Zinc oxide nanowire arrays for silicon core/shell solar cells

Zinc oxide nanowire arrays for silicon core/shell solar cells Asman Tamang,1 Minoli Pathirane,2 Rion Parsons,1 Miriam M. Schwarz,1 Bright Iheanacho,2 Vladislav Jovanov,1 Veit Wagner,1 William S. Wong,2 and Dietmar Knipp1,* 1

Research Center for Functional Materials and Nanomolecular Science, Jacobs University Bremen, 28759 Bremen, Germany 2 Institute of Nanotechnology, University of Waterloo, Waterloo, ON N2L 3G1, Canada *[email protected]

Abstract: The optics of core / shell nanowire solar cells was investigated. The optical wave propagation was studied by finite difference time domain simulations using realistic interface morphologies. The interface morphologies were determined by a 3D surface coverage algorithm, which provides a realistic film formation of amorphous silicon films on zinc oxide nanowire arrays. The influence of the nanowire dimensions on the interface morphology and light trapping was investigated and optimal dimensions of the zinc oxide nanowire were derived. ©2014 Optical Society of America OCIS codes: (310.6845) Thin film devices and applications; (310.6860) Thin films, optical properties; (350.6050) Solar energy.

References and Links 1. 2. 3. 4. 5. 6.

7. 8. 9. 10. 11. 12. 13. 14.

Y. Kuang, K. H. M. van der Werf, Z. S. Houweling, and R. E. I. Schropp, “Nanorod solar cell with an ultrathin a-Si: H absorber layer,” Appl. Phys. Lett. 98(11), 113111 (2011). X. Xie, X. Zeng, P. Yang, H. Li, J. Li, X. Zhang, and Q. Wang, “Radial nip structure SiNW-based microcrystalline silicon thin-film solar cells on flexible stainless steel,” Nanoscale Res. Lett. 7(621), 1–6 (2012). W. J. Nam, L. Ji, T. L. Benanti, V. V. Varadan, S. Wagner, Q. Wang, W. Nemeth, D. Neidich, and S. J. Fonash, “Incorporation of a light and carrier collection management nano-element array into superstrate a-Si: H solar cells,” Appl. Phys. Lett. 99(7), 073113 (2011). M. M. Adachi, M. P. Anantram, and K. S. Karim, “Core-shell silicon nanowire solar cells,” Sci. Rep. 3, 1465 (2013). M. M. Adachi, M. P. Anantram, and K. S. Karim, “Optical Properties of Crystalline-Amorphous Core-Shell Silicon Nanowires,” Nano Lett. 10(10), 4093–4098 (2010). M. J. Naughton, K. Kempa, Z. F. Ren, Y. Gao, J. Rybczynski, N. Argenti, W. Gao, Y. Wang, Y. Peng, J. R. Naughton, G. McMahon, T. Paudel, Y. C. Lan, M. J. Burns, A. Shepard, M. Clary, C. Ballif, F.-J. Haug, T. Söderström, O. Cubero, and C. Eminian, “Efficient nanocoax-based solar cells,” Phys. Status Solidi (RRL)Rapid Res. Lett. 4(7), 181–183 (2010). J. Zhu, Z. Yu, G. F. Burkhard, C.-M. Hsu, S. T. Connor, Y. Xu, Q. Wang, M. McGehee, S. Fan, and Y. Cui, “Optical absorption enhancement in amorphous silicon nanowire and nanocone arrays,” Nano Lett. 9(1), 279– 282 (2009). E. Garnett and P. Yang, “Light Trapping in Silicon Nanowire Solar Cells,” Nano Lett. 10(3), 1082–1087 (2010). D. L. Staebler and C. R. Wronski, “Wronski, Reversible conductivity changes in discharge produced amorphous Si,”Appl. Phys. Lett. 31(4), 292 (1977). O. Lupan, V. M. Guérin, I. M. Tiginyanu, V. V. Ursaki, L. Chow, H. Heinrich, and T. Pauporté, “Well-aligned arrays of vertically oriented ZnO nanowires electrodeposited on ITO-coated glass and their integration in dye sensitized solar cells,” J. Photochem. Photobiol. Chem. 211(1), 65–73 (2010). J. Joo, B. Y. Chow, M. Prakash, E. S. Boyden, and J. M. Jacobson, “Face-selective electrostatic control of hydrothermal zinc oxide nanowire synthesis,” Nat. Mater. 10(8), 596–601 (2011). B. Liu and H. C. Zeng, “Hydrothermal Synthesis of ZnO Nanorods in the Diameter Regime of 50 nm,” J. Am. Chem. Soc. 125(15), 4430–4431 (2003). D. J. Rogers, V. E. Sandana, F. H. Teherani, M. Razeghi, and H. J. Drouhin, “Fabrication of nanostructured heterojunction LEDs using self-forming moth eye type arrays of n-ZnOnanocones grown on p-si (111)substrates by pulsed laser deposition,” Proc. SPIE 7217, Zinc Oxide Materials and Devices IV, 721708, 721708 (2009). M. M. Schwarz, T. Richter, R. Pearson, A. Tamang, T. Balster, D. Knipp, and V. Wagner, “Controlled electrodeposition of ZnO nanostructures for enhanced light scattering properties,” J. Appl. Electrochem. 44, 1–8 (2014).

#203648 - $15.00 USD Received 26 Dec 2013; revised 19 Feb 2014; accepted 19 Feb 2014; published 17 Mar 2014 (C) 2014 OSA 5 May 2014 | Vol. 22, No. S3 | DOI:10.1364/OE.22.00A622 | OPTICS EXPRESS A622

15. J. D. Plummer, M. D. Deal, and P. B. Griffin, Silicon VLSI Technology Fundamentals, Practice and Modeling (Prentice Hall, 2000), Chap. 5. 16. R. L. Puurunen, “Surface chemistry of atomic layer deposition: A case study for the trimethylaluminum/water process,” J. Appl. Phys. 97(12), 121301 (2005). 17. C. C. Tsai, J. C. Knights, G. Chang, and B. Wacker, “Film formation mechanisms in the plasma deposition of hydrogenated amorphous silicon,” J. Appl. Phys. 59(8), 2998–3001 (1986). 18. R. A. Street, Hydrogenated Amorphous Silicon (Cambridge University, 1991), Chap. 2. 19. U. Palanchoke, V. Jovanov, H. Kurz, P. Obermeyer, H. Stiebig, and D. Knipp, “Plasmonic effects in amorphous silicon thin film solar cells with metal back contacts,” Opt. Express 20(6), 6340–6347 (2012). 20. V. Jovanov, X. Xu, S. Shrestha, M. Schulte, J. Hüpkes, M. Zeman, and D. Knipp, “Influence of interface morphologies on amorphous silicon thin film solar cells prepared on randomly textured substrates,” Sol. Energy Mater. Sol. Cells 112, 182–189 (2013). 21. V. Jovanov, X. Xu, S. Shrestha, M. Schulte, J. Hüpkes, and D. Knipp, “Predicting the interface morphologies of silicon films on arbitrary substrates: application in solar cells,” ACS Appl. Mater. Interfaces 5(15), 7109–7116 (2013). 22. A. Lin and J. Phillips, “Optimization of random diffraction gratings in thin-film solar cells using genetic algorithms,” Sol. Energy Mater. Sol. Cells 92(12), 1689–1696 (2008). 23. R. Dewan, I. Vasilev, V. Jovanov, and D. Knipp, “Optical enhancement and losses of pyramid textured thin-film silicon solar cells,” J. Appl. Phys. 110(1), 013101 (2011). 24. A. Čampa, J. Krč, and M. Topič, “Analysis and optimisation of microcrystalline silicon solar cells with periodic sinusoidal textured interfaces by two-dimensional optical simulations,” J. Appl. Phys. 105(8), 083107 (2009). 25. S. Fahr, T. Kirchartz, C. Rockstuhl, and F. Lederer, “Approaching the Lambertian limit in randomly textured thin-film solar cells,” Opt. Express 19(S4 Suppl 4), A865–A874 (2011). 26. R. Dewan, J. I. Owen, D. Madzharov, V. Jovanov, J. Hüpkes, and D. Knipp, “Analyzing nanotextured transparent conductive oxides for efficient light trapping in silicon thin film solar cells,” Appl. Phys. Lett. 101(10), 103903 (2012). 27. A. Tamang, A. Hongsingthong, P. Sichanugrist, V. Jovanov, M. Konagai, and D. Knipp, “Light-Trapping and Interface Morphologies of Amorphous Silicon Solar Cells on Multiscale Surface TexturedSubstrates,” IEEE J. Photovolatics 4(1), 16–21 (2013). 28. S. Benagli, D. Borrello, E. Vallat-Sauvain, J. Meier, U. Kroll, J. Hoetzel, J. Bailat, J. Steinhauser, M. Marmelo, G. Monteduro, and L. Castens, “High efficiency amorphous silicon devices on LPCVD-ZnO TCO prepared in industrial KAI R&D reactor,” 24th European Photo. Solar Energy Conf., 21–25(2009, September). 29. M. Zeman, R. A. C. M. M. van Swaaij, J. W. Metselaar, and R. E. I. Schropp, “Optical modeling of a-Si:H solar cells with rough interfaces: Effect of back contact and interface roughness,” J. Appl. Phys. 88(11), 6436–6443 (2000). 30. C. Battaglia, C.-M. Hsu, K. Söderström, J. Escarré, F.-J. Haug, M. Charrière, M. Boccard, M. Despeisse, D. T. Alexander, M. Cantoni, Y. Cui, and C. Ballif, “Light trapping in solar cells: can periodic beat random?” ACS Nano 6(3), 2790–2797 (2012). 31. 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).

1. Introduction The realization of core / shell nanowire solar cells is a promising route to enhance the energy conversion efficiency and short circuit current of solar cells utilizing materials that are limited by low charge carrier diffusion lengths [1–8]. The 3D architecture of the solar cells allows for the decoupling of electrical and optical properties, so that high conversion efficiencies can be achieved. Furthermore, the 3D arrangement of the nanowires (NWs) allows for an efficient incoupling and trapping of light in the solar cell structure leading to an enhanced quantum efficiency and short circuit current. Experimentally realized core / shell solar cell based on zinc oxide (ZnO) nanowires in combination with amorphous silicon overcoats exhibit high short circuit currents [1–7], while the photo-induced degradation (Staebler-Wronski effect) of the solar cells is reduced [6,9]. The combination of oxide semiconductor based nanowires with silicon thin film technology is a very promising route, since highly ordered oxide based nanowires can be grown at low temperatures on glass or plastic substrates [10–13]. Different growth processes of ZnO nanowires ranging from hydrothermal to electro-deposition and pulse laser deposition have been demonstrated [1,10–14]. The array of nanowires acts as substrate for the growth of the very thin solar cells. The ZnO nanowires are doped by aluminum or gallium, so that they form an electrical contact to the solar cell [1]. Afterwards, the nanowires are over-coated by the thin amorphous silicon p-i-n diode. The deposition process used to cover the array of nanowires must be conformal to ensure all regions of the nanowire surface are covered with an equal amount of p-i-n layers, which would enable efficient generation of photocurrent across the entire surface area of the nanowires. Hence,


atomic layer deposition and chemical vapor deposition processes are most favorable since they exhibit excellent surface coverage [15–18]. In most of the cases, the amorphous silicon films are prepared by plasma enhanced chemical vapor deposition [17,18]. Although promising experimental results have been demonstrated the optimal core/shell dimensions of the solar cell are unknown. The aim of the manuscript is to derive optimal device dimensions of core/shell solar cells. However, a realistic description of the surface and interface morphology is required to study the optical properties of such nanowire solar cells. We have developed a simple approach to determine realistic interface morphologies of the individual layers of the solar cells. The approach is described in section 2. The simulation results are presented and discussed in section 3 before summarizing the results in section 4. 2. Device structures and interface morphologies In the following, the device model of the core/shell nanowire solar cell is described. The device structure was selected to be consistent with experimentally realized core/shell nanowire solar cells [1–7]. Figure 1 (left) shows the Scanning Electron Microscope (SEM) image of an array of zinc oxide nanowires synthesized from a thin Al-doped seed layer. The nanowires were subsequently coated with a thin amorphous silicon layer. A cross-section of the nanowire solar cell is shown in Fig. 1 (right). The amorphous silicon solar cells exhibited an excellent surface coverage of the ZnO nanowires [1]. The silver back reflector is assumed to be flat to minimize parasitic absorption losses compared to a textured silver back reflector [19]. On top of the silver reflector, a 50 nm thick back TCO ZnO film is defined to act as a seed layer for the growth of the ZnO nanowires. The dimensions of the zinc oxide nanowire array are described by the spacing of the nanowires (period of the unit cell, p), the diameter (d), and height of the nanowire (h). The cross section of the nanowire is assumed to be circular. To provide an analysis of the optics in core/shell nanowire arrays a detailed description of the nanowire geometry is required.

Fig. 1. SEM image of substrate solar cell consisting of ~500 nm ZnO NWs grown on an AlZnO seed layer with a coating of ~100 nm of a-Si:H. The sketch to the right shows the structure of the coated layers.


Fig. 2. Film formation (a) in the direction of substrate normal, (b) by 3D surface growth algorithm and (c) by simple geometrical model. For the film formation the metal back contact is used as substrate for the nominal film thickness of d and ZnO nanowire (NW) grows on transparent conductive oxide.

The optimal device geometry of the nanowires can only be determined by a realistic description of the thin film growth on the ZnO nanowire array. Figure 2 shows interface morphologies of silicon films on ZnO nanowires for different growth models. Figure 2(a) shows the interface morphology of a silicon film in which the silicon film grows in the direction of the substrate normal. Such a growth model is commonly assumed when describing the silicon film growth on textured substrates. However, in the case of the ZnO nanowires array the substrate normal is zero for the side walls of the nanowires so that the side walls are not covered. A realistic description of the silicon film formation is given in Fig. 2(b). In order to provide a realistic description of the interface morphology a 3D surface coverage algorithm was developed. The algorithm assumes that the film grows in the direction of the local surface normal. A comparison of measured and calculated silicon thin film morphologies on textured substrates exhibits a very good agreement so that the algorithm enables a realistic interface description of silicon thin film formation. References 20,21 provide details on the surface growth algorithm. The same film formation mechanism was assumed for the front transparent conductive oxide layer. The calculated interface morphologies were used as input parameter in the optical simulations. For a device structure in which the spacing between the nanowires is larger than twice the thickness of the thin amorphous silicon diode and the length of the nanowire is much larger than the diameter of the nanowire the interface morphology can be approximated by the simple geometrical model as shown in Fig. 2(c). The thin amorphous silicon p-i-n solar cell consists of a 10 nm n-layer, an intrinsic amorphous silicon layer (i-layer) and 10 nm p-layer. Since the solar cell is very thin it can be assumed that the diffusion length of the photo-generated carriers is larger than the solar cell thickness. The thickness of the front TCO (ZnO) layer was 100 nm. 3. Results and discussion 3.1 Electrical versus optical properties Preparing very thin amorphous silicon solar cells on ZnO nanowire allows for decoupling of the electrical and optical properties of the solar cell. Due to the short diffusion length of carriers in nanowire solar cells, their electrical properties are similar to that of a solar cell of flat substrates. The same structure behaves optically like a much thicker solar cell due to efficient light trapping within the nanowire array [6]. To describe the differences in the electrical and optical behavior of the solar cell, the “electrical thickness” and “optical thickness” need to be defined. The electrical thickness, telec, is defined as the maximum distance photo-generated electron/hole pairs have to travel, while the optical or effective thickness of i-layer, topt, is the averaged intrinsic layer thickness of the solar cell structure.


The electrical thickness and optical thickness depend on the interface morphologies and/or silicon film growth [Figs. 2(b) and 3]. The electrical and optical thicknesses were compared to the nominal thickness of the intrinsic layer of the solar cell, which is given by ti. The optical thickness, topt, of the absorber/intrinsic (i) layer of the p-i-n diode was calculated by

topt =

1 p2

p

  0

p

0

t ( x, y )dxdy ,

(1)

where, t(x, y) is the local thickness of the absorber layer (i-layer) and p is the period of the circular based unit cell. The influence of the realistic interface morphologies on the core/shell nanowire solar cells is shown in Fig. 3. The interface morphology of the nanowire solar cell is determined by the diameter of the nanowire, height of the nanowire, period or spacing of the unit cell, and thickness of the silicon film, to [4]. The influence of different dimensions of the ZnO array on the interface morphology is shown in Fig. 3. The diameter (d) and height (h) of the nanowire are 120 nm and 200 nm, respectively, while the thickness of the p-i-n diode or amorphous silicon solar cell is assumed to be to = 120 nm. If the period is smaller than 2 × to + d, the gap between the nanowires is completely filled [Fig. 3(a)] or partly filled [Fig. 3(b)]. For such structures, the electrical thickness increases from to to h + to. Since the height of the nanowires is usually distinctly larger than the nominal thickness of the solar cells, creating a maximum charge carrier transport distance that may exceed the carrier diffusion length. For the selected device geometries in Fig. 3, such an effect occurs for nanowire spacing smaller than 500 nm. As a consequence, we will not discuss the optics of these structures. We focused on wire spacings of unit cell period (p>500 nm) for which the electrical thickness of the solar cell is equal to the nominal thickness (to) of the amorphous silicon solar cell. Examples of such structures are shown in Figs. 3(c) and 3(d). The surface morphology for large nanowire spacing is shown in Fig. 3(d). Since the period is distinctly increased, a large fraction of the structure is flat. With increasing period, the optical thickness converges towards the nominal thickness of the solar cell [Fig. 3(d)]. The optical thickness for the structure shown in Fig. 3(c) is distinctly larger compared to structure shown in Fig. 3(d).

Fig. 3. Influence of periods (p, 200 nm (a), 360 nm (b), 600 nm (c) and 840 nm (d)) on the interface morphologies, optical thickness and electrical thickness (telec) at nominal p-i-n diode thickness (to, 120 nm), nanowire height (h, 200 nm), diameter of nanowire (d, 120 nm).


In this study, we focused on structures which exhibit an increased optical thickness while the electrical thickness is equal to the nominal thickness. Hence, we focused our investigation on structures where the p-i-n diode is wrapped around the individual nanowires. Furthermore, it is assumed that the height of the nanowire is larger than the nominal thickness of the p-i-n diode. The structures where the height of the wire is smaller than the nominal thickness of the p-i-n diode correspond to solar cells with textured contact layers [22–27]. 3.2 Optics of core/shell nanowire solar cells The optical wave propagation within the core/shell nanowire solar cells was determined by Finite Difference Time Domain (FDTD) simulations [19,23,26,27]. The Maxwell’s equations were rigorously solved in 3D and the electric fields within the solar cells and absorption of light in individual layers of the solar cell were calculated. To compare the different solar cell structures, the power loss profiles, the quantum efficiency (QE) and the short circuit current were calculated. Firstly, the electric field was calculated in the 3D FDTD simulations for circularly polarized incident light. Secondly, the time average power loss within the amorphous silicon solar cells was calculated from the electric fields for the incident light with amplitude of 1 V/m. Afterwards, the quantum efficiency was calculated and is defined as the ratio of the power absorbed in the i-layer with respect to the total power incident in the unit cell. In the simulations, it was assumed that all photo-generated charge carriers are collected, i.e. internal quantum efficiency was assumed to be 100%. Finally, the short circuit current was calculated from the quantum efficiency of the i-layer. References 19,25 give details on the calculation of these parameters. The calculated power loss profile for a solar cell [Fig. 4(a)] with an i-layer thickness of 100 nm is shown in Figs. 4(b)-4(d). The power loss profile of the solar cell was calculated in 3D and Fig. 4(a) exhibits the power loss profile in the center of the unit cell. The power loss profiles for Figs. 4(b)-4(d) are shown for an incident wavelength of 360 nm, 560 nm and 620 nm, respectively. For a shorter wavelength of 360 nm, the absorption of light mainly occurs in the front TCO due to the high absorption coefficient of ZnO close to the band edge. Longer wavelengths (560 nm and 620 nm) is scattered and diffracted by the nanowire array to enhance absorption of light in absorber layer of the solar cell. The influence of the ZnO nanowire diameter on the quantum efficiency is shown in Fig. 5(a) for a solar cell with a period of 600 nm, a height of 200 nm and i-layer thickness of 100 nm. The diameter of the nanowire was varied from 60 nm to 150 nm. As a reference the QE of a solar cell on the flat substrate is shown in Figs. 5 and 6(b). A comparison of different simulations shows that the diameter of the ZnO nanowire has only a minor influence on the quantum efficiency. The incoupling of light in the array of the solar cell is not affected. Furthermore, the scattering of longer wavelengths light remains almost unchanged. The short circuit current of the solar cell increases from 15.0 mA/cm2 to 15.6 mA/cm2 with increasing diameter. The reference solar cell on the flat substrate exhibits a short circuit current of 7.6 mA/cm2. The optical thickness of i-layer was calculated to be 126 nm for a 50 nm ZnO nanowire diameter and 140 nm for a nanowire diameter of 150 nm for the nominal i-layer thickness of 100 nm. The influence of the nanowire spacing or unit cell period on the quantum efficiency is shown in Fig. 5(b). The ZnO nanowire height was assumed to be 200 nm, while the period was varied from 600 nm to 800 nm. The quantum efficiency was calculated for a nominal ilayer thickness of 100 nm. The optical thickness of i-layer decreases from 136 nm (for a 600 nm period) to 120 nm (for 800 nm period) with increasing period. The increasing period results in a drop of the short circuit current from 15.6 mA/cm2 to 12.3 mA/cm2. For smaller periods, the gap between the nanowires is filled and the electrical thickness exceeds 100 nm nominal i-layer thickness. For larger periods, the optical thickness and the short circuit current steadily decrease. Furthermore, light is diffracted in smaller angles leading to a drop in the short circuit current. For very large periods the quantum efficiencies and short circuit currents converge towards that of the flat substrate and as a result, the nanowire solar cells with larger nanowire spacing behave like a flat solar cell.


Fig. 4. (a) Schematic cross section of the unit cell of an amorphous silicon thin film solar cell based on ZnO nanowire at 400 nm nanowire height, 600 nm unit cell period, 100 nm i-layer thickness and 120 nm nanowire diameter. Simulated power loss profiles under monochromatic illumination at wavelength of (b) 360 nm, (c) 560 nm and (d) 620 nm for the structure (a).

The influence of i-layer thickness on the quantum efficiency is shown in Fig. 6(a). The quantum efficiency increases with increasing thickness of the i-layer. The short circuit currents increase from 13.8 mA/cm2 for an i-layer thickness of 50 nm to 15.8 mA/cm2 for a thickness of 150 nm. The height, diameter and period of the ZnO nanowire were 200 nm, 120 nm and 600 nm, respectively. The optical thickness of nominal i-layer thickness of 50 nm is estimated to be 65 nm. For the nominal thickness of 150 nm the approximated optical thickness of 208 nm was calculated. The increased optical thickness leads to the higher quantum efficiency for wavelengths larger than 520 nm. The nominal thickness of the i-layer also influences the front contact morphology and the incoupling of the shorter wavelength light (360-520 nm). The best light incoupling is achieved for the i-layer thickness of 125 nm. The influence of the nanowire height on the quantum efficiency is shown in Fig. 6(b). The height of the wires was varied from 100 nm to 400 nm. For shorter wavelengths (