Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles

Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles

G Model ARTICLE IN PRESS APSUSC-30552; No. of Pages 6 Applied Surface Science xxx (2015) xxx–xxx Contents lists available at ScienceDirect Applie...

2MB Sizes 2 Downloads 55 Views

G Model

ARTICLE IN PRESS

APSUSC-30552; No. of Pages 6

Applied Surface Science xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Applied Surface Science journal homepage: www.elsevier.com/locate/apsusc

Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles Wen-Jeng Ho ∗ , Chia-Min Chang, Po-Hung Tsai Department of Electro-Optical Engineering, National Taipei University of Technology, No. 1, Sec. 3, Zhongxiao E. Rd., Taipei 10608, Taiwan, ROC

a r t i c l e

i n f o

Article history: Received 4 October 2014 Received in revised form 27 May 2015 Accepted 6 June 2015 Available online xxx Keywords: Metal-assisted chemical etching (MACE) Nanoporous Surface recombination Subwavelength surface structure Thin-film solar cell

a b s t r a c t Surface properties of a thin-film p-on-n silicon solar cell with a subwavelength nanoporous structure fabricated on an emitter layer by using metal-assisted chemical etching (MACE) were investigated through an experiment and simulation. After 10-s MACE processing, the conversion efficiency increased by 43.09% (from 5.64% to 8.07%) was obtained, compared with a reference solar cell without MACE. The simulation result indicated that the surface recombination velocity was an exponential function of the etching time from 0 to 30 s, and showed close agreement with the experimental data. © 2015 Elsevier B.V. All rights reserved.

1. Introduction Light management on the surface of monocrystalline Si (c-Si) is crucial because flat silicon surfaces have high reflectivity, ranging from 35% to 40%. Generally, an antireflection coating (ARC) of SiOx , TiOx , and SiNx materials is provided on a semiconductor device surface to reduce surface reflection [1–3]. However, in a singlelayer ARC, the light incidence angle and spectral range are limited. Another method to reduce the reflectivity is to fabricate micropyramids and subwavelength structures on the device surface. While dry etching using the reactive ion etching (RIE) process [4,5] is the most widely used technology, it is too expensive for solar cells. Wet etching [6], using an aqueous alkaline solution, is cheaper, but its use is limited to c-Si and it is not suitable for multicrystalline silicon (mc-Si) since the mc-Si grains have different orientations, unlike the c-Si grains. Metal-assisted chemical etching (MACE) and perpendicular direction etching can be applied to c-Si and mc-Si to obtain nanoporous structures to show the light trapping effect and increase the optical path through multiple reflections [7–12]. The finite-difference time-domain method [13,14], rigorous coupled-wave analysis [15,16], and finite element method [17] have been widely used for subwavelength simulations. Most simulations have been performed under periodic conditions or for a certain distribution, but in the subwavelength nanoporous

∗ Corresponding author. Tel.: +886 227712171; fax: +886 287733216. E-mail address: [email protected] (W.-J. Ho).

structure obtained using the MACE method, the hole depth, hole diameter, and hole spacing are irregular. Specifying the morphology of subwavelength nanoporous structures is difficult; therefore, the surface of the fabricated solar cell was characterized based on light transmission (T) through the surface (T = 1 − R), which was determined from experimental-data-based simulations and R is surface reflectivity of device, to show the existence of the light trapping effect, which is typical of nanoporous surface structures [18]. Previous research has indicated that nanopores on the surface influence the surface recombination velocity, and, therefore, periodic nanoholes on the surface can be identified based on the effective surface recombination velocity Seff , which can be expressed as Seff = Si Ai /Az=0 , with Ai denoting the area of surface i, Si denoting the surface recombination of surface i, and Az=0 is the area for z = 0 [19]. This equation implies that recombination velocities under distinct surface conditions can be equivalently represented as distinct recombination velocities compared with plane z = 0; thus, the recombination velocity increases with the nanohole depth. In this study, the electrical and optical properties of the nanoporous emitter surface of a thin-film solar cell were experimentally characterized and simulated, and an attempt was made to separate the effects of the electrical properties and optical properties on photovoltaic performances because it help us analyze solar cell performance effectively. In previous studies, it took a considerable amount of experimental time to identify the optimal etching time because the sample size, etching solution concentration, and temperature influenced the device surface structure. Here,

http://dx.doi.org/10.1016/j.apsusc.2015.06.033 0169-4332/© 2015 Elsevier B.V. All rights reserved.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033

G Model APSUSC-30552; No. of Pages 6

ARTICLE IN PRESS W.-J. Ho et al. / Applied Surface Science xxx (2015) xxx–xxx

2

the simulations combined with experiment are performed to study the performance of thin-film Si solar cells with subwavelength nanoporous emitter profiles. 2. Experiment In this study, chemical vapor deposition was used to grow a 5.0-␮m-thick n− -Si base layer and a 0.87-␮m-thick p+ -Si emitter layer on an n+ -Si (1 0 0) wafer. The doping concentration of the base was approximately 2.6 × 1013 cm−3 and that of the emitter layer was approximately 4.3 × 1017 cm−3 . The MACE process involved three steps. First, after RCA clean processing, a 20-nmthick silver film was directly deposited on the emitter layer surface using e-beam evaporation. The pressure was 2.67 × 10−4 Pa and the ˚ deposition rate was 0.8 A/s for a 20-mA emission current. Next, the wafer was placed in a rapid thermal annealing (RTA) chamber in a N2 atmosphere at 300 ◦ C for 10 min to convert the island Ag-films to Ag nanoparticles. Finally, the samples were soaked in HF:H2 O2 :H2 O (8:1:40) solution for 1, 5, 10, 15 and 30 s. In MACE, H2 O2 caused silver nanoparticles to dissociate into silver ions, and the silver ions captured electrons from silicon and reconverted to silver nanoparticles. After the loss of electrons, silicon were oxidized to SiO2 by water and etched off by HF. After the MACE process,

the Ag nanoparticles were removed by an HNO3 etching solution to obtain a subwavelength nanoporous layer. The reflectivity of etched sample was measured as a function of the MACE time by using a UV/Vis/NIR spectrophotometer (Perkin–Elmer Lambda 35) in the wavelength range of 350–1050 nm. For solar cell fabrication, a 300-nm-thick Al film was deposited on the nanoporous emitter layer by using photolithograph photoresist patterns, and a 15-nm Ti/300-nm Al film was deposited on the rear side through e-beam evaporation. To ensure close ohmic contacts between the semiconductor and the metal, the samples were subsequently annealed in the RTA chamber at 450 ◦ C for 10 min in a N2 atmosphere after front side deposition. The depth of the nanopores was estimated for various etching times by using side-view scanning electron microscopy (SEM), and the SEM images for MACE times of (a) 1 s, (b) 5 s, (c) 10 s, (d) 15 s, and (e) 30 s are shown in Fig. 1. The nanopore depth for the MACE time of 10 s was approximately 405 nm, which was half of the emitter thickness, and that for the MACE time of 15 s was 524 nm. After the MACE time of 30 s, the depth was approximately 786 nm near the emitter-base layer interface. From the cross-section image, all the nanopores were perpendicular to the surface. The photovoltaic performance of the thin-film solar cells with various MACE times was measured using photovoltaic

Fig. 1. Side-view scanning electron microscopy (SEM) with MACE times of (a) 1 s, (b) 5 s, (c) 10 s, (d) 15 s, and (e) 30 s.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033

G Model APSUSC-30552; No. of Pages 6

ARTICLE IN PRESS W.-J. Ho et al. / Applied Surface Science xxx (2015) xxx–xxx

Fig. 2. Reflectivity as a function of the wavelength of the solar cells with MACE times of 0, 1, 5, 10, 15, and 30 s. The trend of the plot of average reflectivity versus etching time is also shown in the inset.

current–voltage (I–V) measurements under one-sun illumination (AM 1.5G, 1000 mW/cm2 ) at 25 ◦ C. Before each measurement, the solar simulator (XES-151S, San-Ei Electric Co., Ltd.) was calibrated using a National Renewable Energy Laboratory-certified crystalline silicon reference cell (PVM-236). The external quantum efficiency (EQE, Enli Technology Co., Ltd.) was measured to determine how the subwavelength structure influenced the device. To discuss determine electrical and optical properties of the thin-film solar cell with a subwavelength nanoporous structure, a simulation was performed using the APSYS commercial software. 3. Results and discussion Fig. 2 shows the reflectivity of the solar cells as a function of the wavelength. The reflectivity decreased with an increase in the MACE time. The total reflection decreased faster with an increase in the etching time. Which aspect of the phenomenon matches the subwavelength. For a MACE time of 1 s, the silver was quickly dissociated by H2 O2 , and, therefore, only shallow holes were formed. Subsequently, the reflection with MACE time >1 s was totally down because the hole depth continued to increase with continued etching. Etching for 5, 10, and 15 s resulted in peak reflectivity corresponding to wavelengths of 475, 720, and 980 nm, which was caused by the high specific surface pore with periodic wave reflection, and the peak shifted to longer wavelengths as the etching time increased. The trend of the plot of average reflectivity versus etching time is shown in the inset of Fig. 2. The average reflectivity shows that there is no major change after the etch time of 15 s, and the etching depth shows that the etching rate slightly slows down at long etching times because the H2 O2 continues to be consumed. Fig. 3 shows photovoltaic current density–voltage (J–V) curves of thin-film solar cells with distinct etching times under AM1.5G one-sun illumination at 25 ◦ C. The bare thin-film silicon solar cell that was not etched was examined first, and at the Voc value of 0.532 V, Jsc was 15.31 mA/cm2 , the fill factor (FF) was 0.69, and  was 5.64%. Compared with the Voc value for the bare cell, Voc increased up to 10 s and then decreased after 15 s. The reason is that the surface recombination velocity exceeded the light trapping enhancement as well as Jsc . A maximal efficiency of 8.07% and an efficiency enhancement of 43.09% were obtained for the cell with a MACE time of 10 s relative to the values of these parameters for the bare solar cell. For the MACE time of 30 s, the result shows that the light trapping effect continued to exist; however, the light trapping

3

Fig. 3. Photovoltaic current density–voltage (J–V) curves of solar cells with MACE times of 0, 1, 5, 10, 15, and 30 s. Table 1 Photovoltaic performance of solar cells with MACE times of 0, 1, 5, 10, 15, and 30 s. Cell Bare cell Etching 1s Etching 5s Etching 10s Etching 15s Etching 30s

Jsc (mA/cm2 )

Voc (V)

FF (%)

 (%)

 Enhancement (%)

15.31 16.52 20.04 21.09 21.01 20.12

0.532 0.540 0.548 0.548 0.545 0.531

69 71 72 70 70 71

5.64 6.30 7.88 8.07 8.00 7.55

11.70 39.72 43.09 41.84 33.87

contribution was lower than the recombination losses, and thus, the efficiency decreased. Photovoltaic performance of solar cells with MACE times of 0, 1, 5, 10, 15, and 30 s are summarized in Table 1. Fig. 4 shows the measured EQE response of thin-film solar cells as a function of the wavelength and MACE time. For the MACE time of 1 s, the EQE values slightly decreased at wavelengths shorter than 450 nm. However, this decrease apparently stopped at 450 nm and increased for longer wavelengths. For 5-s etching, the EQE values decreased considerably at short wavelengths (350–500 nm), but increased markedly in the long-wavelength band ( > 550 nm). To discuss the electrical and optical properties of solar cells with subwavelength surface-structure, the APSYS commercial software

Fig. 4. The measured EQE response of solar cells with MACE times of 0, 1, 5, 10, 15, and 30 s.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033

G Model

ARTICLE IN PRESS

APSUSC-30552; No. of Pages 6

W.-J. Ho et al. / Applied Surface Science xxx (2015) xxx–xxx

4

Fig. 6. Measured etching depth as a function of the etching time and surface recombination velocity as a function of the etching time, which obtained from exponential fitting, for simulation. Fig. 5. Simulation data of photovoltaic J–V and EQE response for various surface recombination velocities.

was used to simulate EQE and J–V performances, and the simulation structure was set to be identical to the experimental structure. The front side and rear side were connected through ohmic contacts, and the back surface recombination was set to be zero to focus on the front side surface recombination. The light source AM1.5 and Si material data were specified from the database in APSYS. The optical properties were set by light transmission through the surface, and the amount of transmission was determined from experimental data by using the expression T = 1 − R [18]. Three simulations were performed under (a) the constant light transmission power (set as 55%) and the various surface recombination velocities, (b) the different light transmission power determined from experiment (T = 1 − R) and various surface recombination velocities, and (c) the different light transmission power determined from experiment data and constant surface recombination velocity (1250 m/s). A comparison of the experimental and simulation results can provide clues for improving the solar cell conversion efficiency. Fig. 5 shows the simulation data for the J–V characteristics and EQE response for various surface recombination velocities. In the simulation, the surface recombination velocity was varied and the power transmission was set as 55%. The EQE spectrum at the surface recombination velocity of 1000 m/s was used to fit the experimental data, and the effect of the surface recombination velocity on EQE and J–V was examined from 1000 to 9000 m/s. The short-circuit current (Jsc ) changed from 15.89 to 14.23 mA/cm2 and the efficiency changed from 7.8% to 6.8% as the surface recombination velocity changed from 1000 to 9000 m/s. The EQE at the wavelength of 350 nm changed from 31.68% to 7.08% as the surface recombination velocity increased from 1000 to 9000 m/s, and it further proved that the surface recombination velocity influenced the short wavelengths tremendously. In previous research [19], the effect surface recombination velocity (Seff ) of the periodic 2D nanoporous model can be seen as: Seff =

 SA i i

i Az=0

= Sz=0 × n ×

 2x+2z  2x

= Sz=0 × n ×

 x+z  x

thus facilitating estimation of the surface recombination velocity (Sz = 0). The surface recombination velocity and etching depth as a function of the etching time are shown in Fig. 6. The fitting result was 1250 m/s for the bare cell, and a value of 1,000,000 m/s [20–22] was used for the 30-s etching time to represent a worse surface state because in simulation, the efficiency saturates at surface recombination velocities set larger than 1,000,000 m/s for a 30-s etching time. The fitting curve was set as an exponential slope because it presented a tremendous change in the surface recombination velocity. After the simulation electrical properties were set, the optical properties were set by considering various extents of light transmission in the experimental data. Fig. 7 shows the simulation results for the EQE response and J–V characteristics, which are summarized in Table 2. For the MACE time of 10 s, the efficiency reached the highest value of 9.99% for a surface recombination velocity of 10,199 m/s. However, for the MACE time of 10 s, the efficiency decreased to 9.81% at the surface recombination velocity of 1,000,000 m/s. In short, the nanopores on the surface may degrade the electrical properties and decrease the efficiency, but they enhance the optical properties. Therefore, the target was to optimize the efficiency of solar cells by light trapping and surface recombination effects.

(1)

where n denotes the number of holes, z represents the hole depth, and x denotes the hole radius. Therefore, as the depth z and the hole number n increased, the effective surface recombination velocity increased. However, in our experiment, HF still slightly etched the Si solar cell surface and the silver ions dissociated by H2 O2 may diffuse, so the holes number n would be increased and became nonperiodic condition. Here, the simulation Jsc fit the experimental Jsc ,

Fig. 7. Simulation results of photovoltaic J–V curves and EQE responses and different surface recombination velocity and light transmission.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033

G Model

ARTICLE IN PRESS

APSUSC-30552; No. of Pages 6

W.-J. Ho et al. / Applied Surface Science xxx (2015) xxx–xxx Table 2 Simulation results of photovoltaic performance for various surface recombination velocities and various transmission lights. Surface recombination velocity (m/s)

Average reflectivity (%)

Jsc (mA/cm2 )

 (%)

1250 (planar) 1507 (MACE 1 s) 7500 (MACE 5 s) 10,199 (MACE 10 s) 31,737 (MACE 15 s) 1,000,000 (MACE 30 s)

38.21 28.85 7.61 4.27 2.62 1.17

15.31 17.05 20.33 20.55 20.54 20.32

7.55 8.42 9.89 9.99 9.94 9.81

 Enhancement (%)

11.52 30.99 32.31 32.05 29.93

Table 3 Simulation results of photovoltaic performance for various transmission lights (with all surface recombination velocity set 1250 m/s). Cell

Average reflectivity (%)

Jsc (mA/cm2 )

 (%)

Bare cell Etching 1s Etching 5s Etching 10s Etching 15s Etching 30s

38.21 28.85 7.61 4.27 2.62 1.17

15.31 17.25 22.46 23.00 23.53 23.87

7.55 8.55 11.26 11.54 11.81 12.00

 Enhancement (%)

13.25 49.14 52.85 56.42 58.94

The final simulation was only for various extents of light transmission (with constant surface recombination velocity of1250 m/s), and the results are shown in Table 3. The efficiency increased up to 12% because the increase in the surface recombination velocity was not of concern. The result of this simulation also indicated that if the nanoporous surface had good passivation, the efficiency may be 10% to 12%. The experimental and simulated efficiencies are shown in Fig. 8. The difference between simulation and experiment data is also shown in this figure. From the experimental result, the efficiency was optimal for the 10-s MACE. The simulation of the exponential surface recombination matched the experimental result that the 10-s MACE was optimal. The difference between simulation and experimental values continued to increase, which led to another finding: the magnitude of surface recombination influences the efficiency, and therefore, the optimal result may correspond to an etching time between 5 s and 10 s. In this simulation, the simple relation T = 1−R was used, and this method did not consider absorption of light by the nanoporous layer. However, in the simulation, the surface area in the proposed

Fig. 8. Efficiency enhancement of simulation and experiment, and the efficiency enhancement difference between simulation and experiment.

5

structure was the same and did not decrease, so it was under the same baseline. Accordingly, the efficiency difference between experimental data and simulation data may result from the difference in Rs and Rsh between these two data sets or from the deviation of the initial Jsc fitting. Although this error would become more noticeable as the nanoporous layer increased in thickness, this way can easily improve the performance of the new solar cell. 4. Conclusion The subwavelength nanoporous structure on a solar cell surface, fabricated using MACE and its electrical and optical properties were studied through a simulation and an experiment. From the experiment, the highest efficiency of 8.07% was obtained for the cell with MACE 10 s. However, the simulation under exponential function of surface recombination velocity, the highest efficiency of 9.99% was observed under surface recombination velocity set as 10,199 m/s and using transmission power provided by MACE 10 s experimental data. From the difference of efficiency enhancement between simulation and experiment, the optimal result may correspond to an etching time between 5 s and 10 s. Thus, the experiment and simulation provided useful information to obtain the optimal photovoltaic performance of thin-film Si solar cells with subwavelength nanoporous on the emitter profiles. Acknowledgements The authors would like to thank the National Science Council of the Republic of China for financial support under Grant NSC-1002221-E-027-053-MY3 and MOST 103-2221-E-027-049-MY3. References [1] J. Li, A.M. DeBerardinis, L. Pu, M.C. Gupta, Optical properties of solutionprocessable semiconducting TiOx thin films for solar cell and other applications, Appl. Opt. 51 (2012) 1131–1136. [2] F. Wünsch, D. Klein, A. Podlasly, A. Ostmann, M. Schmidt, M. Kunst, Lowtemperature contacts through SixNy-antireflection coatings for inverted a-Si:H/c-Si hetero-contact solar cells, Sol. Energy Mat. Sol. Cells 93 (2009) 1024–1028. [3] D. Zhang, I.A. Digdaya, R. Santbergen, R.A.C.M.M. Van Swaaij, P. Bronsveld, M. Zeman, J.A.M. Van Roosmalen, A.W. Weeber, Design and fabrication of a SiOx/ITO double-layer anti-reflective coating for heterojunction silicon solar cells, Sol. Energy Mat. Sol. Cells 117 (2013) 132–138. [4] D.H. Macdonald, A. Cuevas, M.J. Kerr, C. Samundsett, D. Ruby, S. Winderbaum, A. Leo, Texturing industrial multicrystalline silicon solar cells, Sol. Energy 76 (2004) 277–283. [5] P.K. Singh, R. Kumar, M. Lal, S.N. Singh, B.K. Das, Effectiveness of anisotropic etching of silicon in aqueous alkaline solutions, Sol. Energy Mat. Sol. Cells 70 (2001) 103–113. [6] S. Winderbaum, O. Reinhold, F. Yun, Reactive ion etching (RIE) as a method for texturing polycrystalline silicon solar cells, Sol. Energy Mat. Sol. Cells 46 (1997) 239–248. [7] C. Chartier, S. Bastide, C. Lévy-Clément, Metal-assisted chemical etching of silicon in HF-H2 O2 , Electrochim. Acta 53 (2008) 5509–5516. [8] S. Chattopadhyay, X. Li, P.W. Bohn, In-plane control of morphology and tunable photoluminescence in porous silicon produced by metal-assisted electroless chemical etching, J. Appl. Phys. 91 (2002) 6134–6140. [9] S. Koynov, M.S. Brandt, M. Stutzmann, Black nonreflecting silicon surfaces for solar cells, Appl. Phys. Lett. 88 (2006). [10] K. Peng, M. Zhang, A. Lu, N.B. Wong, R. Zhang, S.T. Lee, Ordered silicon nanowire arrays via nanosphere lithography and metal-induced etching, Appl. Phys. Lett. 90 (2007). [11] 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). [12] D. Li, L. Wang, N. Zhou, Z. Feng, X. Zhong, D. Yang, Formation of nanostructured emitter for silicon solar cells using catalytic silver nanoparticles, Appl. Surf. Sci. 264 (2013) 621–624. [13] A. Deinega, I. Valuev, B. Potapkin, Y. Lozovik, Minimizing light reflection from dielectric textured surfaces, J. Opt. Soc. Am. A 28 (2011) 770–777. [14] C.H. Chiu, P. Yu, H.C. Kuo, C.C. Chen, T.C. Lu, S.C. Wang, S.H. Hsu, Y.J. Cheng, Y.C. Chang, Broadband and omnidirectional antireflection employing disordered GaN nanopillars, Opt. Expr. 16 (2008) 8748–8754.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033

G Model APSUSC-30552; No. of Pages 6 6

ARTICLE IN PRESS W.-J. Ho et al. / Applied Surface Science xxx (2015) xxx–xxx

[15] D. Lehr, M. Helgert, M. Sundermann, C. Morhard, C. Pacholski, J.P. Spatz, R. Brunner, Simulating different manufactured antireflective sub-wavelength structures considering the influence of local topographic variations, Opt. Expr. 18 (2010) 23878–23890. [16] G. Demésy, F. Zolla, A. Nicolet, M. Commandré, C. Fossati, The finite element method as applied to the diffraction by an anisotropic grating, Opt. Expr. 15 (2007) 18089–18102. [17] J.H. Schmid, P. Cheben, S. Janz, J. Lapointe, E. Post, D.X. Xu, Gradient-index antireflective subwavelength structures for planar waveguide facets, Opt. Lett. 32 (2007) 1794–1796. [18] L. Remache, E. Fourmond, A. Mahdjoub, J. Dupuis, M. Lemiti, Design of porous silicon/PECVD SiOx antireflection coatings for silicon solar cells, Mater. Sci. Eng. B: Solid State Adv. Technol. 176 (2011) 45–48.

[19] K. Xiong, S. Lu, D. Jiang, J. Dong, H. Yang, Effective recombination velocity of textured surfaces, Appl. Phys. Lett. 96 (2010) 193107/1–193107/193107. [20] U. Gangopadhyay, S. Roy, S. Garain, S. Jana, S. Das, Comparative simulation study between n-type and p-type silicon solar cells and the variation of efficiency of n-type solar cell by the application of passivation layer with different thickness using AFORS HET and PC1D, IOSR J. Eng. 2 (8) (2012) 41–48. [21] R. Stangl, A. Froitzheim, W. Fuhs, Thin film silicon emitters for crystalline silicon solar cells, epitaxial, amorphous or microcrystalline, PV Eur. – From PV Technol. Energy Solut. (2002) 123–126. [22] M.M. Hilali, P. Hacke, J.M. Gee, Two-dimensional modeling of EWT multicrystalline silicon solar cells and comparison with the IBC solar cell, in: IEEE 4th World Conference on Photovolt Energy Conversion, 2007, pp. 1299–1303.

Please cite this article in press as: W.-J. Ho, et al., Simulation and characterization of performance of thin-film silicon solar cells with subwavelength nanoporous emitter profiles, Appl. Surf. Sci. (2015), http://dx.doi.org/10.1016/j.apsusc.2015.06.033