A novel scheme of direct electrical contact on vertically aligned silicon nanowire (SiNW) axial p-n junction is demonstrated by means of oblique-angle deposition of slanted indium-tin-oxide (ITO) film for photovoltaic applications. The slanted ITO film exhibits an acceptable resistivity of 1.07x10−3Ω-cm underwent RTA treatment of T = 450°C, and the doping concentration and carrier mobility by Hall measurement amount to 3.7x1020cm−3 and 15.8cm2/V-s, respectively, with an n-type doping polarity. Because of the shadowing effect provided by the SiNWs, the incident ITO vapor-flow is deposited preferentially on the top of SiNWs, which coalesces and eventually forms a nearly continuous film for the subsequent fabrication of grid electrode. Under AM 1.5G normal illumination, our axial p-n junction SiNW solar cell exhibits an open circuit voltage of VOC = 0.56V, and a short circuit current of JSC = 1.54 mA/cm2 with a fill factor of FF = 30%, resulting in a total power conversion efficiency of PEC = 0.26%.
© 2012 OSA
The main constraint of widespread utilization of silicon (Si) solar cell (SC) stems from the high cost of pure materials that supports sufficient long diffusion length of minority carrier and effective collection of photogenerated carrier [1, 2]. To overcome the cost issue, several approaches, such as the use of thin-film architecture and nanostructured geometry, have been widely investigated [3–9]. Among them, the Si nanowire (SiNW) has attracted much attention due to its natural profile associated with the core-shell geometry, which has a benefit of orthogonalizing the direction of light absorption and charge-carrier collection [10–12]. Yet the open-circuit voltage (VOC) of core-shell SiNW SC (especially for the thin and long SiNW, i.e. < 1μm in diameter and >10μm in length) is much sensitive to the trap density in the depletion and quasi-neutral regions versus planar Si SC, entirely due to the increased effective area of radial p-n junction . Thus as compared to its planar counterpart, the core-shell SiNW SC generally exhibits the smaller VOC value [14–17], and that hinders the ultimate performance of power conversion efficiency (PEC). Although the to-date reported PEC of core-shell SiNW SC is around 10%, it is still far below the record efficiency of 25% for Si wafer-based SC . Recently Wong et al. have reported that the photogenerated carriers are mainly concentrated in the nanowire itself, primarily attributable to the confinement effect of incident photons . Therefore, even for SiNW SC embedded with axial p-n junction, the collection of charge-carrier is facilitated without compromising of VOC value, as the excess recombination of charged-carrier associated with trap density is significantly reduced. Yet for such axial p-n junction SiNW configuration, making an electrical contact related to depositing a continuous conductive film on the top of SiNWs is difficult, as its junction sidewall is completely exposed to the air and thus easily induces undesirable leakage electrical path of charged-carrier during cell fabrication . Perraud et al. conducts a feasible approach , which buries SiNW arrays in a spin-on glass (SOG) matrix and then planarizes the front surface by chemical-mechanical polishing (CMP) for the subsequent deposition of conductive film, to demonstrate an axial p-n junction SiNW SC. However, the complete filling of SOG and planarization process of CMP with acceptable uniformity and thickness is demanding and that also adds the extra process and manufacture cost of cell fabrication, adversely affecting the popularization of SiNW solar cell. In this work, we demonstrate a general strategy involving oblique-angle deposition scheme that directly integrates SiNWs embedded with axial p-n junction into photovoltaic devices.
Figure 1(a) illustrates fabrication process of the proposed SiNW SC using oblique-angle deposition scheme. The primary process sequence includes following steps: (1) Formation of p-n junction on n-type (0.02 Ω-cm) (100)-oriented crystalline Si substrate by means of the diborane (B2H6) diffusion in the quartz furnace at 850°C; (2) immersion in aqueous solution of 4.8M HF/0.005M AgNO3 to facilitate the redox reaction and form Ag nano-particles on the top of wafer surface (4Ag+ + Si + 6F-→4Ag + [SiF6]2-) ; (3) metal-induced anisotropic chemical etching to form SiNWs by immersing in aqueous solution of 4.8M HF/0.5M H2O2 with assistance of Ag nano-particles that functions as the catalyst for Si oxidation (Si + 2H2O2 + 6HF→H2SiF6 + 4H2O) ; (4) chemical cleaning by rinsing HNO3 solution to dissolve the remaining Ag nano-particles and remove other contaminations in SiNWs; and (5) oblique-angle deposition by RF magnetron sputtering to selectively grow slanted indium-tin-oxide (ITO), which interconnects the top (p-type) of individual SiNWs and provides a continuous and conductive film for the subsequent mask-evaporation of silver front grid. The corresponding energy-band diagram of the device is also plotted in the figure. Figure 1(b) shows the cross-sectional scanning electron microscopy (SEM) image of the device. The top-view SEM image of slanted ITO film (left-up) and cross-sectional SEM image of SiNWs (left-down) are also inserted in the figure. Accordingly, the nanowires are randomly distributed and vertically aligned with 2μm in length, which is controllable by the duration time of chemical etching. The typical wire diameter from SEM is 10–100nm with 35% area density, similar to previous reports using the same SiNW synthesis approach . Furthermore, because of the shadowing effect provided by the SiNWs [24, 25], the incident ITO vapor-flow (θ = 60°) is deposited preferentially on top of the SiNWs, and it eventually coalesces altogether. This coalescence forms an optically transparent and electrically conductive ITO film (d = 400nm) with a nearly continuous surface morphology (RMS = 30.7nm), providing an excellent platform for the subsequent fabrication of grid electrode.
3. Results and discussion
Figure 2(a) shows the X-ray diffraction (XRD) patterns of the slanted ITO film for different rapid thermal annealing (RTA) treatments (T = 250–475°C) to improve the crystallographic quality and electrical conductivity. Accordingly, the diffraction peaks of slanted ITO film that underwent RTA treatment exhibit the crystallographic cubic structure with preferred orientations in the (211), (222), (400), (440), and (622) planes, those were observed at 2θ = 21.8°, 30.86°, 35.68°, 51.22°, and 60.78°, respectively. With the increasing of RTA temperature till to T = 450°C, the measured diffraction peaks become higher and sharper, especially for the (222) diffraction peak. It suggests that the degree of crystallinity of the slanted ITO film is gradually improved for higher RTA temperature, attributable to the sufficient thermal energy that provides the diffusion of deposited atoms, and thus redistributes them much orderly; however, further increasing of RTA treatment to T = 475°C will adversely affect the crystallographic quality as the (222) diffraction peak of slanted ITO film is obviously reduced. Figure 2(b) plots the full-width at half maximum (FWHM) of the (222) diffraction peak and the measured resistivity of slanted ITO film versus the RTA temperature. Indeed the lowest FWHM of 0.6° was observed on the RTA temperature of T = 450°C, which corresponds to 14.3 nm of the grain size of slanted ITO film estimated by the Scherrer formula . The electrical analysis of slanted ITO film is characterized by means of four-point probe and Hall measurement. The Hall measurements are taken in a Van-der-Pauw geometry at room temperature under a nominal magnetic field of 1.5T. Accordingly, the measured resistivity also exhibits the lowest value of 1.07x10−3Ω-cm at RTA temperature of T = 450°C, which is one order of magnitude higher than that of planar one due to the inherited porosity structure of slanted ITO film that increases the electron scattering . Nevertheless it is still acceptable to perform as a transparent conductive film for photovoltaic applications, and the RTA temperature is therefore determined to be T = 450 °C. The sheet resistance of slanted ITO film with a measured thickness of 350nm is 30.6Ω/☐, while the corresponding doping concentration and carrier mobility amount to 3.7x1020cm−3, and 15.8cm2/V-s, respectively, with an n-type doping polarity.
The following discussion concerns the optical characteristics of the samples of interest. Three different samples, including the planar Si wafer coated with and without quarter-wavelength antireflection coating (ARC), and the proposed SiNWs solar cell, were fabricated and compared in this work. Figure 3(a) shows the photographs of representative samples with the identical sizes of 1cm × 1cm. The planar Si wafer (left) has a mirror-like surface, and becomes to reddish colour with ARC (center). Visually, the proposed SiNW SC (right) displays the darkest appearance compared to the other samples, suggesting the oblique-angle deposition of slanted ITO film barely affects the low reflectivity of SiNWs . Furthermore, the image contour is uniform with reasonable fluctuations, indicates that the fabrication process of proposed SiNW SC is stable and reliable. In the top of Fig. 3(b), we plot the measured reflectivity versus the incident wavelength of λ = 400–700nm for all samples. The profile of measured reflectivity of planar Si wafer (black line) decreases monotonically from R = 48.5% to R = 33.9% as the incident wavelength increases, because of the slight decrease of the refractive index of Si with wavelength. On average, the reflectivity of the planar Si wafer is R = 38.1%. The measured reflectivity is reduced for unmodified Si wafer with the quarter-wavelength ARC (blue line), and exhibits nearly perfect anti-reflectivity at the incident wavelength of λ = 500nm where the strongest irradiance over the solar spectrum occurs. However, the corresponding average reflectivity of R = 21.3% is still too high to harvest sunlight efficiently. Significantly, the proposed SiNW SC shows a reduction in reflectivity of one order of magnitude over the full spectrum of visible-light, and the measured profile stays steady with variation of incident wavelengths. This decrease is primarily caused by the randomly distributed SiNWs, which effectively scatters and traps incident photons, resulting in the suppression of optical reflection between interfaces. As a result, an extremely low reflectivity with average value of R = 5.1% is achievable over the visible-light spectrum. It shall be noted here that in the industry of solar cell, the device surface was generally textured by the chemical wet etching or reactive-ion dry etching before antireflection coating, which significantly reduces reflection losses in solar cells and achieves a low reflectivity of about 10% . However, in this work, the quarter-wavelength ARC was deposited directly on the planar Si wafer without the treatment of surface texture. Thus the average reflectivity of our Si wafer deposited with ARC is almost twice that of the industry sample with surface treatment. Nevertheless, the average reflectivity of our proposed SiNWs SC is around half that of the industry sample, suggesting its great potential for photovoltaic applications.
To evaluate the effect of incident angle of sunlight on the PEC throughout the operating day of a non-tracking solar cell, Fig. 3(b) also performs the angular-dependent reflectivity, R(θ, λ), based on a incident light of He-Ne laser (λ = 632.8nm), and a broad range of incident angles (θ = 10°–70°), as well as for the transverse electric (TE) and transverse magnetic (TM) polarizations. Accordingly, the measured profiles of the all sample display a fundamental behavior of the external reflection , i.e. for TM light, the reflectivity gradually decreases with θ until reaching the Brewster angle, thus leading to lower reflectance than that of TE light. Nevertheless, the measured reflectivity of the proposed SiNW SC is relatively insensitive to θ and the polarization of incident light. Additionally, the corresponding reflectivity is significantly lower than those of the other samples for both TE and TM polarizations. Above observations associated with the angular-dependent reflectivity shows that the proposed SiNW SC has a great advantage over the other two competitors in terms of the collection of solar energy regardless of sun’s location in the sky. Similarly, the optical absorption of all samples obtained by A(θ, λ) = 1–R(θ, λ) were plotted in Fig. 3(c), where the transmission is negligible because the thickness of the wafer substrate is larger than 200 μm. Across the incident wavelengths studied here, the peak absorption of planar Si wafer (top) is relatively low, limiting the practical applications for solar cells. With the assistance of the quarter-wavelength ARC (middle), the optical absorption is mainly increased at incident wavelengths of λ = 500 ± 50nm, in which the destructive interference of incident light occurs, causing A(θ, λ) to exhibit a band-like profile. Thus the overall enhancement of optical absorption is still insignificant. Restated, the proposed SiNWs SC (bottom) cannot only suppresses the optical reflection, but virtually eliminates the angular sensitivity of A(θ, λ). Consequently, the corresponding A(θ, λ) is dramatically enhanced and nearly angle-independent over the full visible-light spectrum.
To characterize the electrical properties of the proposed SiNW SC, current density versus voltage (J-V) is measured both in dark conditions and AM 1.5G normal illumination (100mW/cm2, 1sun) at room temperature. Figure 4(a) displays the J-V characteristics of the proposed SiNWs SC on a semi-log scale both in the dark and under light. The photography of actual device with the dimension of 2cmx2cm is also inserted in the figure. According to the dark J-V curve, leakage currents in the order of 9.95x10−5A at −1V are measured and the rectification ratio of 45 ( ± 1V) is achieved, exhibiting a well-behaving current rectification of p-n junction. The series and shunt resistances extracted from the dark J-V curve are 118Ω and 7.5kΩ, respectively. As compared to the standard bulk Si solar cells , the relatively large series resistance of the proposed SiNW SC mainly stems from a doping mis-alignment of the type p/n+ for top SiNW/slanted ITO, which induces a potential barrier and increases the contact resistance hindering the extraction of drifted carriers (holes) outside the device, as the depiction of energy-band diagram shown in Fig. 1(a). While the much smaller shunt resistance is mainly due to the high surface to volume ratio in the SiNWs, which exacerbates the surface recombination that generally accompanies with high values of p-n junction reverse current (JR = 3.7 μA) and ideality factor (n = 2.9) . Additionally, such high JR and n values also indicate a high density of localized states in the SiNWs, which acts as generation-recombination centers of phtogenerated minority carriers , and diminishes the ultimate PEC of the device as well. It suggests that for our future study the surface passivation in the proposed SiNW SC is necessary as such structure retains an extremely high surface-to-volume ratio. For comparison, the J-V curves of SiNW arrays deposited with standard ITO film (perpendicular angle, θ = 0°) both in the dark and under AM 1.5G illumination were plotted and inserted in Fig. 4(a). The measured J-V curves of the SiNW arrays deposited with standard ITO film in the dark and under light are almost identical, and are symmetrical for the applied bias-voltage from −1 to 1V. Notably, no current rectification behavior is observed. It is because that the perpendicular ITO vapor-flow was deposited conformally onto surfaces of nanowires, bypassing the p-n junction region of SiNWs and thus exhibiting the bulk-like electrical property. Most importantly, it indicates that our proposed scheme by means of oblique-angle deposition of slanted ITO film to make an electrical contact on the axial p-n junction SiNW SC is reliable and promising.
Figure 4(b) re-plots the J-V curves in linear scale both in the dark and under light to identify several important parameters of the proposed SiNW SC. The power-density versus voltage curve is also inserted in the figure. Under AM 1.5G illumination, a clear boost of measured current was observed in the reverse-bias region. The device exhibits an open circuit voltage (VOC) of 0.56V, and a short circuit current (JSC) of 1.54mA/cm2 with a fill factor (FF) of 30% resulting in a total PEC of 0.26%. The small values of JSC and FF of the device can mainly be blamed to the un-optimized alignment of doped polarity obstructing the collection of drifted current, as well as the undesirable surface recombination consuming the photogenerated minority carriers, all of which leads the device’s maximum power-density as small as 0.26mW/cm2.
In conclusion, this work presents a novel approach using oblique-angle deposition of slanted ITO film to integrate metal-induced chemical-etching SiNW arrays into solar cells. Our prototype SiNW solar cells show efficiencies up to about 0.26%, much lower than that of standard bulk Si solar cell, yet with great potential based on the analysis of angular-dependent reflectivity. Ongoing research in the future to enhance the efficiency of the device is focused on decreasing series and shunt resistances by appropriately aligning doped polarity of contact materials, and by diminishing the interfacial recombination of photocarriers via surface passivation.
The authors gratefully acknowledge financial support from the National Science Council of Republic of China (ROC) in Taiwan (contract No. NSC–100–2112–M–003–006–MY3) and from the National Taiwan Normal University (NTNU100-D-01).
References and links
1. L. Tsakalakos, “Nanostructures for photovoltaics,” Mater. Sci. Eng. 62(6), 175–189 (2008). [CrossRef]
2. C. A. Wolden, J. Kurtin, J. B. Baxter, I. Repins, S. E. Shaheen, J. T. Torvik, A. A. Rockett, V. M. Fthenakis, and E. S. Aydil, “Photovoltaic manufacturing: Present status, future prospects, and research needs,” J. Vac. Sci. Technol. A 29(3), 030801 (2011). [CrossRef]
3. A. Fujisaka, S. Kang, L. Tian, Y. L. Chow, and A. Belyaev, “Implant-cleave process enables ultra-thin wafers without kerf loss,” Photovoltaics World, pp. 21–24, Issue: May/Jun (2011).
5. O. Gunawan, K. Wang, B. Fallahazad, Y. Zhang, E. Tutuc, and S. Guha, “High performance wire-array silicon solar cells,” Prog. Photovolt. Res. Appl. 19(3), 307–312 (2011). [CrossRef]
6. K. Hadobás, S. Kirsch, A. Carl, M. Acet, and E. F. Wassermann, “Reflection properties of nanostructure-arrayed silicon surfaces,” Nanotechnology 11(3), 161–164 (2000). [CrossRef]
8. V. V. Iyengar, B. K. Nayak, and M. C. Gupta, “Optical properties of silicon light trapping structures for photovoltaics,” Sol. Energy Mater. Sol. Cells 94(12), 2251–2257 (2010). [CrossRef]
9. J. Li, H. Yu, S. M. Wong, G. Zhang, X. Sun, P. G.-Q. Lo, and D.-L. Kwong, “S nanopillar array optimization on Si thin films for solar energy harvesting,” Appl. Phys. Lett. 95(3), 033102 (2009). [CrossRef]
11. B. Tian, X. Zheng, T. J. Kempa, Y. Fang, N. Yu, G. Yu, J. Huang, and C. M. Lieber, “Coaxial silicon nanowires as solar cells and nanoelectronic power sources,” Nature 449(7164), 885–889 (2007). [CrossRef] [PubMed]
12. S. W. Boettcher, J. M. Spurgeon, M. C. Putnam, E. L. Warren, D. B. Turner-Evans, M. D. Kelzenberg, J. R. Maiolo, H. A. Atwater, and N. S. Lewis, “Energy-conversion properties of vapor-liquid-solid-grown silicon wire-array photocathodes,” Science 327(5962), 185–187 (2010). [CrossRef] [PubMed]
13. B. M. Kayes, H. A. Atwater, and N. S. Lewis, “Comparison of the device physics principles of planar and radial p-n junction nanorod solar cells,” J. Appl. Phys. 97(11), 114302 (2005). [CrossRef]
14. L. Tsakalakos, J. Balch, J. Fronheiser, B. A. Korevaar, O. Sulima, and J. Rand, “Silicon nanowire solar cells,” Appl. Phys. Lett. 91(23), 233117 (2007). [CrossRef]
16. S. Perraud, S. Poncet, S. Noël, M. Levis, P. Faucherand, E. Rouvière, P. Thony, C. Jaussaud, and R. Delsol, “Full process for integrating silicon nanowire arrays into solar cells,” Sol. Energy Mater. Sol. Cells 93(9), 1568–1571 (2009). [CrossRef]
18. M. A. Green, “The path to 25% silicon solar cell efficiency: history of silicon cell evolution,” Prog. Photovolt. Res. Appl. 17(3), 183–189 (2009). [CrossRef]
19. S. M. Wong, H. Y. Yu, J. S. Li, G. Zhang, G. Q. Lo, and D. L. Kwong, “Design high-efficiency Si nanopillar-array-textured thin-film solar cell,” IEEE Electron Device Lett. 31(4), 335–337 (2010). [CrossRef]
20. K. Rasool, M. A. Rafiq, C. B. Li, E. Krali, Z. A. K. Durrani, and M. M. Hasan, “Enhanced electrical and dielectric properties of polymer covered silicon nanowire arrays,” Appl. Phys. Lett. 101(2), 023114 (2012). [CrossRef]
22. X. Li and P. W. Bohn, “Metal-assisted chemical etching in HF/H2 O2 produces porous silicon,” Appl. Phys. Lett. 77(16), 2572 (2000). [CrossRef]
23. K. Peng, X. Wang, and S. T. Lee, “Silicon nanowire array photoelectrochemical solar cells,” Appl. Phys. Lett. 92(16), 163103 (2008). [CrossRef]
24. K. Robbie, J. C. Sit, and M. J. Brett, “Advanced techniques for glancing angle deposition,” J. Vac. Sci. Technol. B 16(3), 1115–1122 (1998). [CrossRef]
25. Y. J. Lee, S.-Y. Lin, C.-H. Chiu, T.-C. Lu, H.-C. Kuo, S.-C. Wang, S. Chhajed, J. K. Kim, and E. F. Schubert, “High output power density from GaN-based two-dimensional nanorod light-emitting diode arrays,” Appl. Phys. Lett. 94(14), 141111 (2009). [CrossRef]
26. M. I. Mendelson, “Average grain size in polycrystalline ceramics,” J. Am. Ceram. Soc. 52(8), 443–446 (1969). [CrossRef]
27. X. Xiao, G. Dong, J. Shao, H. He, and Z. Fan, “Optical and electrical properties of SnO2:Sb thin films deposited by oblique angle deposition,” Appl. Surf. Sci. 256(6), 1636–1640 (2010). [CrossRef]
28. Y.-C. Yao, M.-T. Tsai, H.-C. Hsu, L.-W. She, C.-M. Cheng, Y.-C. Chen, C.-J. Wu, and Y.-J. Lee, “Use of two-dimensional nanorod arrays with slanted ITO film to enhance optical absorption for photovoltaic applications,” Opt. Express 20(4), 3479–3489 (2012). [CrossRef] [PubMed]
29. D. H. Macdonald, A. Cuevas, M. J. Kerr, C. Samundsett, D. Ruby, S. Winderbaum, and A. Leo, “Texturing industrial multicrystalline silicon solar cells,” Sol. Energy 76(1-3), 277–283 (2004). [CrossRef]
30. M. Born and E. Wolf, Principles of optics 7th edition. Cambridge University Press, Cambridge, U.K., 46 (1999).
31. I. Tobı’as, C. del Can˜izo, J. Alonso, Handbook of Photovoltaic Science and Engineering (Wiley, New York, 2004).
32. M. A. Green, Solar Cells: Operating Principles, Technology and System Applications (Prentice-Hall, Englewood Cliffs, New Jersey, 1982).