Broadband absorption enhancement in a-Si:H film ... - OSA Publishing

Broadband absorption enhancement in a-Si:H thin-film solar cells sandwiched by pyramidal nanostructured arrays Chuanhao Li,1,2 Liangping Xia,2 Hongtao Gao,2 Ruiying Shi,1,4Chen Sun,1,2 Haofei Shi,3 and Chunlei Du2,3,* 1 Physics Department, Sichuan University, Chengdu 610064, China Institute of Optics and Electronics, Chinese Academy of Sciences, P.O. Box 350, Chengdu 610209, China 3 Chongqing institute of green and intelligent technology, Chinese Academy of Sciences, Chongqing, 401122, China 4 [email protected] * [email protected] 2

Abstract: A new thin-film solar cell structure with a broadband absorption enhancement is proposed. The active a-Si:H film is sandwiched by two periodic pyramidal structured layers. The upper dielectric pyramidal layer acts as matching impedance by gradual change of the effective refractive index to enhance the absorption of the active layer in the short wavelength range. The lower metallic pyramidal layer traps light by the excitation of Fabry–Perot (FP) resonance, waveguide (WG) resonance and surface plasmon (SP) mode to enhance the absorption in the long wavelength range. With the cooperation of the two functional layers, a broadband absorption enhancement is realized. The structure parameters are designed by the cavity resonance theory, which shows that the results are accordant with the finite-difference time-domain (FDTD) simulation. By optimizing, the absorption of the sandwich structure is enhanced up to 48% under AM1.5G illumination in the 350–900 nm wavelength range compared to that of bare thin-film solar cells. ©2012 Optical Society of America OCIS codes: (230.7370) Waveguides; (250.5403) Plasmonics; (040.5350) Photovoltaic.

References and links 1.

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). 2. W. Wang, S. Wu, K. Reinhardt, Y. Lu, and S. Chen, “Broadband light absorption enhancement in thin-film silicon solar cells,” Nano Lett. 10(6), 2012–2018 (2010). 3. J. N. Munday and H. A. Atwater, “Large integrated absorption enhancement in plasmonic solar cells by combining metallic gratings and antireflection coatings,” Nano Lett. 10, 000-000 (2010). 4. S. Chhajed, M. F. Schubert, J. K. Kim, and E. F. Schubert, “Nanostructured multilayer graded-index antireflection coatings for Si solar cell with broadband and omnidirectional characteristics,” Appl. Phys. Lett. 93(25), 251108 (2008). 5. B. M. Kayes, H. A. Atwater, and N. S. Lewis, “Comparison of the device physics principles of planar and radial pn junction nanorod solar cells,” J. Appl. Phys. 97(11), 114302 (2005). 6. 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). 7. F. C. Chen, J. L. Wu, C. L. Lee, Y. Hong, C. H. Kuo, and M. H. Huang, “Plasmonic-enhanced polymer photovoltaic devices incorporating solution-processable metal nanoparticles,” Appl. Phys. Lett. 95(1), 013305 (2009). 8. M. Schmid, R. Klenk, M. Ch. Lux-Steiner, M. Topic, and J. Krc, “Modeling plasmonic scattering combined with thin-film optics,” Nanotechnology 22(2), 025204 (2011). 9. W. S. Koh, Y. Akimov, Y. Li, M. S. Soh, W. P. Goh, and H. S. Chu, “Optical Enhancement with Plasmonic Nanoparticles in Organic Bulk-Heterojunction Solar Cells,” Optical Society of America (2010). 10. L. Xia, H. Gao, H. Shi, X. Dong, and C. Du, “A Wideband Absorption Enhancement for P3HT: PCBM Addressing by Silver Nanosphere Array,” J. Comput. Theor. Nanosci. 8(1), 27–30 (2011). 11. J. Y. Lee and P. Peumans, “The origin of enhanced optical absorption in solar cells with metal nanoparticles embedded in the active layer,” Opt. Express 18(10), 10078–10087 (2010).

#166639 - $15.00 USD Received 13 Apr 2012; revised 31 May 2012; accepted 12 Jun 2012; published 9 Jul 2012 (C) 2012 OSA 10 September 2012 / Vol. 20, No. S5 / OPTICS EXPRESS A589

12. L. Qiao, D. Wang, L. Zuo, Y. Ye, J. Qian, H. Chen, and S. He, “Localized surface plasmon resonance enhanced organic solar cell with gold nanospheres,” Appl. Energy 88(3), 848–852 (2011). 13. N. N. Lal, B. F. Soares, J. K. Sinha, F. Huang, S. Mahajan, P. N. Bartlett, N. C. Greenham, and J. J. Baumberg, “Enhancing solar cells with localized plasmons in nanovoids,” Opt. Express 19(12), 11256–11263 (2011). 14. C. Min, J. Li, G. Veronis, J. Y. Lee, S. Fan, and P. Peumans, “Enhancement of optical absorption in thin-film organic solar cells through the excitation of plasmonic modes in metallic gratings,” Appl. Phys. Lett. 96(13), 133302 (2010). 15. R. A. Pala, J. White, E. Barnard, J. Liu, and M. L. Brongersma, “Design of Plasmonic Thin film Solar Cells with Broadband Absorption Enhancements,” Adv. Mater. (Deerfield Beach Fla.) 21(34), 3504–3509 (2009). 16. M. A. Sefunc, A. K. Okyay, and H. V. Demir, “Plasmonic backcontact grating for P3HT: PCBM organic solar cells enabling strong optical absorption increased in all polarizations,” Opt. Express 19(15), 14200–14209 (2011). 17. H. Shen, P. Bienstman, and B. Maes, “Plasmonic absorption enhancement in organic solar cells with thin active layers,” J. Appl. Phys. 106(7), 073109 (2009). 18. V. E. Ferry, L. A. Sweatlock, D. Pacifici, and H. A. Atwater, “Plasmonic nanostructure design for efficient light coupling into solar cells,” Nano Lett. 8(12), 4391–4397 (2008). 19. V. E. Ferry, M. A. Verschuuren, H. B. T. Li, R. E. I. Schropp, H. A. Atwater, and A. Polman, “Improved red-response in thin film a-Si: H solar cells with soft-imprinted plasmonic back reflectors,” Appl. Phys. Lett. 95(18), 183503 (2009). 20. A. Abass, H. Shen, P. Bienstman, and B. Maes, “Angle insensitive enhancement of organic solar cells using metallic gratings,” J. Appl. Phys. 109(2), 023111 (2011). 21. J. Zhu, C. M. Hsu, Z. Yu, S. Fan, and Y. Cui, “Nanodome solar cells with efficient light management and self-cleaning,” Nano Lett. 10(6), 1979–1984 (2010). 22. E. D. Palik and G. Ghosh, Handbook of optical constants of solids (Academic Press, 1985). 23. T. I. Kim, J. H. Kim, S. J. Son, and S. M. Seo, “Gold nanocones fabricated by nanotransfer printing and their application for field emission,” Nanotechnology 19(29), 295302 (2008). 24. C. M. Hsu, S. T. Connor, M. X. Tang, and Y. Cui, “Wafer-scale silicon nanopillars and nanocones by Langmuir-Blodgett assembly and etching,” Appl. Phys. Lett. 93(13), 133109 (2008).

1. Introduction Thin-film solar cells of hydrogenated amorphous silicon (a-Si:H) possess the characteristic advantages of non-toxicity, abundance and mature processing technology [1, 2]. The conversion efficiency of thin-film cells is directly related to optical absorption of the active layer and transfers of the carriers. The carrier transport of a-Si:H is poor, especially the diffusion length of the short minority carrier is only about 300 nm. In order to collect photocarriers adequately, the thickness of the active layer should be several times shorter than the diffusion length. However, the ultrathin active layer leads to a poor absorption. To enhance optical absorption in thin-film cells, sufficient researches have been explored recently. These achieved absorption enhancements are attributed to the introduction of certain nanostructures that contain the multilayer antireflection (AR) coatings [3, 4], dielectric nanopillars [5, 6] and nanotips [1], metallic nanoparticles [7–12], nanovoids [13], nanogratings [14–19], and nanotips [20, 21]. Among these structures, the nanotip periodic arrays are found to perform a comparable absorption enhancement of both the improved efficiency and the broadened band in thin-film cells. The dielectric nanotips at the top interface of cells are commonly utilized to match the surface impedance, and hence the surface reflection loss is reduced. The metallic nanotips introduced in thin-film cells commonly play the role of trapping light by exciting surface plasmon (SP) resonance to localize incident fields into subwavelength regions, and hence the reflection from the back contact is restrained. However, a single enhanced mechanism commonly works in a narrow wavelength range. And a broadband absorption enhancement of the nanotip structure results from confused physical mechanisms, which is not clear enough at present. What’s more, the enhancement theory model of the nanotip structure is absence of the structure parameter design. In this study, a new solar cell structure with the active a-Si:H film sandwiched by two periodic pyramidal nanostructured arrays is proposed. The sandwich cell structure is designed into two individuals, the upper impedance-matching structure and the lower light-trapping structure. The upper impedance-matching structure with an ITO/a-Si:H pyramidal array on the top is designed based on the impedance match. The lower light-trapping structure with an Ag pyramidal array on the back contact is designed as a cavity resonance model. The cavity resonance model which excites the FP resonance and WG resonance is built by optimizing the structure parameters. Finally, with the


cooperation of the two functional individuals, a broadband absorption enhancement is achieved in the sandwich structure. The overall absorption is greatly enhanced up to 48% under AM1.5 illumination, which is remarkable compared to the common proposed a-Si:H thin-film cells (within the range of 10~30%). As the pyramidal structures in the cell structure are 2D periodic distributed arrays, the broadband absorption enhancement is independent with the incidence polarization. 2. Theory design and analyze As shown in Fig. 1, to make clear the absorption enhancement mechanisms and design the structure parameters, the proposed sandwich thin-film solar cell structure is divided into two parts, the impedance-matching structure and the light-trapping structure. For the impedance-matching structure, the flat a-Si:H layer is placed at the bottom of a multiple pyramidal array — an ITO pyramidal array as the transparent anode and an a-Si:H pyramidal array surrounded by ITO, in addition the ITO and the a-Si:H pyramidal array share the same size. For the light-trapping structure, the flat a-Si:H layer is placed at the top of an Ag pyramidal array, and the space between these Ag pyramids is filled with ITO in order to ensure carriers to be collected adequately. By combining the two individuals, the proposed sandwich cell structure is obtained as shown in the same figure. The surface reflection loss and the light-trapping mechanisms are analyzed in the two individuals independently. In either of the two individuals, the mechanisms are theoretically explained, and the parameters are logically designed. Then the commercial software Lumerical Solutions based on the 3D finite-difference time-domain (FDTD) method is used to demonstrate these designed results. Finally, the optimum structure parameters of the two individuals are selected as the parameters of the sandwich pyramid structure to achieve a broadband absorption enhancement.

Fig. 1. Schematic of the sandwich pyramid structure divided into two parts. The first part is the impedance-matching structure and the second part is the light-trapping structure.

Considering the bandgap of a-Si:H and AM1.5G spectrum distribution, the 350-900 nm wavelength range is selected as the incidence. The dispersive dielectric constants of materials are referenced from Palik [22], and only normal incidence is considered. In addition, the EMS memory of the workstation is up to 192 GB to match the enormous calculation of the 3D simulations. 2.1 Impedance-matching structure The surface reflection is caused when the refractive index changes suddenly at the interface, and the reflection efficiency is related to the difference of the indices. The index of ITO approximates to 1.8 and the index of a-Si:H shown in Fig. 2(a) possesses a large real part in the incidence wavelength range, indicating that the sudden change of the index at the interface between a-Si:H and ITO is larger than that at the interface between air and ITO in a-Si:H thin-film cells. Hence in this study, besides the ITO pyramidal array is introduced to match the impedance at the air/ITO interface by gradual change of the effective index, the a-Si:H layer is also designed to be a pyramidal array to match the impedance at the ITO/a-Si:H interface. In addition, to eliminate the sudden change of the index at the interfaces completely, the filling fraction of the ITO/a-Si:H array is set to be 1. To investigate the impedance match when the ITO/a-Si:H pyramidal array is introduced, the effective index is calculated by averaging the refractive indices of air, ITO and a-Si:H by volume [1] in the impedance-matching structure with the formula for the 2D pyramidal arrays: #166639 - $15.00 USD Received 13 Apr 2012; revised 31 May 2012; accepted 12 Jun 2012; published 9 Jul 2012 (C) 2012 OSA 10 September 2012 / Vol. 20, No. S5 / OPTICS EXPRESS A591

N eff   FFi  ni

(1)

i

where FFi and ni are the filling fraction and the refractive index of the ith layer respectively. The calculated result in Fig. 2(b) shows that the effective index gradually changes from the top of the cell to the active layer, indicating that the impedance match from the air to the active layer is obtained. Therefore most of the incidence enters the active layer through the top surface, namely the transmission through the top surface approaches to 1. Now the absorption enhancement of the impedance-matching structure is discussed. The imaginary part of the a-Si:H index shown in Fig. 2(a) is very large in the short wavelength range but rapidly decreases in the long wavelength range. Consequently, as the incidence almost enters the active layer due to the impedance match, the extra transmission through the top surface in the short wavelength range is effectively absorbed in the active layer, but that in the long wavelength range is mainly reflected back from the back contact and makes less contribution to the absorption enhancement. As a result, the aim of introducing the impedance-matching structure is at enhancing the absorption of the active a-Si:H layer in the short wavelength range. When the height of these pyramids is fixed at h1’ = h2’ = 100 nm, the simulated absorption spectra with a series of periods are shown in Fig. 2(c). The absorption is obviously enhanced in the short wavelength range compared to that of bare a-Si:H thin-film cells. With a larger period, the absorption spectrum causes a red shift, but the absorption in the short wavelength range gets worse. To realize a remarkable absorption enhancement in the short wavelength range and a relatively widened absorption band, the period is selected to be 200 nm.

Fig. 2. (a) Real and imaginary part of the a-Si:H refractive index. (b) Effective refractive index profiles along the perpendicular direction in the impedance-matching structure. Inset: cross-sectional view of the impedance-matching structure. (c) Absorption spectra with a series of P1’/P2’ in the impedance-matching structure compared to bare a-Si:H thin-film cells.

2.2 Light-trapping structure 2.2.1 Theory and principle As the impedance-matching structure has increased the absorption in the short wavelength range, the light-trapping structure is expected to work in the long wavelength range. Here the physical mechanisms of trapping light are discussed and the structure parameters are designed theoretically in the light-trapping structure. The cavity resonance can be excited in the active a-Si:H layer of the light-trapping structure when the reflection from the back contact and the reflection from the ITO/a-Si:H interface satisfy the phase matching condition. The phase matching condition is satisfied in the active layer:

2k  h 3’ + 1 + 2 = 2m

(2)

where k is the wave vector in the a-Si:H film along the perpendicular direction, h3’ is the thickness of the a-Si:H film, 1, 2 are the phase changes at the top interface and bottom interface of the a-Si:H film respectively, and m is the resonance mode number. Since the layer below the a-Si:H film is composed of the Ag pyramidal array and the filled ITO, the effective index of this layer is estimated to be neff = nAg·FFAg + nITO·FFITO. So the phase


changes 1 and 2 are: 1 = arctan (Im(r21)/Re(r21)), where r21 = (a-Si/ka-Si:H-ITO/kITO)/(a-Si/ka-Si:H + ITO/kITO), ka-Si:H = na-Si:Hk0, kITO = nITOk0, k0 is wave vector at vacuum; 2 = arctan (Im(r23)/Re(r23)), where r23 = (a-Si/ka-Si:H-eff/keff)/(a-Si/ka-Si:H + eff/keff), keff = neff k0, eff = neff2. When the layer below the a-Si:H layer acts as a pure reflector, the FP resonance is excited, and the incidence transmits only along the perpendicular direction, hence k = ka-Si:H. When an extra transverse momentum is supplied by the 2D Ag pyramidal array, the WG resonance is excited, but the k is converted, namely k2 + k∥2 = ka-Si:H2. The transverse wave vector k∥ satisfies the relationship:

k //  i 2  j 2

2 P4'

(3)

where i and j are the orders in the x and y directions, P4’ is the period of the Ag pyramidal array. 2.2.2 Calculation and analysis The FP mode is calculated by Eq. (2) and the result is shown in Fig. 3(a). The circle marked blue line stands for the relationship between the resonance wavelength and the thickness of the a-Si:H film, which reveals that two orders of the FP mode (m = 1, 2) is excited when the thickness h3’ ranges from 50 nm to 130 nm. The first order of the FP resonance is excited when h3’ is smaller than 110 nm, accordingly m = 1; the second order is excited when h3’ is larger than 110 nm, accordingly m = 2. To demonstrate the theory result, the 3D FDTD simulation is carried out, and the simulated absorption spectra is shown in the same figure. The solid blue line stands for the FP resonance, which accords well with the theory result. In addition, the FP mode is little related to the structure parameters of the Ag pyramidal array according to Eq. (2), hence the resonance wavelength is hardly changed when the period of the Ag array changes as the simulated solid blue line shown in Fig. 3(b). The WG mode is also analyzed. Based on Eq. (2) and (3), the theory result is calculated and shown as the square marked dark line in Fig. 3(a) and 3(b). The simulated WG mode is marked by the solid dark line. In Fig. 3(a), when the a-Si:H film is thinner than 100 nm, the simulation accords well with the theory. The transverse momentum vector supplied by the Ag pyramidal array mainly exists near the Ag structure. It will get weaker if the distance from the Ag pyramidal array gets farther. By this reason, when the active a-Si:H layer gets thicker, the WG resonance gets weaker and gradually overlapped with the FP mode. In Fig. 3(b), it shows that the simulated result and the theory are remarkably accorded, and the resonance wavelength causes a red shift as the period of the Ag pyramidal array increases. Considering the absorption band and efficiency, the thickness of the a-Si:H film h3’ = 83.33 nm and the period of the Ag pyramidal array P4’ = 310 nm are selected based on the absorption spectra in Fig. 3(a) and 3(b). In the optimized condition, the absorption spectrum of the light-trapping structure is shown in Fig. 3(c). By contrast with bare a-Si:H thin-film cells, the absorption is obviously enhanced in the long wavelength range. In addition, there are four peaks as signed in the spectrum. The FP resonance leads to the absorption peak A, and the peak B stands for the WG resonance.


Fig. 3. (a) Absorption spectra with varying the thickness of the a-Si:H film and illumination conditions in the light-trapping structure when W4’ = P4’ = 310 nm and h4’ = 120 nm are fixed. (b) Absorption spectra with varying P4’ and illumination conditions in the light-trapping structure when W4’ = P4’, h3’ = 83.33 nm and the shape of the Ag pyramids h4’/P4’ = 12/31 are fixed. (c) Absorption spectrum of the light-trapping structure with the optimum structure parameters (red line) compared to bare a-Si:H thin-film cells (black line). Inset: cross-sectional view of the light-trapping structure. (d) Electric field distribution on a logarithmic scale in the light-trapping structure at the absorption peak C.

In order to clear the absorption peak C in the long wavelength range, the corresponding electric field intensity distribution is simulated as shown in Fig. 3(d). The light field is strongly localized on the surface of the Ag pyramids, which means the excitation of the SP mode. Further, the electric vectors that reflect the oscillation of the electric field is calculated and described by the blue arrows. The oscillation contains two parts: the oscillation at the bottom of the Ag pyramids is mainly located in the lower ITO material, which has no contribution to the absorption enhancement of the active a-Si:H layer; the oscillation at the top of the Ag pyramids penetrates into the a-Si:H film effectively, which leads to the absorption enhancement. In addition, the field distribution shows that there is a depth of the localized field penetrating into the a-Si:H film. When the thickness of the a-Si:H film is not thick enough, the localized field would penetrate into the transparent electrode ITO, and then the absorption enhancement of the a-Si:H layer is low. It accords with the absorption peaks marked by the lower white line shown in Fig. 3(a). Hence in order to obtain a large absorption enhancement, the a-Si:H film should be thick enough. Through calculation the absorption peak D is due to another FP resonance in the cavity formed by the combination of the a-Si:H layer and the transparent electrode ITO layer. Due to the index of the a-Si:H is much larger than that of ITO, the resonance is mainly located in the active layer, and accordingly the absorption of the active layer is enhanced. 3. Results The impedance-matching structure and the light-trapping structure have been independently designed and the enhancement mechanisms are analyzed. When the two individuals are combined into the sandwich pyramid structure, the structure parameters of the sandwich pyramid structure are chosen to be the same as those optimized in the two individuals. The thickness of the flat active layer in the sandwich pyramid structure is set to be h3 = h3’- h2’/3 = 50 nm based on the equivalent principle, where h3’ = 83.33 nm is the optimized thickness of the a-Si:H film in the light-trapping structure and h2’ = 100 nm is the #166639 - $15.00 USD Received 13 Apr 2012; revised 31 May 2012; accepted 12 Jun 2012; published 9 Jul 2012 (C) 2012 OSA 10 September 2012 / Vol. 20, No. S5 / OPTICS EXPRESS A594

height of the a-Si:H pyramids in the impedance-matching structure. The absorption enhancement is defined as:

 Enh  

900 nm

350 nm 900 nm 350 nm

PAM 1.5( )A 2( )d  PAM 1.5( )A 1( )d 

(4)

where PAM1.5() is the ratio of the incidence at the wavelength of  under AM1.5G illumination. A2() and A1() are the absorption efficiency in the sandwich pyramid solar cell and contrast bare a-Si:H thin-film cells at the wavelength of , respectively. With the structure parameters designed, the absorption spectrum of the 3D sandwich pyramid structure shown in Fig. 4(a) is simulated, and the corresponding result is shown in Fig. 4(b). The solid blue curve shows that the absorption spectrum of the sandwich pyramid structure approximatively combines the enhanced bands of the impedance-matching structure and the light-trapping structure as expected. As a result, the absorption band of the sandwich pyramid structure is widely extended. Calculated by Eq. (4), the absorption enhancement is up to 48%. In addition, fabrication of such periodic pyramidal nanostructured arrays is technically feasible. One method to fabricate dielectric pyramidal arrays is a combination of Electron-Beam Lithography and Nanoimprint Lithography [23]. With the mold fabricated by electron-beam lithography, the pyramidal array can be duplicated by a conventional nanoimprint process. Another method is a combination of Nanosphere Lithography and Reactive Ion Etching [24]. Based on the dielectric pyramidal array, the effective metallic pyramidal array can be effectively formed by depositing a thin-film layer of metal on the surface of the dielectric pyramidal array.

Fig. 4. (a) A 3D conceptual schematic of the sandwich pyramid structure. Inset: cross-sectional view of the sandwich pyramid structure. (b) Absorption spectrum of the sandwich pyramid structure, the impedance-matching structure and the light-trapping structure compared to bare a-Si:H thin-film cells.

4. Conclusion Generally, the solar cell structure with the a-Si:H film sandwiched by two pyramidal nanostructured arrays is proposed and the corresponding enhancement mechanisms are analyzed. By dividing the sandwich pyramid structure into two independent individuals — the impedance-matching structure and the light-trapping structure, the absorption mechanisms are rigorously discussed in theory and the parameters are logically designed. The introduction of the upper dielectric pyramidal array restrains the surface reflection loss based on the impedance match and leads to the absorption enhancement in the short wavelength range, while the introduction of the lower Ag pyramidal array restrains the reflection from the back contact by the excitation of the FP resonance, the WG resonance, the SP mode and leads to the absorption enhancement in the long wavelength range. With the combination of the two function layers in the a-Si:H thin-film cell, a broadband absorption enhancement is achieved in the sandwich pyramid structure and the overall absorption is greatly enhanced up to 48% under AM1.5 illumination.


Acknowledgment This work was supported by the Chinese Nature Science Grant Nos. 11074251, 11174281 and 91123032. Chongqing Science & Technology Commission Distinguished Young Scholars Program (cstc2012jjjq002).
