illumination, the I–V properties and the electrical properties of the produced solar cells were measured by an AM 1.5 G double beam Solar simulator (YSS-50A, Yamashita Denso) and characterized by fourpoint probe and Hall effect measurement in the van der Pauw configuration (HL 550PC, Bio-rad) at room temperature. X-ray diffraction (XRD) patterns were obtained by 40 kV-20 mA CuKa radiation (UltimaIV, Rigaku). Optical transmission of the films was measured using a UV–visible spectrophotometer (UV-4100, Hitachi).
3. Results and discussion Resistivity and deposition rate as a function of deposition power for ZnO:Ga thin films deposited at 200 1C can be seen for Ar-only ambient in Fig. 1 (a) and Ar+O2 ambient in Fig. 1 (b). Clearly, the deposition rate increases monotonically and ˚ identically from 4 to 18 A/s as the deposition power increases from 500 to 1900 W in both figures, which is to say that the tested gas pressures and gas mixtures have no significant effect on the deposition rate. Specifically, the deposition power directly refers to the number of atoms sputtered from the target. For higher sputtering power, the sputtered species get a higher energy that contributes to the film growth. These higher energy particles have high-surface mobility and therefore, a higher growing process at the surface takes place [16–18]. With regard to resistivity, the films fabricated in Ar ambient at 1.06 Pa show little resistivity variance with increasing sputtering power, remaining almost
ARTICLE IN PRESS K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
399
Fig. 1. Resistivity and deposition rate as a function of the deposition power for ZnO:Ga thin films deposited under: (a) Ar and (b) Ar + O2 gas ambient at various gas pressures. Fig. 2. X-ray diffraction patterns of ZnO:Ga films for: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar + O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
constant at 3.5 10 3 O cm. Under 0.74 Pa Ar ambient, however, there seems to be a significant but small increase in resistivity, changing from 3 10 3 to 4.5 10 3 O cm with increasing sputtering power. The reason for increasing resistance is unclear, but we speculate morphological differences related to the number and type of defects. Fig. 1 (a) shows that the lowest resistances occur at slowest deposition rates, i.e. lowest sputtering powers. In general, slow deposition and crystallization processes favor better crystal quality, i.e. fewer defects. Resistance is even lower at the lower 0.74 gas pressure, under which conditions there would be less crowding and ‘‘premature’’ bonding with the growing ZnO:Ga surface. Thus it seems reasonable that, under pure Ar carrier, the observed changes in resistance are related to the morphology of the deposited film, with faster deposition resulting in reduced crystal quality. In contrast, the films fabricated under Ar+ O2 ambient at both tested pressures show dramatic and very similar decrease of resistivity with increasing sputtering power. At the lowest sputtering power of 500 W, the resistivity curves for the films produced under both gas pressures show an average resistivity of approximately 2 10 2 O cm. At 1200 W, however, the resistivity for both curves has decreased dramatically to around 0.45 10 2 O cm, i.e. about one-fourth of the resistivity obtained at the lowest deposition power. We attribute the higher film resistance at the lowest deposition power to vacancy filling by the oxygen species contained in the Ar+ O2 gas ambient, thus reducing the number of oxygen vacancies in the samples. Conversely, increasing the deposition rate by increasing the deposition power naturally and inevitably reduces the probability of vacancy filling by the oxygen species, which in turn causes the samples to have resistivity nearly equal to that of the samples fabricated in Ar-only ambient (compare Fig. 1 (a) and (b) at 1200 W). Fig. 2 illustrates the XRD patterns for the ZnO:Ga films deposited under various atmospheres. In the case of the Ar+ O2 gas mixture, the intensity of the (0 0 2) peak is clearly enhanced. Obviously, the crystallization of a ZnO:Ga thin film produced by DC magnetron sputtering is significantly improved under Ar+ O2 gas ambiance. Table 1 shows the resistivity, mobility and carrier concentrations of the ZnO:Ga films fabricated under different gas atmospheres. When the oxygen species is included in the gas mixture, the film exhibits a resistivity of 5.2 10 3 O cm, a carrier concentration of 1.17 1020 cm 3 and a mobility of 10.3 cm2/V s. The observed improvement over the withoutoxygen deposition conditions seems due to improved ZnO:Ga
Table 1 Resistivity, Hall mobility and carrier concentrations of ZnO:Ga films fabricated under Ar and Ar+ O2. Gas ambient
Resistivity (O cm)
Ar Ar +O2
3.6 10 5.2 10
3 3
Hall mobility (cm2/V s)
Carrier concentration. (cm 3)
8.42 10.3
2.06 1020 1.17 1020
Fig. 3. Transparency versus wavelength for: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar+ O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
crystallization, as indicated by the XRD results and supported by the SEM data presented below. Comparison of transmittance of the ZnO:Ga films prepared under different gas sources as a function of wavelength is shown in Fig. 3. Curve (a) represents Ar-only at low-deposition power, while curve (b) represents Ar +O2 at high-deposition power. Although both samples exhibit the same transmittance versus wavelength trend, the ZnO:Ga film obtained by DC magnetron sputtering under Ar+ O2 gas ambient has significantly higher
ARTICLE IN PRESS 400
K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
Fig. 4. SEM micrographs of ZnO:Ga thin films prepared under: (a) Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness and (b): Ar + O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness.
Table 2 I–V properties of p–i–n mc-Si solar cells produced under: (A) Ar and (B) Ar + O2. Cell ID
Front electrode
Back electrode
DVOC
DJSC
DFF
DZ
Cell Cell Cell Cell
SnO2:F SnO2:F/ZnO:Ga_(A) SnO2:F/ZnO:Ga_(A) SnO2:F/ZnO:Ga_(B)
ZnO:Ga_(A)/Ag ZnO:Ga_(A)/Ag ZnO:Ga_(B)/Ag ZnO:Ga_(B)/Ag
1 1 1.02 1.02
1 1.17 1.19 1.23
1 1.08 1.08 1.08
1 1.26 1.31 1.36
A B C D
transmittance. Notably, Ar-only is less transparent than Ar+O2 under all fabrication conditions. Moreover, changing the fabrication sputtering power and/or gas pressure for both gas mixtures shows no obvious change in transparency. The observed changes in both resistance and transparency are reflected in the SEM data shown in Fig. 4 (a) (Ar, low power (500 W), 0.74 Pa, 200 1C, 1000 A˚ thickness) and 4(b) (Ar+ O2, high power (1900 W), 0.74 Pa, 200 1C, 1000 A˚ thickness). The smaller grains seen in Fig. 4 (a) commonly indicate worse crystal quality. The larger grains and more consistent grain structure seen in Fig. 4 (b) typically indicate better crystal quality and improved transparency. To study the influence of ZnO:Ga film as transparent front and back contact electrodes on mc-Si thin film solar cells, we fabricated single junction solar cells with the structure Glass/ SnO2:F/ZnO/p–mc-Si:H/i–mc-Si:H/n–mc-Si:H/ZnO/Ag with various TCO electrodes. The open-circuit voltage (DVOC), short-circuit current density (DJSC), fill factor (DFF), and efficiency (DZ) of these cells are depicted in Table 2. It may be seen that both JSC and fill factor are lowest in the case of the SnO2:F-coated substrate. For a SnO2:F/ZnO:Ga substrate, in all cases the JSC and fill factor are higher than for SnO2 alone. The cell prepared on the SnO2: F-coated glass substrate (Cell A) shows low values for efficiency, short-circuit current density (JSC), open-circuit voltage (VOC) and fill factor. This is due to degradation of the SnO2:F film under the hydrogen plasma, which is an essential processing condition for fabrication of the p–mc-Si:H layer. Comparison of cells B (Ar only) and D (Ar+O2) shows superior efficiency in cell D, which is attributed mainly to the higher optical transmission of the ZnO:Ga contact electrodes, transmission having been enhanced by preparation under Ar+ O2 ambient. Cells B (Ar only) and D (Ar+O2) have the identical cell configuration, but cell D shows a short-circuit current density of 1.23, significantly better than the 1.17 of cell B. Further, efficiency of cell D is 1.36, significantly higher than the 1.26 of cell B. These results clearly show that use of Ar+O2 ambient during DC magnetron sputtering preparation of ZnO:Ga thin films can improve the performance of mc-Si thin film solar cells. The improved performance via Ar+O2 ambient can be attributed to two effects, namely enhanced transparency of the ZnO:Ga TCO electrodes (both front and back) and also reduced resistivity of the
ZnO:Ga TCO material. However, the use of Ar+O2 at our lowest deposition rate results in high film resistance. It is only under our two higher sputtering powers (and consequentially higher deposition rates) that the film resistance drops to acceptable levels.
4. Conclusions This paper has investigated the relative transparency and conductivity of ZnO:Ga thin films deposited by DC magnetron sputtering under various working powers (500, 1200, and 1900 W), process gas conditions (Ar, Ar/O2 =50/0.2 sccm) and working pressures (0.74 Pa, 1.06 Pa). Process temperature in all cases was held at 200 1C. Working pressure was found to have no effect on deposition rate. For Ar-only ambient at low-deposition power, higher working pressure resulted in slightly higher film resistance. At middle deposition power, however, this effect disappeared and, at high-deposition power, higher working pressure reversed this effect. For all deposition conditions, addition of O2 to the Ar carrier gas improved the optical transmittance of the films because of better crystalline structure. Higher transmittance occurred in the (0 0 2) orientation along the c-axis. Unfortunately, use of Ar+O2 at low-deposition power resulted in sharply increased resistivity of the deposited film. The high resistivity disappeared, however, when deposition was performed at middle or high-sputtering power. The sputtering conditions which produced the lowest resistivity and highest transmittance were used under both Ar-only (500 W, 0.74 Pa) and Ar+O2 (1900 W, 0.74 Pa) ambient to produce SnO2:F/ ZnO:Ga bi-layer front contact electrodes and ZnO:Ga/Ag back reflector contact electrodes for mc-Si thin film solar cells. The performance of the resulting solar cells were compared for ZnO:Ga TCO fabrication under Ar-only versus Ar+O2. Efficiency was lower for SnO2-only mc-Si thin film solar cells since the SnO2 TCO degrades under hydrogen plasma. The JSC and fill factor were higher when the mc-Si thin film solar cells were fabricated using SnO2/ZnO:Ga because the ZnO:Ga layer protected the SnO2 from plasma damage. Of the SnO2/ZnO:Ga contact electrodes, the cell fabricated under Ar+O2 showed better JSC compared to fabrication under Ar-only. This latter effect is attributed to the greater transparency of the ZnO:Ga produced under Ar+O2, which allows more light into the front of the cell and reduced light absorption in the back contact layer. It is expected that the results of this study will contribute meaningfully to the development of the next generation of transparent conducting materials.
Acknowledgement This work was supported by the National Science Council of Taiwan under Contract nos. NSC-97-2221-E-006-239-MY2.
ARTICLE IN PRESS K.-C. Lai et al. / Solar Energy Materials & Solar Cells 94 (2010) 397–401
References [1] K. Baedeker, Uber die elektrische Leitfahigkeit und die thermoelektrische Kraft einiger Schwermetallverbindungen, Ann. Phys. 22 (1907) 749–766. [2] G. Haacke, Transparent conducting oxides, Ann. Rev. Mater. Sci. 7 (1977) 73–93. [3] S.J. Hsieh, C.C. Chen, W.C. Say, Process for recovery of indium from ITO scraps and metallurgic microstructures, Mater. Sci. Eng. B 158 (2009) 82–87. [4] L. Hu, D.S. Hecht, G. Gruner, Infrared transparent carbon nanotube thin films, Appl. Phys. Lett. 94 (2009) 081103-1–081103-3. [5] M. Kubon, E. Boehmer, F. Siebke, B. Rech, C. Beneking, H. Wagner, Solution of the ZnO/p contact problem in a-Si:H solar cells, Sol. Energy Mater. Sol. Cells 41/42 (1996) 485–492. [6] R. Das, T. Jana, S. Ray, Degradation studies of transparent conducting oxide: a substrate for microcrystalline silicon thin film solar cells, Sol. Energy Mater. Sol. Cells 86 (2005) 207–216. [7] J.C. Lee, V. Dutta, J. Yoo, J. Yi, J. Song, K.H. Yoon, Superstrate p–i–n a-Si:H solar cells on textured ZnO:Al front transparent conduction oxide, Superlattices Microstruct. 42 (2007) 369–374. [8] P.K. Song, M. Watanabe, M. Kon, A. Mitsui, Y. Shigesato, Electrical and optical properties of gallium-doped zinc oxide films deposited by dc magnetron sputtering, Thin Solid Films 411 (2002) 82–86. [9] V. Assuncao, E. Fortunato, A. Marques, H. Aguas, I. Ferreira, M.E.V. Costa, R. Martins, Influence of the deposition pressure on the properties of transparent and conductive ZnO:Ga thin-film produced by r.f. sputtering at room temperature, Thin Solid Films 427 (2003) 401–405. [10] A. Tiburcio-Silver, A. Sanchez-Juarez, A. Avila-Garcia, Properties of galliumdoped ZnO deposited onto glass by spray pyrolysis, Sol. Energy Mater. Sol. Cells 55 (1998) 3–10.
401
[11] V. Khranovskyy, U. Grossner, V. Lazorenko, G. Lashkarev, B.G. Svensson, R. Yakimova, PEMOCVD of ZnO thin films, doped by Ga and some of their properties, Superlattices Microstruct. 39 (2006) 275–281. [12] E. Fortunato, V. Assuncao, A. Goncalves, A. Marques, H. Aguas, L. Pereira, I. Ferreira, P. Vilarinho, R. Martins, High quality conductive gallium-doped zinc oxide films deposited at room temperature, Thin Solid Films 451–452 (2004) 443–447. [13] T. Yamada, A. Miyake, S. Kishimoto, H. Makino, N. Yamamoto, T. Yamamoto, Effects of substrate temperature on crystallinity and electrical properties of Ga-doped ZnO films prepared on glass substrate by ion-plating method using DC arc discharge, Surf. Coat. Technol. 202 (2007) 973–976. [14] X.D. Zhang, Y. Zhao, Y.T. Gao, F. Zhu, C.C. Wei, X.L. Chen, J. Sun, G.F. Hou, X.H. Geng, S.Z. Xiong, Influence of front electrode and back reflector electrode on the performances of microcrystalline silicon solar cells, J. Non-Cryst. Solids 352 (2006) 1863–1867. [15] E. Fortunato, D. Ginley, H. Hosono, D.C. Paine, Transparent conducting oxides for photovoltaics, MRS Bull. 32 (2007) 242–247. [16] Q.B. Ma, Z.Z. Ye, H.P. He, J.R. Wang, L.P. Zhu, B.H. Zhao, Substrate temperature dependence of the properties of Ga-doped ZnO films deposited by DC reactive magnetron sputtering, Vacuum 82 (2008) 9–14. [17] H. Gsomez, M. de la L. Olvera, Ga-doped ZnO thin films: effect of deposition temperature, dopant concentration, and vacuum-thermal treatment on the electrical, optical, structural and morphological properties, Mater. Sci. Eng. B 134 (2006) 20–26. [18] X. Yu, J. Ma, F. Ji, Y. Wang, X. Zhang, C. Cheng, H. Ma, Effects of sputtering power on the properties of ZnO:Ga films deposited by r.f. magnetron-sputtering at low temperature, J. Cryst. Growth 274 (2005) 474–479.