Next Article in Journal
The Current Status of Hydrogen Storage Alloy Development for Electrochemical Applications
Next Article in Special Issue
Efficiency Improvement of HIT Solar Cells on p-Type Si Wafers
Previous Article in Journal
Structural Response of Polyethylene Foam-Based Sandwich Panels Subjected to Edgewise Compression
Previous Article in Special Issue
Preparation and Optoelectronic Characteristics of ZnO/CuO-Cu2O Complex Inverse Heterostructure with GaP Buffer for Solar Cell Applications
Article Menu

Export Article

Materials 2013, 6(10), 4565-4573; https://doi.org/10.3390/ma6104565

Article
Development of Screen-Printed Texture-Barrier Paste for Single-Side Texturization of Interdigitated Back-Contact Silicon Solar Cell Applications
1
Department of Chemistry, National Cheng Kung University, Tainan 70101, Taiwan
2
Department of Electro-Optical Engineering, National Formosa University, Huwei, Yunlin 63201, Taiwan
3
Institute of Mechanical and Electro-Mechanical Engineering, National Formosa University, Huwei, Yunlin 63201, Taiwan
*
Authors to whom correspondence should be addressed.
Received: 10 July 2013; in revised form: 8 October 2013 / Accepted: 12 October 2013 / Published: 17 October 2013

Abstract

:
Continuous cost reduction of silicon-based solar cells is needed to lower the process time and increase efficiency. To achieve lower costs, screen-printed texture-barrier (SPTB) paste was first developed for single-side texturization (ST) of the interdigitated back-contact (IBC) for silicon-based solar cell applications. The SPTB paste was screen-printed on silicon substrates. The SPTB paste was synthesized from intermixed silicate glass (75 wt %), a resin binder (ethyl cellulose ethoce: 20 wt %), and a dispersing agent (fatty acid: 5 wt %). The silicate glass is a necessity for contact formation during firing. A resin binder and a dispersing agent determine the rheology of the SPTB paste. In this work, by modulating various parameters, including post SPTB firing, alkali texturing, and removal of the SPTB, the ST of IBC silicon solar cells was achieved. Since the advantages of the SPTB paste include low toxicity and prompt formation of the texture-barrier, SPTB is potentially suited for simple fabrication at low-cost for solar cell applications. The cost of the SPTB is around $100/kg which is lower than the SiH4/NH3 gas ambient used in plasma-enhanced chemical vapor deposition (PECVD). Thus, the expensive Si3N4 film deposited by PECVD using SiH4 and NH3 gas ambient for silicon solar cells can be replaced by this SPTB.
Keywords:
screen-printed texture-barrier (SPTB); interdigitated back-contact (IBC); photovoltaic silicon solar cells; paste

1. Introduction

Screen-printed mono- and multi-crystalline silicon solar cells have been widely used for industrial solar cells [1,2,3,4,5,6,7]. However, both the front contact and rear side of the screen-printed silicon solar cells (SPSSCs) have been the major obstacle for conversion efficiency (CE) enhancements. This is due to optical loss of the top-contact coverage shading and poor surface recombination. Since maximum CE of the screen-printed mono-crystalline silicon solar cells is around 19.5%, the improvement of SPSSC efficiency is highly desirable. Thus, a CE of 24.2% for interdigitated back-contact (IBC) silicon solar cells has attracted great attention [8,9,10]. Compared with conventional solar cells with front side metal grids, IBC solar cells with interdigitated contact of opposite polarity on the back surface of the cells have various advantages, including improved photo-generation, reduced grid series resistance, improved blue photo-response, and simplified module assembly [11]. In recent years, the IBC solar cells have cells made with aluminum-alloyed emitters using screen-printed technology [12,13]. Moreover, the random pyramids with height of around 2–10 μm in alkali texturing solution were overly larger for the rear side metallization of the IBC solar cells. Therefore, single-side texturization (ST) in IBC solar cells with a flattened back side is preferable. Previous literature reported that the ST on the front surface of the IBC solar cell was achieved by covering a SiNx protection layer on the rear side of the silicon substrate [13,14]. The high costs of a SiNx texture-barrier as deposited by plasma-enhanced chemical vapor deposition (PECVD) are here identified. The cost of the screen-printed texture-barrier (SPTB) is around $100/kg which is lower than the SiH4/NH3 gas ambient used in PECVD. Following this, our study demonstrates a SPTB paste that was fabricated simply at low-cost using screen-printed technology.

2. Experimental Section

To develop the SPTB paste for ST applications of the IBC silicon solar cells, all square samples with 4 cm2 were cut from (100)-oriented p-type monocrystalline silicon wafer with ρ = 0.5 − 3 Ω cm. The sheet resistance of the p-type Si(100) substrate was measured to be around 130 Ω/sq. Prior to the SPTB process, all wafers were cleaned with a wet process. Wet processes, including an H2SO4/H2O2 mixture in a ratio of 4:1 (volume) at 120 °C, an NH4OH/H2O2/H2O mixture in a ratio of 1:4:20 (volume) at 75 °C and an HCl/H2O2/H2O mixture in a ratio of 1:1:6 (volume) at 75 °C, were used to remove residues and contaminants. The dipping time in all chemical processes was 10 min. Diluted HF (DHF) with HF and H2O in a ratio of 1:100 (volume) was used to remove native oxide. Following chemical processes, all wafers were rinsed with deionized water for 10 min. Then, the prepared SPTB paste was screen-printed onto the cleaned wafer. The process of making SPTB consists of mixing silicon glass powder (75 wt %), organic resin binder (ethyl cellulose ethoce: 20 wt %), and dispersing agent (fatty acid: 5 wt %) in an agitation tank. The ready-mixed materials are dispersed in a roll mill to finally form the SPTB paste. The silicate glass in the SPTB paste is a necessity for contact formation during firing. A resin binder and a dispersing agent determine the rheology of the SPTB paste. The viscosity of the SPTB paste can be tuned by the ratio of the silicate glass, the resin binder, and the dispersing agent. Then, the stacked SPTB paste/p-type Si(100) substrates were fired at 840 °C for 10, 15, 20, 30, 40, 60, 80 and 100 min, respectively. Furthermore, the SPTB paste/p-type Si(100) substrates were textured in a mixture solution of 1.7% potassium hydroxide (KOH) for the ST process. The temperature and the texturing time of the mixture solution were maintained at 83 °C for 15, 20, 25 and 30 min, respectively. After the ST process, the SPTB pastes protected on p-type Si(100) substrates were removed by the mixed HCl/H2O2 solution at 10, 30 and 60 min, respectively. The process used an HCl/H2O2/H2O mixture in a ratio of 1:1:4 (volume) at 75 °C to remove the SPTB protected layer. Finally, all surface morphologies and cross-section images were detected by a scanning electron microscopy (SEM) (Hitachi, S3000, Tokyo, Japan). All sheet resistances of the surfaces were evaluated by four-point probe current-voltage meters. Since the SPTB paste is similar to the Al paste used in mono-crystalline silicon solar cells, a large size wafer such as 156 × 156 mm2 can be screen-printed by design of the large screen.

3. Results and Discussion

Firstly, to confirm the quality of the texturization, the p-type Si(100) substrates were textured in a mixed solution of 1.7% KOH. The temperature and texturing time of the texturing solution were maintained at 83 °C for 15, 20, 25 and 30 min, respectively. The cross-section images detected by SEM for (a) 15, (b) 20, (c) 25 and (d) 30 min, respectively, are shown in Figure 1.
Figure 1. Cross-section images of the p-type Si(100) substrate after texturing at (a) 15, (b) 20, (c) 25 and (d) 30 min, respectively, were detected by scanning electron microscopy (SEM). The surface morphologies of the p-type Si(100) with the texturing time of 25 min are shown in the inset of (c).
Figure 1. Cross-section images of the p-type Si(100) substrate after texturing at (a) 15, (b) 20, (c) 25 and (d) 30 min, respectively, were detected by scanning electron microscopy (SEM). The surface morphologies of the p-type Si(100) with the texturing time of 25 min are shown in the inset of (c).
Materials 06 04565 g001
The height of the pyramid was measured to be around 1–7 μm. Since pyramids are formed by anisotropical KOH etching on a single crystalline silicon (111) surface, the height of the pyramid was increased by increasing texturing time. Compared with the sample textured at 25 min, a lower height of the pyramid was obtained for samples textured at 15 and 20 min. On the other hand, some top conoids of the pyramid were etched for the sample textured at 30 min, as shown in Figure 1c. The reflection could be increased by the flattened top surface of the pyramid. Thus, according to the surface morphologies and the cross-section image detected by SEM as shown in Figure 1, a suitable texturing time of 25 min would be expected in this work. Therefore, the ST process of the IBC silicon solar cells is achieved at 83 °C for 25 min.
To demonstrate the ST of the IBC silicon solar cell, all SPTB layer/p-type Si(100) substrates with firing at 10, 20, 40, 60, 80, and 100 min, respectively, were textured in a 1.7% KOH solution for 25 min. Figure 2 shows the cross-section images of all SPTB layer/p-type Si(100) substrates. Figure 3 shows the surface morphologies and the cross-section images of the p-type Si(100) substrates without a SPTB layer side. The height of the pyramid was measured to be around 4∓7 μm. It can be clearly demonstrated that the SPTB layer/p-type Si(100) substrates fired below 60 min cannot be adopted for ST due to the KOH solution that passes through the voids in the SPTB layer (Figure 2a–c). On the contrary, there is no texturization between the SPTB layer/p-type Si(100) substrates interface for the SPTB layer fired above 60 min (Figure 2d,f). The adhesion between the SPTB layer/p-type Si(100) substrate interface increases on increasing the firing time. Therefore, the SPTB layer/p-type Si(100) substrate fired at 840 °C for 60 min are suitable for the preparation of the ST.
Figure 2. Cross-section images of screen-printed texture-barrier (SPTB) layer/p-type Si(100) substrates fired at 840 °C for (a) 10, (b) 20, (c) 40, (d) 60, (e) 80, and (f) 100 min, respectively, and textured in a 1.7% KOH solution for 25 min.
Figure 2. Cross-section images of screen-printed texture-barrier (SPTB) layer/p-type Si(100) substrates fired at 840 °C for (a) 10, (b) 20, (c) 40, (d) 60, (e) 80, and (f) 100 min, respectively, and textured in a 1.7% KOH solution for 25 min.
Materials 06 04565 g002
Figure 3. The SPTB layer/p-type Si(100) substrates were textured in 1.7% KOH solution for 25 min. (a) the surface morphology and (b) the cross-section image of the p-type Si(100) substrate without a SPTB layer was detected by SEM.
Figure 3. The SPTB layer/p-type Si(100) substrates were textured in 1.7% KOH solution for 25 min. (a) the surface morphology and (b) the cross-section image of the p-type Si(100) substrate without a SPTB layer was detected by SEM.
Materials 06 04565 g003
To investigate the surface morphologies of the SPTBs deposited on the p-type Si(100) substrates with various firing times, the stacked SPTB/p-type Si(100) substrate structures were fired at 840 °C for (a) 10, (b) 15, (c) 30 and (d) 60 min, respectively, as shown in Figure 4. The results show that many voids in the SPTB layers were detected for the 10 min sample. The voids in the SPTB layers decreased with increasing firing time of the SPTB layer as shown in Figure 4d. This is due to the flow of the SPTB paste determined by high-temperature firing. When alkali texturing is presented, the KOH solution passes through the voids in the SPTB layer and then etches the silicon surface. On the other hand, the SPTB layer with long time firing is difficult to remove after single-side texturization. Fewer voids result in shorter etching times of the silicon surface, thus introducing a trade-off between maintaining fewer voids and achieving a high removal rate of the SPTB layer after alkali texturing.
The process sequences of ST are screen-printed SPTB paste, fired SPTB layer/p-type Si(100) substrate, texturing in KOH solution, and removal of the SPTB paste from the stacked SPTB layer/p-type Si(100) substrate. Thus, after alkali texturing of the SPTB layer/p-type Si(100) substrate, the SPTB layer should have been removed. To investigate the removal efficiency of the SPTB layer, the stacked SPTB layer/p-type Si(100) substrates were fired at 840 °C for 60 min and textured in 1.7% KOH solution for 25 min. Then, the surface morphologies and the cross-section images of the SPTB layer/p-type Si(100) substrate etched in the mixed HCl/H2O2 solution at (a,b) 10, (c,d) 30, (e,f) 60 min, respectively, were detected by SEM as shown in Figure 5. The results show that some SPTB residue was detected on the surface of the p-type Si(100) substrate etched in the mixed HCl/H2O2 solution at 10 min. Conversely, the SPTB layer can be removed under the SPTB layer/p-type Si(100) substrate etched in the mixed HCl/H2O2 solution at 60 min. The removal efficiency can be enhanced by increasing etching time.
To confirm the SPTB residue on the p-type Si(100) surface, the sheet resistances of the p-type Si(100) surfaces with the removal of the SPTB layer were measured by four-point probe. The sheet resistances of the SPTB layer/p-type Si(100) substrate etched in mixed HCl/H2O2 solution at 10, 30, and 60 min, respectively, were measured and are shown in Figure 6. The sheet resistance of original p-type Si(100) substrate is around 130 Ω/sq. The removal efficiency is enhanced with increasing etching time. Compared with original sheet resistance, higher sheet resistance was obtained for the SPTB layer/p-type Si(100) substrate etched in mixed HCl/H2O2 solution at 10 min. This is due to the SPTB residue on the surface. On the other hand, the sheet resistance of the SPTB layer/p-type Si(100) substrate etched in mixed HCl/H2O2 solution at 60 min was around 130 Ω/sq. Thus, the SPTB layer was completely removed by the HCl/H2O2 solution at 60 min. These results were consistent with the results as shown in Figure 5.
Figure 4. Surface morphologies of screen-printed SPTB on p-type Si(100) substrates fired at 840 °C for (a) 10, (b) 15, (c) 30, and (d) 60 min, respectively.
Figure 4. Surface morphologies of screen-printed SPTB on p-type Si(100) substrates fired at 840 °C for (a) 10, (b) 15, (c) 30, and (d) 60 min, respectively.
Materials 06 04565 g004
Figure 5. The SPTB layer/p-type Si(100) substrates were fired at 840 °C for 60 min and textured in 1.7% KOH solution for 25 min. Then, surface morphologies and cross-section images of the SPTB layer/p-type Si(100) substrates etched in mixed HCl/H2O2 solution at (a,b) 10, (c,d) 30, (e,f) 60 min, respectively, were detected by SEM.
Figure 5. The SPTB layer/p-type Si(100) substrates were fired at 840 °C for 60 min and textured in 1.7% KOH solution for 25 min. Then, surface morphologies and cross-section images of the SPTB layer/p-type Si(100) substrates etched in mixed HCl/H2O2 solution at (a,b) 10, (c,d) 30, (e,f) 60 min, respectively, were detected by SEM.
Materials 06 04565 g005aMaterials 06 04565 g005b
Figure 6. Sheet resistances of the SPTB layer/p-type Si(100) substrates etched in mixed HCl/H2O2 solution at 10, 30, and 60 min, respectively.
Figure 6. Sheet resistances of the SPTB layer/p-type Si(100) substrates etched in mixed HCl/H2O2 solution at 10, 30, and 60 min, respectively.
Materials 06 04565 g006
Finally, to demonstrate the ST of the IBC silicon solar cell, the SPTB layer/p-type Si(100) substrates were fired at 10, 20, 40, 60, 80, and 100 min, respectively. Then, all SPTB layer/p-type Si(100) substrates were textured in 1.7% KOH solution for 25 min. Finally, all SPTB layer/p-type Si(100) substrates were etched in the mixed HCl/H2O2 solution for 60 min. Figure 7 shows all surface morphologies of the p-type Si(100) substrate after SPTB removal detected by SEM. Again, the slight texturing on the surface between the SPTB layer and p-type Si(100) substrate was obtained for the SPTB layer/p-type Si(100) substrate fired below 60 min. With short firing time of the SPTB paste, the KOH solution passes easily through the voids on the SPTB layer. Thus, the ST of the IBC silicon solar cell can be demonstrated by the SPTB layer/p-type Si(100) substrate fired above 60 min. These results are consistent with Figure 2.
Figure 7. The SPTB layer/p-type Si(100) substrates were fired at (a) 10, (b) 20, (c) 40, (d) 60, (e) 80 and (f) 100 min, respectively. Then, all SPTB layer/p-type Si(100) substrates were textured in 1.7% KOH solution for 25 min. Finally, the SPTB layer/p-type Si(100) substrates were etched in mixed HCl/H2O2 solution for 60 min. All surface morphologies of the p-type Si(100) substrate after SPTB removal were detected by SEM.
Figure 7. The SPTB layer/p-type Si(100) substrates were fired at (a) 10, (b) 20, (c) 40, (d) 60, (e) 80 and (f) 100 min, respectively. Then, all SPTB layer/p-type Si(100) substrates were textured in 1.7% KOH solution for 25 min. Finally, the SPTB layer/p-type Si(100) substrates were etched in mixed HCl/H2O2 solution for 60 min. All surface morphologies of the p-type Si(100) substrate after SPTB removal were detected by SEM.
Materials 06 04565 g007

4. Conclusions

The screen-printed texture-barrier (SPTB) paste for single-side texturization (ST) of IBC silicon solar cell applications has been demonstrated. The SPTB paste consists of silicate glass (75 wt %), a resin binder (20 wt %) and a dispersing agent (5 wt %). The results show that the voids in the SPTB layer were decreased with increased firing times for the SPTB layer. There is no texturization between the SPTB layer/p-type Si(100) substrates interface for the SPTB paste fired above 60 min. The SPTB layer can be removed by mixed HCl/H2O2 solution over 60 min. The complete removal of the SPTB layer can be demonstrated by the surface morphology and the sheet resistance of the silicon surface. Furthermore, the cost of the screen-printer and the high-temperature furnace is cheap, around half of that of plasma-enhanced chemical vapor deposition (PECVD). The cost of the SPTB is around $100/kg which is lower than the SiH4/NH3 gas ambient used in PECVD. Thus, the fabricated cost of the IBC solar cells can be reduced by this new texturization-barrier paste.

Acknowledgments

The authors would like to thank the National Science Council of the Republic of China for the financial support under the contract number NSC 102-2221-E-150-040, and NSC 101-2622-E-150-020-CC3. Technical support from National Nano Device Laboratories (NDL) of the R.O.C., Common Laboratories for Micro/Nano Science and Technology of National Formosa University, and Advanced Electronic Materials (ITK) Inc. are also acknowledged.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Neuhaus, D.-H.; Munzer, A. Industrial silicon wafer solar cells. Adv. OptoElectron. 2007, 2007, 24521:1–24521:15. [Google Scholar] [CrossRef]
  2. Duerickx, F.; Szlufcik, J. Defect passivation of industrial multicrystalline solar cells based on PECVD silicon nitride. Sol. Energy Mater. Sol. Cells 2002, 72, 231–246. [Google Scholar] [CrossRef]
  3. Singh, P.; Shivaprasad, S.-M.; Lal, M.; Husain, M. Angle-dependent XPS analysis of silicon nitride film deposited on screen-printed crystalline silicon solar cell. Sol. Energy Mater. Sol. Cells 2009, 93, 19–24. [Google Scholar] [CrossRef]
  4. Liu, C.-W.; Cheng, C.-L.; Dai, B.-T.; Yang, C.-H.; Wang, J.-Y. Fabrication and photovoltaic characteristics of coaxial silicon nanowire solar cells prepared by wet chemical etching. Int. J. Photoenergy 2012, 2012, 701809:1–701809:8. [Google Scholar]
  5. Müller, J.; Bothe, K.; Gatz, S.; Plagwitz, H.; Schubert, G.; Brendel, R. Recombination at local aluminum-alloyed silicon solar cell base contacts by dynamic infrared lifetime mapping. Energy Procedia 2001, 8, 337–342. [Google Scholar] [CrossRef]
  6. Aberle, A.-G.; Hezel, R. Progress in low-temperature surface passivation of silicon solar cells using remote-plasma silicon nitride. Prog. Photovol. 1997, 5, 29–50. [Google Scholar] [CrossRef]
  7. Cheng, C.-L.; Yang, J.-Y. Hydrothermal synthesis of Eu3+-doped Y(OH)3 nanotubes as downconversion materials for efficiency enhancement of screen-printed monocrystalline silicon solar cells. IEEE Electron Device Lett. 2012, 33, 697–699. [Google Scholar] [CrossRef]
  8. Cousins, P.J.; Smith, D.D.; Luan, H.-C.; Manning, J.; Dennis, T.D.; Waldhauer, A.; Wilson, K.E.; Harley, G.; Mulligan, W.P. Generation 3: Improved Performance at Lower Cost. In 35th IEEE Photovoltaic Specialists Conference, Honolulu, HI, USA, 20–25 June 2010.
  9. Cuevas, A.; Kerr, M.-J.; Samundsett, C.; Ferrazza, F.; Coletti, G. Millisecond minority carrier lifetimes in n-type multicrystalline silicon. Appl. Phys. Lett. 2002, 81, 4952–4954. [Google Scholar] [CrossRef]
  10. Glunz, S.-W.; Rein, S.; Lee, J.-Y.; Warta, W. Minority carrier lifetime degradation in boron-doped czochralski silicon. J. Appl. Phys. 2001, 90, 2397–2404. [Google Scholar] [CrossRef]
  11. Mulligan, W.-P.; Cudzinovic, M.-J.; Pass, T.; Smith, D.; Kaminar, N.; McIntosh, K.; Swanson, R-.M. Solar cell and method of manufacture. U.S. Patent US7,897,867B1, 1 March 2011. [Google Scholar]
  12. Galbiati, G.; Mihailetchi, V.-D.; Halm, A.; Roescu, R.; Kopecek, R. Results on n-type IBC solar cells using industrial optimized techniques in the fabrication processing. Energy Procedia 2011, 8, 421–426. [Google Scholar] [CrossRef]
  13. Woehl, R.; Krause, J.; Granek, F.; Biro, D. 19.7% efficient all-screen-printed back-contact back-junction silicon solar cell with aluminum-alloyed emitter. IEEE Electron Device Lett. 2011, 32, 345–347. [Google Scholar] [CrossRef]
  14. Bock, R.; Mau, S.; Schmidt, J.; Brendel, R. Back-junction back-contact n-type silicon solar cells with screen-printed aluminum-alloyed emitter. Appl. Phys. Lett. 2010, 96, 263507:1–263507:3. [Google Scholar] [CrossRef]
Materials EISSN 1996-1944 Published by MDPI AG, Basel, Switzerland RSS E-Mail Table of Contents Alert
Back to Top