Effective light trapping by hybrid nanostructure for crystalline silicon solar cells

Solar Energy Materials & Solar Cells 140 (2015) 180–186

Contents lists available at ScienceDirect

Solar Energy Materials & Solar Cells journal homepage: www.elsevier.com/locate/solmat

Effective light trapping by hybrid nanostructure for crystalline silicon solar cells Yahui Liu a, Wei Zi a, Shengzhong (Frank) Liu a,b,n, Baojie Yan a,b a Key Laboratory of Applied Surface and Colloid Chemistry, National Ministry of Education, Institute for Advanced Energy Materials, School of Materials Science and Engineering, Shaanxi Normal University, Xi'an 710119, China b Dalian Institute of Chemical Physics, Dalian National Laboratory for Clean Energy, Chinese Academy of Sciences, Dalian 116023, China

art ic l e i nf o

a b s t r a c t

Article history: Received 26 December 2014 Received in revised form 30 March 2015 Accepted 7 April 2015 Available online 29 April 2015

Ag nanoparticles (NPs), as etching catalyst, are applied onto the pyramid textured surface of the semifinished single crystalline silicon solar cell by spraying a solution containing AgNO3 and sodium citrate. Upon chemical etching, nanopits are formed with the Ag NPs staying inside of the nanopits. Complete solar cells with and without the Ag-assisted etching are fabricated to study the effects of the hybrid nanostructure. It is found that the optical reflection is effectively reduced by the nanostructure and the incident light is harvested more effectively for enhanced external quantum efficiency (EQE) by as much as 17% for the cells before the SiNx anti-reflection coating, and the EQE can be further enhanced by applying an electric bias 1 V during the EQE measurement. It is believed that when the solar cell with the Ag NPs is illuminated, the local surface plasmon resonance (LSPR) is induced at the specific wavelength regions around 405 and 810 nm. For ultraviolet light, the LSPR effect dominates the increases of the EQE while the effective medium effect is also believed to be responsible for the effective light trapping. & 2015 Elsevier B.V. All rights reserved.

Keywords: Silicon Solar cell Light trapping Ag nanoparticle Surface plasmon resonance Quantum efficiency

1. Introduction Among various semiconductor materials, silicon is the material of choice for photovoltaic (PV) applications due to its low cost, abundance on earth, non-toxicity, long-term stability, and well-established technology. The silicon based solar cells account for 90% in overall PV installations. As one of the primary costs for silicon photovoltaic cells stems from the materials consumption, e.g., Si wafer alone represents almost a half of the module costs [1–3]. The most effective cost-cutting measure is by increasing solar cell efficiency. Light trapping by confining the light within the active semiconductor layers to promote the absorption, may accomplish both efficiency improvement and cost cutting, it is therefore a favorable and effective strategy. Nanoparticles and nanostructures have been successfully used for the light-trapping applications [4,5]. In fact, a facile method using metal-assisted chemical etching has been developed to prepare large-area silicon nanostructure based on metal-catalyzed site-specific anisotropic Si etching mechanism [6–9]. It is simple, low-cost, and large-area compatible for generating Si nanostructures, including nanowires, nanoholes and nanopillars [6,7,9–11]. A few novel methods have been proposed to enhance the light trapping effect. For example, the introduction of local surface n

Corresponding author. Tel.: þ 86 029 8153 0785. E-mail address: [email protected] (S. (Frank) Liu).

http://dx.doi.org/10.1016/j.solmat.2015.04.019 0927-0248/& 2015 Elsevier B.V. All rights reserved.

plasmon resonance (LSPR) effect induced by metallic nanostructures has been widely recognized to be able to boost the light absorption of solar cells. The LSPR effect of noble metal nanoparticles (NPs) has attracted considerable attention in PV research due to its special characteristics: when the metallic NPs are excited by light at specific wavelengths, the collective movement of conduction electrons builds up polarization charges on the particle surface. The key phenomena lie in two main aspects. Firstly, the LSPR can excite electron–hole pairs or increase the rate of electron– hole formation in the semiconductor by transferring the plasmonic energy from the metal to the semiconductor [12–15]. Secondly, the LSPR boosts the light absorption of solar cells due to the far-field scattering from the metal NPs [16,17]. Both the shape and size of the metal nanoparticles are found to be key factors in coupling between the metal particles and the dielectric, and thereby the overall enhancement in the optical absorption efficiency [18–21]. The LSPR effect is currently being exploited in various applications including molecular sensing [22–24], light focusing [25], near-field optical microscopy [26], and subwavelength photonics [27]. Various metal nanoparticles are found to support the LSPR effect. In particular, Ag, Au, Cu, and Al nanoparticles have been explored to enhance the solar cell efficiency [28–32]. In this work, we found that a hybrid nanostructure that combines advantages of nanopits and Ag NPs effectively enhances the light absorption and thereby quantum efficiency (QE) of the solar cells. The Ag NPs were prepared as the catalyst to etch the surface

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

of single crystalline silicon surface, resulting in a hybrid nanostructure including silicon nanopits and Ag NPs within the nanopits. It is found that the solar cells with the hybrid nanostructure have significantly enhanced short-circuit current density (JSC) by 17%. At the same time, the QE response clearly supports the interpretation that the increased current is from the LSPR effect.

181

samples were measured by using a QE measurement system (QTEST STATION 500TI, CROWNTECH) equipped with integrating sphere. The QE measurement system was also used to record incident photon current conversion efficiency spectra before and after the chemical etching.

3. Results and discussion 2. Experimental Semi-finished single crystalline silicon solar cells, acquired from a production line after the alkali anisotropic etching but prior to the SiNx anti-reflection (AR) coating, are used for this study. The Ag NPs were prepared by reducing AgNO3 (0.01 M) using sodium citrate (7  10  3 M) at about 100 °C. To apply the Ag NPs onto the silicon solar cell surface in a well-controlled fashion, a single crystalline silicon solar cell was heated to slightly above 100 °C first, the preformed Ag NPs colloid suspension was then spraydeposited onto the preheated solar cell surface using a pressure regulated air brush spray system. As the water solvent evaporates as it reaches the solar cell surface maintained at above 100 °C, the Ag NPs are “fixed” onto the cell surface. A chemical etching process, using DI water–HF–H2O2 (volume ratio¼30:4:1, CHF ¼40%, CH2 O2 ¼ 30%), was used to form silicon nanopits on the wafer surface with a diameter about 30 nm on the pyramid structure. The Ag NPs can be easily removed by reacting with nitric acid for 30 min. Both the sample surface and its cross-section were examined using a scanning electron microscope (SEM, S4800, HITACHI). A Kratos Axis Ultra-XPS system was used with a monochromatic Al–Kα source. The AR properties of structured

It is known that the Ag NPs in colloid suspension exhibit characteristic peaks in the UV–vis absorption spectrum. Fig. 1 (a) shows the UV–vis absorption spectra of Ag NPs prepared with different diameters at 25, 35, 45, and 60 nm. It is clear that the Ag NPs show distinguished LSPR peaks at 404, 412, 425, and 433 nm. Particularly, the Ag NP diameter has a well-established relationship with its LSPR peak position. Fig. 1(b) shows XPS of Ag NPs deposited on the silicon wafer, with two peaks at 368 eV and 374 eV attributed to atomic Ag. Fig. 1(c) shows a SEM image of the 25 nm Ag NPs deposited on a silicon wafer surface, and the measurement is consistent with what calculated from the UV–vis [33,34]. Fig. 1(d) shows the SEM image of the cross-section of the Si substrates after etching. It is obvious that in addition to an array of pyramidal structures about several micrometers, there are smaller nanopits with an average diameter  30 nm on each pyramid. Changing chemical etching time from 1 to 12 min does not alter surface morphology. In order to observe the depth of nanopits, the same method was employed onto the planar single crystalline silicon. Careful observation shows that longer etching time resulted in deeper pits (Fig. 2). When the sample was etched for 1 min, the nanopits were not as obvious. However, increasing the chemical etching time to 3 min, 5 min, 10 min, and 12 min,

Fig. 1. (a) UV–vis absorption spectra of Ag NPs in colloid suspension with different sizes. (b) XPS of the Ag NPs. (c) SEM image of the Ag NPs on Si substrate. (d) Crosssectional SEM image of the solar cell after etching.

182

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

Fig. 2. Cross-sectional SEM images of crystalline silicon wafers after etching for (a) 1 min, (b) 3 min, (c) 5 min, (d) 10 min, (e) 12 min.

results in the depth with 60 nm, 110 nm, 200 nm, and 250 nm respectively. Fig. 3(a) shows the total reflection measurement using an integrating sphere in a wavelength range of 400–1100 nm. Comparing to the conventional pyramid-structured sample that shows a typical reflectance  12% from 500 to  1000 nm and even higher reflectance in shorter and longer wavelength ranges. It shows greatly decreased reflection as a result of the chemical etching, moreover, the longer the etching time, the lower the reflection. Particularly, for the sample with 10 min chemical etching, the average reflectance is lowered to o 5%, even lower than the unetched solar cell with the SiNx AR coating that shows 7% average reflectance. The QE is a good measure for how efficiently a solar cell converts photons into electrons as a function of wavelength. We have measured the external QE (EQE) in the wavelength range 400– 1100 nm, as shown in Fig. 3(b). Without the secondary texture, the reference sample with only the first order pyramidal texture gives an EQE about 79% in spectrum range from 600 to 800 nm. When the cell is etched using the Ag NPs for 1 min, the EQE increased to

 82%, reaching to the level of the unetched cell with the SiNx AR coating for the most of the spectrum range. Further increasing the chemical etching time to 3 min, 5 min, and 10 min, improves the EQE to 88%, 91%, and 95%, respectively. The EQE value is seen to decrease when the chemical etching time is further extended beyond 10 min, likely due to defect formation that would lead to the recombination of the holes and electrons [27,35–37]. It appears that 10 min etching gives the best EQE result 490% in the wavelength range of 530–850 nm, increased by  17% comparing to the EQE measurement without etching. It is apparent that the samples with the secondary nanostructure have significantly increased EQE than the reference sample with only the pyramid texture over the wavelength range 450–1100 nm, confirming that the secondary nanostructure is indeed effective in light trapping. Moreover, Fig. 3(b) shows that the longer the etching time, the higher the EQE. The internal QE (IQE) is a more direct measure of intrinsic quality of the solar cell structure as it excludes loss due to the optical reflection. The IQE is defined as follow:

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

183

Fig. 3. (a) Reflectance spectra, (b) EQE and (c) IQE of crystalline silicon solar cells with hybrid nanostructure after different etching times and unetched cell with SiNx. (d) EQE of the hybrid-nanostructured crystalline silicon solar cells measured with and without  1 V electric bias.

IQE=EQE/(1 − R)

(1)

where R is the reflectance. Fig. 3(c) shows the IQE as a function of etch time. It is clear that the cell without etching shows the lowest IQE  88% in spectrum range 580–780 nm. As chemical etching time increases, the IQE goes up and it reaches the maximum  97% at 10 min etch time. When the etch time is further extended to 12 min, IQE drops to 94%. As the 12 min etching gives almost identical reflection as that for the 10 min one, the IQE deterioration is likely due to the carrier recombination caused by surface defects. The unetched cell with the SiNx AR coating shows significant improvement in both EQE and IQE in the short wavelength range o500 nm, indicating that the SiNx coating does have a passivation effect. In order to investigate the carrier recombination, an electric bias of  1 V was applied to the solar cell during the EQE measurement. Fig. 3(d) shows that the bias indeed improves the EQE from 95% to 97%, implying that there are still some electron–hole recombination losses in the device. The significant impact on reflectance and QE can be explained by two effects: effective medium effect [38–41] of the nanopits and the LSPR effect of the Ag NPs [12–15]. The pyramid structure at the micrometer (1–2 mm) scale leads to the light-harvesting effect due to the strong forward scattering based on Mie theory [42]. As the light wave strikes the pyramids, it diffracts into several beams and partly rebounces between the pyramids, preventing light from escaping back to air. Thus, the pyramid structure acts as a buffer layer to compensate for the large mismatch in effective refractive indexes between the air and the silicon material, as illustrated in Fig. 4(a). In this case, although the refractive index changes gradually, the AR effect is not very effective because the feature sizes are much larger than the wavelengths of the light, and the light seems to illuminate on

Fig. 4. Schematic illustrations showing the relevance between the structure and the effective refractive index profile: (a) air to pyramid structure, (b) air to hybrid structure. The yellow points represent Ag NPs. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

inclined silicon surfaces. When the smaller nanopits are introduced by chemical etching onto the pyramids, they behave as an effective medium to improve the impedance mismatch of the refractive index from air to silicon, leading to gradually increased effective refractive index [4,43,44] (Fig. 4(b)). Because the sizes of the nanopits are much smaller than the light wavelength, the gradually increased effective refractive index provides a broadband AR outcome. As the incident light is reflected at different depths from the surface, the suppression of the reflection over a wide spectrum range through destructive interferences when the light waves with different phases partially cancel one another

184

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

[40,41,45], resulting in lower reflection, increased light trapping and therefore improved QE. To analyze the effect of the Ag NPs, and EQE of solar cells were measured before and after removing the Ag NPs. Fig. 5 shows the enhancement of the EQE (the difference between the EQE with Ag NPs and that after the Ag NPs dissolved) versus wavelength. The samples with the Ag NPs exhibit greater light absorption than the samples without the Ag NPs in the range of 350–500 nm, showing an evident peak centered at 405 nm, the same wavelength as the UV–vis absorption peak of the Ag NPs. The EQE increases demonstrate existence of the LSPR effect of the Ag NPs that lead to more electron–hole pairs in the semiconductor. There are three possible mechanisms that maybe related to the phenomenon. The first one suggests that the LSPR causes enhanced local electromagnetic field, leading to the local generation of electron–hole pairs [13,14,46–49]. As the rate of electron–hole formation is proportional to the local intensity of the electric field, the rate of electron–hole formation may increase by a few orders of magnitude when the LSPR effect occurs [3,15]. The second one is based on a non-radiative mechanism that plasmonic energy transfer takes the form of a resonant energy transfer (RET) process [14]. The RET process directly excites electron–hole pairs non-radiatively through the relaxation of the localized surface plasmon dipole. Another possible reason is called “hot electron” [15,50–53]. Non-radiative decay in Ag NPs can likely take place through intraband excitations within the conduction band. After non-radiative surface plasmon decay, surface plasmons in the Ag NPs can transfer energies to hot electrons. Because there is a Schottky barrier between the Ag NPs and silicon, hot electrons with energies higher than the Schottky barrier energy can be injected into the silicon with an emission efficiency dependent upon their energy level. After injection of hot electrons into the silicon, the Ag NPs are left positively charged because of electronic depletion. For this mechanism to be effective, the silicon material is required to be in intimate contact with the Ag NPs to transport the generated holes to the counter electrode, keeping the charge balance and sustaining an electric current. It is apparent in Fig. 5 that the EQE increases as the sample is etched with longer time up to 10 min, indicating that the Ag NPs contribute to a stronger LSPR effect when they gradually approach to the PN junction. Evidently, reducing the spacing between the Ag NPs

and the PN junction enables an effective overlap between the active layer and the concentrated electromagnetic field excited by the Ag NPs, and by overlapping this electromagnetic field with the active layer, the highly localized field can excite more electron–hole pairs locally without the need for the photons to travel too far [54,55]. Meanwhile, there is a weak peak at 810 nm, exactly twofold of the main peak wavelength 405 nm. There could be three possible interpretations. (1) The Ag NPs are in contact with two different interface materials, namely Si and air. Very likely the first one has the long wavelength resonance at 810 nm and the second one has the short wavelength resonance at 405 nm, both enhance light trapping [56]. (2) The second possibility is that the 810 nm peak is from the dipole excitation and the 405 nm one from the quadruple excitation. (3) In addition, two lower energy photons at 810 nm wavelength may combine at the Ag NPs to work like one high energy photon at 405 nm via the well-known upper conversion mechanism [57,58]. It is also known that the surface of the Ag NPs can be oxidized to form a thin surface layer Ag2O film, leading to red-shift of the plasmonic peak and intensity change [59]. The JSC can be directly calculated from the EQE data using the following equation:

JSC = q

∫ QE(λ) S (λ) dλ

(2)

where q is the electron charge and S(λ) is the standard spectral photon density (Air Mass 1.5 global, AM 1.5G). By integrating the experimentally measured EQE with the standard AM 1.5G solar spectrum, the JSC value can rise from 28.85 mA/cm2 to 33.79 mA/cm2 with a maximal enhancement of 17% due to the light trapping effect from the hybrid nanostructure, as shown in Table 1. Upon depositing the SiNx AR layer, the complete hybrid structured solar cell devices were characterized under the AM 1.5G illumination for its current–voltage (I–V) performance, including open-circuit voltage (VOC), JSC, fill factor (FF), and efficiency (η) of the cells with different etching times as summarized in Table 2. The solar cell without etching shows VOC of 586 mV, JSC of 34.81 mA/cm2, FF of 73.06%, and η of 14.90%, whereas the best solar cell etched for 5 min shows improved JSC to 35.52 mA/cm2, FF 62.23%, and η 12.42%. The net current gain by the NPs and the nanopits in the completely finished solar cell is about 2%, even after the SiNx AR layer has already provided effective AR effect. We assumed that the decrease of FF and VOC are caused by contamination and lack of perfection in our current process. However, with improvement in passivation and processing, we believe that the high JSC can be retained without sacrificing VOC or FF. Therefore, it is expected that high solar cell efficiency can be achieved using the technique.

4. Conclusion

Fig. 5. Enhancement of EQE (the difference between the EQE with Ag NPs and that after the Ag NPs dissolved) for etched crystalline silicon solar cell with Ag NPs after different etching times.

In this letter, a simple method is introduced to form Ag NPs to etch the pyramid structured single crystalline silicon solar cell to form a special hybrid nanostructure. The hybrid nanostructure can effectively trap light by effective medium effect of the nanopits and the LSPR effect of the Ag NPs, resulting in decreased reflection loss and increased EQE to 95% and JSC by as much as 17% for the cells without SiNx AR coating. The EQE measured under a  1 V

Table 1 JSC of crystalline silicon solar cell with hybrid structure (without SiNx). Etching time (min) 2

JSC before removing Ag NPs (mA/cm ) JSC enhancement (%)

0

1

3

5

10

12

28.85 –

29.83 3.39

31.81 10.26

32.98 14.32

33.79 17.12

32.86 13.90

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

Table 2 Efficiency of crystalline silicon solar cell with hybrid structure (with SiNx). Etching time (min)

0

1

3

5

10

12

JSC (mA/cm2) VOC (mV) FF (%) η (%)

34.81 586 73.06 14.90

34.69 577 56.54 11.32

34.81 566 60.48 11.92

35.52 562 62.23 12.42

35.30 559 58.48 11.54

34.70 552 44.77 8.58

bias is as high as 97%. For the incident light with a wavelength from 350 nm to 500 nm, the LSPR effect of the Ag NPs occurs. After depositing the SiNx AR layer, the best hybrid structured solar cell device shows JSC increase by 2% to 35.52 mA/cm2, FF to 62.23%, and η to 12.42%.

Acknowledgment The authors acknowledge support from the Overseas Talent Recruitment Project (B14041), the 1000-Talent-plan (1110010341), the National University Research Fund (GK261001009), supported by Program for Changjiang Scholars and Innovative Research Team in University (IRT_14R33).

References [1] J. Poortmans, V. Arkhipov, Thin Film Solar Cells: Fabrication, Characterization and Application, first edition,. John Wiley & Sons Ltd., Chichester, 2006. [2] J. Li, H. Yu, Y. Li, Aligned Si nanowire-based solar cells, Nanoscale 3 (2011) 4888. [3] V.E. Ferry, J.N. Munday, H.A. Atwater, Design considerations for plasmonic photovoltaics, Adv. Mater. 22 (2010) 4794. [4] W.Q. Xie, J.I. Oh, W.Z. Shen, Realization of effective light trapping and omnidirectional antireflection in smooth surface silicon nanowire arrays, Nanotechnology 22 (2011) 065704. [5] Q. Yang, X.A. Zhang, A. Bagal, W. Guo, C.-H. Chang, Antireflection effects at nanostructured material interfaces and the suppression of thin-film interference, Nanotechnology 24 (2013) 235202. [6] K.-Q. Peng, Y.-J. Yan, S.-P. Gao, J. Zhu, Synthesis of large-area silicon nanowire arrays via self-assembling nanoelectrochemistry, Adv. Mater. 14 (2002) 1164. [7] K. Peng, Y. Wu, H. Fang, X. Zhong, Y. Xu, J. Zhu, Uniform, axial-orientation alignment of one-dimensional single-crystal silicon nanostructure arrays, Angew. Chem. Int. Ed. 44 (2005) 2737. [8] Z. Huang, H. Fang, J. Zhu, Fabrication of silicon nanowire arrays with controlled diameter, length, and density, Adv. Mater. 19 (2007) 744. [9] Z. Huang, N. Geyer, P. Werner, J. de Boor, U. Gosele, Metal-assisted chemical etching of silicon: a review, Adv. Mater. 23 (2011) 285. [10] X. Li, P.W. Bohn, Metal-assisted chemical etching in HF/H2O2 produces porous silicon, Appl. Phys. Lett. 77 (2000) 2572. [11] K. Peng, Y. Yan, S. Gao, J. Zhu, Dendrite-assited growth of silicon nanowires in electroless metal deposition, Adv. Funct. Mater. 13 (2003) 127. [12] K.M. Mayer, J.H. Hafner, Localized surface plasmon resonance sensors, Chem. Rev. 111 (2011) 3828. [13] H.A. Atwater, A. Polman, Plasmonics for improved photovoltaic devices, Nat. Mater. 9 (2010) 205. [14] S.K. Cushing, J. Li, F. Meng, T.R. Senty, S. Suri, M. Zhi, M. Li, A.D. Bristow, N. Wu, Photocatalytic activity enhanced by plasmonic resonant energy transfer from metal to semiconductor, J. Am. Chem. Soc. 134 (2012) 15033. [15] S. Linic, P. Christopher, D.B. Ingram, Plasmonic-metal nanostructures for efficient conversion of solar to chemical energy, Nat. Mater. 10 (2011) 911. [16] K.R. Catchpole, A. Polman, Plasmonic solar cells, Opt. Express 16 (2008) 21793. [17] S.K. Sardana, V.S.N. Chava, E. Thouti, N. Chander, S. Kumar, S.R. Reddy, V. K. Komarala, Influence of surface plasmon resonances of silver nanoparticles on optical and electrical properties of textured silicon solar cell, Appl. Phys. Lett. 104 (2014) 073903. [18] C.J. Murphy, T.K. Sau, A.M. Gole, C.J. Orendorff, J. Gao, L. Gou, S.E. Hunyadi, T. Li, Anisotropic metal nanoparticles: synthesis, assembly, and optical applications, J. Phys. Chem. B 109 (2005) 13857. [19] A. Henglein, Small-particle research: physicochemical properties of extremely small colloidal metal and semiconductor particles, Chem. Rev. 89 (1989) 1861. [20] S. Link, M.A. El-Sayed, Spectral properties and relaxation dynamics of surface plasmon electronic oscillations in gold and silver nanodots and nanorods, J. Phys. Chem. B 103 (1999) 8410.

185

[21] J. Pérez-Juste, I. Pastoriza Santos, L.M. Liz-Marzán, P. Mulvaney, Gold nanorods: synthesis, characterization and applications, Coord. Chem. Rev. 249 (2005) 1870. [22] M. Moskovits, Surface-enhanced spectroscopy, Rev. Mod. Phys. 57 (1985) 783. [23] K. Kneipp, Y. Wang, H. Kneipp, L.T. Perelman, I. Itzkan, R.R. Dasari, M.S. Feld, Single molecule detection using surface-enhanced Raman scattering (SERS), Phys. Rev. Lett. 78 (1997) 1667. [24] S.M. Nie, S.R. Emery, Probing single molecules and single nanoparticles by surface-enhanced Raman scattering, Science 275 (1997) 1102. [25] K. Li, M.I. Stockman, D.J. Bergman, Self-similar chain of metal nanospheres as an efficient nanolens, Phys. Rev. Lett. 91 (2003) 227402-1. [26] T. Kalkbrenner, M. Ramstein, J. Mlynek, V. Sandoghdar, A single gold particle as a probe for apertureless scanning near-field optical microscopy, J. Microsc. 202 (2001) 72. [27] H. Ditlbacher, J.R. Krenn, G. Schider, A. Leitner, F.R. Aussenegg, Two-dimensional optics with surface plasmon polaritons, Appl. Phys. Lett. 81 (2002) 1762. [28] A. Pors, A.V. Uskov, M. Willatzen, I.E. Protsenko, Control of the input efficiency of photons into solar cells with plasmonic nanoparticles, Opt. Commun. 284 (2011) 2226. [29] M.L. de Souza, P. Corio, A.G. Brolo, Cu nanoparticles enable plasmonicimproved silicon photovoltaic devices, Phys. Chem. Chem. Phys. 14 (2012) 15722. [30] S.H. Lim, W. Mar, P. Matheu, D. Derkacs, E.T. Yu, Photocurrent spectroscopy of optical absorption enhancement in silicon photodiodes via scattering from surface plasmon polaritons in gold nanoparticles, J. Appl. Phys. 101 (2007) 104309. [31] Y. Zhang, X. Chen, Z. Ouyang, H. Lu, B. Jia, Z. Shi, M. Gu, Improved multicrystalline Si solar cells by light trapping from Al nanoparticle enhanced antireflection coating, Opt. Mater. Express 3 (2013) 489. [32] M. Losurdo, M.M. Giangregorio, G.V. Bianco, A. Sacchetti, P. Capezzuto, G. Bruno, Enhanced absorption in Au nanoparticles/a-Si:H/c-Si heterojunction solar cells exploiting Au surface plasmon resonance, Sol. Energy Mater. Sol. Cells 93 (2009) 1749. [33] L. Xu, M. Yin, S. Liu, Agx@WO3 core–shell nanostructure for LSP enhanced chemical sensors, Sci. Rep. 4 (2014) 6745. [34] L. Xu, M. Yin, S. Liu, Superior sensor performance from Ag@WO3 core–shell nanostructure, J. Alloy. Compd. 623 (2015) 127. [35] K.R. Catchpole, A. Polman, Design principles for particle plasmon enhanced solar cells, Appl. Phys. Lett. 93 (2008) 191113. [36] Hao-Chih Yuan Jihun Oh, Howard M. Branz, An 18.2%-efficient black-silicon solar cell achieved through control of carrier recombination in nanostructures, Nat. Nanotechnol. 7 (2012) 743. [37] Hao-Chih Yuan, Vernon E. Yost, Matthew R. Page, Paul Stradins, Daniel L. Meier, Howard M. Branz, Efficient black silicon solar cell with a densitygraded nanoporous surface: optical properties, performance limitations, and design rules, Appl. Phys. Lett. 95 (2009) 123501. [38] P. Beckman, A. Spizzichno, The Scattering of Electromagnetic Waves from Rough Surfaces, first edition,. Pergamon Press, Oxford, 1963. [39] J.-Y. Jung, Z. Guo, S.-W. Jee, H.-D. Um, K.-T. Park, J.-H. Lee, A strong antireflective solar cell prepared by tapering silicon nanowires, Opt. Express 18 (2010) A286. [40] Y.-C. Chao, C.-Y. Chen, C.-A. Lin, J.-H. He, Light scattering by nanostructured anti-reflection coatings, Energy Environ. Sci. 4 (2011) 3436. [41] W.-R. Wei, M.-L. Tsai, S.-T. Ho, S.-H. Tai, C.-R. Ho, S.-H. Tsai, C.-W. Liu, R.-J. Chung, J.-H. He, Above-11%-efficiency organic–inorganic hybrid solar cells with omnidirectional harvesting characteristics by employing hierarchical photon-trapping structures, Nano Lett. 13 (2013) 3658. [42] J.R. Meyer-Arendt, Introduction to Classical and Modern Optics, fourth edition,. Prentice Hall, New York, 1995. [43] H.-C. Chang, K.-Y. Lai, Y.-A. Dai, H.-H. Wang, C.-A. Lin, J.-H. He, Nanowire arrays with controlled structure profiles for maximizing optical collection efficiency, Energy Environ. Sci. 4 (2011) 2863. [44] S. Zhong, Z. Huang, X. Lin, Y. Zeng, Y. Ma, W. Shen, High-efficiency nanostructured silicon solar cells on a large scale realized through the suppression of recombination channels, Adv. Mater. 27 (2015) 555. [45] L.-K. Yeh, K.-Y. Lai, G.-J. Lin, P.-H. Fu, H.-C. Chang, C.-A. Lin, J.-H. He, Giant efficiency enhancement of GaAs solar cells with graded antireflection layers based on syringelike ZnO nanorod arrays, Adv. Energy Mater. 1 (2011) 506. [46] M.A. Mahmoud, W. Qian, M.A. EI-Sayed, Following charge separation on the nanoscale in Cu2O–Au nanoframe hollow nanoparticles, Nano Lett. 11 (2011) 3285. [47] Z. Liu, W. Hou, P. Pavaskar, M. Aykol, S.B. Cronin, Plasmon resonant enhancement of photocatalytic water splitting under visible illumination, Nano Lett. 11 (2011) 1111. [48] L. Wang, C. Clavero, Z. Huba, K.J. Carroll, E.E. Carpenter, D. Gu, R.A. Lukaszew, R.A. Lukaszew, Plasmonics and enhanced magneto–optics in core–shell Co–Ag nanoparticles, Nano Lett. 11 (2011) 1237. [49] D.B. Ingram, S. Linic, Water splitting on composite plasmonic-metal/semiconductor photoelectrodes: evidence for selective plasmon-induced formation of charge carriers near the semiconductor surface, J. Am. Chem. Soc. 133 (2011) 5202. [50] C. Clavero, Plasmon-induced hot-electron generation at nanoparticle/metaloxide interfaces for photovoltaic and photocatalytic devices, Nat. Photonics 8 (2014) 95. [51] C. Sönnichsen, T. Franzl, T. Wilk, G. von Plessen, J. Feldmann, Drastic reduction of plasmon damping in gold nanorods, Phys. Rev. Lett. 88 (2002) 077402.

186

Y. Liu et al. / Solar Energy Materials & Solar Cells 140 (2015) 180–186

[52] M. Rycenga, C.M. Cobley, J. Zeng, W. Li, C.H. Moran, Q. Zhang, D. Qin, Y. Xia, Controlling the synthesis and assembly of silver nanostructures for plasmonic applications, Chem. Rev. 111 (2011) 3669. [53] S. Mubeen, G. Hernandez-Sosa, D. Moses, J. Lee, M. Moskovits, Plasmonic photosensitization of a wide band gap semiconductor: converting plasmons to charge carriers, Nano Lett. 11 (2011) 5548. [54] M. Gu, Z. Ouyang, B. Jia, N. Stokes, X. Chen, N. Fahim, X. Li, M.J. Ventura, Z. Shi, Nanoplasmonics: a frontier of photovoltaic solar cells, Nanophotonics 1 (2012) 235. [55] N.F. Fahim, Z. Ouyang, B. Jia, Y. Zhang, Z. Shi, Enhanced photocurrent in crystalline silicon solar cells by hybrid plasmonic antireflection coatings, Appl. Phys. Lett. 101 (2012) 261102.

[56] S. Pillai, M.A. Green, Plasmonics for photovoltaic applications, Sol. Energy Mater. Sol. Cells 94 (2010) 1481. [57] A. Podlipensky, J. Lange, G. Seifert, H. Graener, I. Cravetchi, Second-harmonic generation from ellipsoidal silver nanoparticles embedded in silica glass, Opt. Lett. 28 (2003) 716. [58] M.G. Papadopoulos, Andrzej J. Sadlej, Jerzy Leszczynski, Non-Linear Optical Properties of Matter, first edition,. Springer, Netherlands, 2006. [59] Anton Kuzma, Martin Weis, Sona Flickyngerova, et al., Influence of surface oxidation on plasmon resonance in monolayer of gold and silver nanoparticles, J. Appl. Phys. 112 (2012) 103531.