Optimised Antireflection Coatings using Silicon Nitride ... - CyberLeninka

0 downloads 0 Views 535KB Size Report
programs such as PC1D. The measured and modelled reflection data are in good agreement, see Fig. 2. In. Table 1, the results are summarised for various SiNx ...
Available online at www.sciencedirect.com

EnergyProcedia Procedia1500 (2011) Energy (2012) 78000–000 – 83

Energy Procedia www.elsevier.com/locate/procedia

International Conference on Materials for Advanced Technologies 2011, Symposium O

Optimised Antireflection Coatings using Silicon Nitride on Textured Silicon Surfaces based on Measurements and Multidimensional Modelling Shubham Duttaguptaa,b,*, Fajun Maa,b, Bram Hoexa, Thomas Muellera and Armin G. Aberlea,b a

Solar Energy Research Institute of Singapore, National University of Singapore, 7 Engineering Drive 1, Block E3A, Singapore 117574, Singapore b Department of Electrical and Computer Engineering, National University of Singapore, 4 Engineering Drive 3, Block E4, Singapore 117576, Singapore

Abstract Plasma-deposited silicon nitride (a-SiNx:H, or briefly, SiNx) is currently the state-of-the-art antireflection coating for silicon wafer solar cells. It simultaneously reduces front-side optical reflection and provides surface and bulk passivation. Silicon nitride films with higher refractive index typically provide a higher level of crystalline silicon surface passivation in the as-deposited state, but the resulting solar cells suffer from a degraded blue response as the films become more absorbing. Hence, it is important to consider all loss mechanisms while optimising SiNx antireflection coatings for silicon wafer solar cells. In this work, the refractive index (n) of the SiNx films is varied from 1.9 to 2.7. The reflection and absorption losses of textured Si wafers coated with various SiNx films are quantified using 2D modelling. It is shown that SiNx films with n = 2.0 (at  = 633.3 nm) and thickness of 70 nm provide a weighted average reflectance (WAR1000) of less than 2.5 % and a weighted average transmission (WAT1000) of more than 97 % on textured mono-Si wafers, combined with a very low saturation current density of 100 fA/cm2 on 70 Ω/sq n+ layers. This shows that very good optical and excellent surface passivation quality can be realised on textured silicon wafers using inline deposited plasma silicon nitride. © by by Elsevier Ltd.Ltd. Selection and/orand/or peer-review under responsibility of the organizing committee © 2011 2011Published Published Elsevier Selection peer-review under responsibility of Solar Energyof International Conference on Materials for Advanced Technologies. Research Institute of Singapore (SERIS) – National University of Singapore (NUS). Keywords: Antireflection coatings; optical properties; silicon wafer solar cells; silicon nitride; 2D modelling

* Corresponding author. Tel.: +65 6516 5779; fax: +65 6775 1943 E-mail address: [email protected]

1876-6102 © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the organizing committee of International Conference on Materials for Advanced Technologies. doi:10.1016/j.egypro.2012.02.009

2

Shubham Duttagupta et al. / Energy Procedia 15 (2012) 78 – 83 S. Duttagupta et al./ Energy Procedia 00 (2011) 000–000

1. Introduction Plasma-deposited silicon nitride films are widely used nowadays as the standard antireflection coating (ARC) for industrial silicon wafer solar cells. The advantages of this ARC are well documented in the literature [1-6]. An efficient ARC reduces the reflection at the front surface of the solar cells, thereby increasing the photocurrent and thus the PV efficiency. Various materials have been considered as ARC for silicon wafer solar cells, but SiNx is nowadays the dominant material, for several reasons: i) it has a tuneable refractive index (1.9-2.9) which gives the flexibility to optimise the coating for applications in air or within a PV module; ii) it provides an excellent level of surface passivation; iii) it contains a high amount of hydrogen that facilitates bulk passivation, which is particularly important for multicrystalline Si wafer solar cells [1-6]. Although SiNx thin films are very well suited for reducing front surface reflection of Si wafer cells, the films have a significant parasitic absorption at wavelengths below 500 nm that cannot be neglected [3, 7]. Moreover, ARCs are typically studied on planar substrates and typically only the refractive index at a single wavelength (commonly 633 nm) is considered in the optimisation. In this paper, we investigate single-layer SiNx films as ARC on polished and textured mono-Si wafers. The experimental characterisation uses spectroscopic ellipsometry and spectrophotometry. This is complemented by further investigation of the SiNx films by 2D modelling in Sentaurus using ray-tracing and the transfer matrix method. The performance of the SiNx films is quantified by considering reflection losses, absorption losses, and the surface passivation quality on n+ diffused surfaces. 2. Experimental details 2.1. Sample preparation In this study the SiNx films were deposited in a commercial inline remote microwave-powered plasmaenhanced chemical vapour deposition (PECVD) machine (SiNA-XS, Roth & Rau), using an ammoniasilane plasma. The gas flows and gas ratios were varied to change the refractive index of the SiN x films. The substrate temperature, deposition pressure, and plasma power were set at 450 °C, 0.27 mbar, and 3500 W, respectively, which were previously found by us to be the optimal conditions for crystalline silicon surface passivation [8]. It should be noted that the actual sample temperature is approximately ~50 °C below the substrate temperature in this type of PECVD machine. The carrier transport speed was adjusted to obtain the desired thickness of the SiNx film. Moderately doped n-type Cz Si (4-5 Ωcm) wafers with thickness of 180 µm were used as substrates. The wafers were subjected to a 20 % KOH treatment at 80 °C for 6 minutes, resulting in the removal of 12 µm of silicon on each side to remove the saw damage. Then, alkaline texturing was carried out in KOH/IPA/DI water solution for 30 minutes at 80 °C, which resulted in a textured surface with upright random pyramids with a size of up to 10 µm. Subsequently, the wafers were cleaned with the standard RCA (Radio Corporation of America) cleaning sequence [9]. The textured wafers were then diffused in a tube furnace using POCl3 as the phosphorus source, giving a 70 Ω/sq P-doped n+ layer on both surfaces (n+/n/n+ structure). The optical properties of the SiNx thin films were determined using spectroscopic ellipsometry (SE, GES5E, Sopralab, France), whereby the SiNx films were deposited onto polished Fz Si wafers. These

79

80

Duttagupta et al. Procedia / Energy Procedia (2012) 78 – 83 S.Shubham Duttagupta et al./ Energy 00 (2011)15 000–000

samples were analysed in the 300-1000 nm wavelength range, to exclude the impact of reflection from the rear side of the polished wafers on the SE measurement. The ellipsometry data were analysed using a model-based analysis, whereby the SiNx film was modelled using the Tauc-Lorentz formalism [7]. The hemispherical reflection of the samples was measured using spectrophotometry (Lambda 950, PerkinElmer, USA) in the 300-1000 nm range. Reflection values for wavelengths above 1000 nm were not taken into account, as this part of the reflection spectrum is dominated by light that is reflected at the rear of the solar cell and is not absorbed in the silicon. From these reflection data we calculate the weighted average reflectance (WAR1000), which is a measure of the total reflected photon flux divided by the total incident photon flux of the AM1.5G spectrum. The passivation quality of the SiNx thin films on textured mono-Si wafers was quantified by the saturation current density J0 of the n+ layers. The saturation current density J0 was determined from contactless photoconductance decay measurements in both the quasi-steady-state and transient mode (Sinton WCT-100), using the relation proposed by Kane and Swanson [10, 11]. The J0 values were also determined after firing of the samples in an industrial fast firing furnace (Despatch, model Ultra-flex). 3. Results 3.1. Optical properties of inline PECVD a-SiNx:H films The optical properties of plasma silicon nitride films strongly depend on the deposition conditions in the PECVD reactor. Figure 1 shows the wavelength dependent refractive index and extinction coefficient for eight SiNx films that were grown with different NH3/SiH4 gas ratios on polished Si wafers. It is evident that SiNx has a significant absorption for short-wavelength light, especially for higher refractive index films. These results are very comparable to other studies reported in the literature [3], however the SiNx films deposited here exhibit a slightly lower extinction coefficient for similar refractive index SiNx films. This could indicate that these SiNx films have a relatively high mass density [12].

(a)

(b)

Fig. 1. (a) Refractive index and (b) extinction coefficient for SiNx films deposited with different SiH4/NH3 gas ratios.

3

Shubham Duttagupta et al. / Energy Procedia 15 (2012) 78 – 83 S. Duttagupta et al./ Energy Procedia 00 (2011) 000–000

4

3.2. Optimised a-SiNx:H films for solar cell application An optimal SiNx film for silicon wafer solar cells should have low optical losses and, simultaneously, a high level of surface passivation. SiNx films with a refractive index (at  = 633.3 nm) in the range of 1.9 to 2.7 were deposited onto the textured Si wafers. For every refractive index, the film thickness was varied from 55 to 80 nm in order to keep the optical thickness (n×d) of the film constant (as shown in Table 1). Figure 2(a) shows the measured reflectance of the silicon nitride coated textured Si wafers, demonstrating that WAR1000 values of 2.3% can be obtained for textured mono-Si wafers coated with a single SiNx film. In addition to a low reflection, it is important to achieve good surface passivation. Figure 2(b) shows the J0 of SiNx passivated n+ layers as a function of the refractive index of the SiNx films, before and after a standard industrial firing process.

(a)

(b)

Fig. 2. (a) Reflectance measurements on a planar Si wafer, a textured Si wafer, and a textured Si wafer coated with a SiNx film having n = 2.15 and d = 70 nm. The latter shows good agreement between measured data (WAR 1000 = 2.3%) and simulated data (WAR1000 = 2.4%) (the symbols are the measured data); (b) Measured saturation current densities (J0) of SiNx passivated textured n+ layers, before and after firing.

4. 2D Modelling With the experimentally determined optical properties of the SiNx films, the weighted average reflectance (WAR1000), weighted average absorption (WAA1000), weighted average transmission (WAT1000) up to a wavelength of 1000 nm was calculated with the help of 2D modelling using ray tracing and the transfer matrix method (Sentaurus, Synopsys, USA) [13]. The WAT1000 accounts for both reflection and absorption losses and basically quantifies the photon flux available for photocurrent generation. It is emphasised that modelling textured surfaces is a multidimensional problem and that an accurate modelling of optical losses at textured silicon surfaces is not possible using 1D simulation programs such as PC1D. The measured and modelled reflection data are in good agreement, see Fig. 2. In Table 1, the results are summarised for various SiNx films on planar as well as pyramid-textured Si wafers. Figure 3 shows the saturation current density (J0) of the n+ layers (left axis) and WAT1000 (right axis) as a function of the refractive index. It can be concluded from Fig. 3 that SiNx films with 2.0 ≤ n < 2.3 have a high transmission in combination with a good surface passivation after industrially firing and thus are suitable for solar cell applications.

81

82

Duttagupta et al. Procedia / Energy Procedia (2012) 78 – 83 S.Shubham Duttagupta et al./ Energy 00 (2011)15 000–000

5

Table 1: Calculated solar weighted average reflection, absorption and transmission for the various SiNx films on planar and textured mono-Si wafers. The optical thickness of the SiNx films was kept constant at ~150 nm in the simulations.

NH3/SiH4 ratio

n at 2.0 eV

k at 3.44 eV

α at 3.44 eV

d (nm)

Planar

Textured

WAR

WAA

WAT

WAR

WAA

WAT

5

1.90

0

0

79

10.5

0.0

89.5

2.5

0.0

97.5

3.5

1.95

1.11 x10-3

3.79 x102

77

10.4

0.0

89.6

2.6

0.0

97.4

3

2.00

8.46 x10-3

2.89 x103

75

10.3

0.2

89.5

2.6

0.3

97.1

2.5

2.03

2.67 x10-2

9.10 x103

74

10.3

0.6

89.1

2.5

0.9

96.6

-2

4

95.9

2

2.15

7.07 x10

70

10.2

1.1

88.6

2.4

1.7

1.5

2.37

3.04 x10-1

1.04 x105

64

12.5

4.1

83.4

2.8

5.7

91.5

1.2

2.60

4.99 x10-1

1.70 x105

58

16.1

5.9

78.0

3.8

8.0

88.2

1

2.70

6.20 x10-1

2.12 x105

56

17.6

7.5

74.9

4.3

10.1

85.6

2.41 x10

Fig. 3. Measured J0 and solar weighted average transmission (WAT1000) as a function of the refractive index of the silicon nitride film. The silicon nitride films with 2.0 ≤ n < 2.3 have higher photon transmission and excellent firing stable passivation. The lines are guides to the eye.

5. Conclusions The optical properties of SiNx films deposited by remote inline PECVD onto c-Si wafers were studied for a wide refractive index range. It was shown that the absorption in these films is significant and should be considered in the optimisation of the final solar cells. Reflectance and photoconductance measurement were done to obtain the relevant information. Furthermore, 2D modelling results were found to be in good agreement with measured reflection data, which allowed the extraction of the weighted average transmission of the SiNx films by taking into account both reflection and absorption losses. We showed that the total photon losses can be limited to below 3% for SiNx films that have an excellent level of surface passivation on textured n+ layers.

Shubham Duttagupta et al. / Energy Procedia 15 (2012) 78 – 83 S. Duttagupta et al./ Energy Procedia 00 (2011) 000–000

6

Acknowledgements SERIS is sponsored by the National University of Singapore and Singapore’s National Research Foundation through the Singapore Economic Development Board. This work was sponsored by the grant NRF2009EWT-CERP001-056 from the National Research Foundation of Singapore.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]

Nagel H, Aberle AG, Hezel R. Optimised antireflection coatings for planar silicon solar cells using remote PECVD silicon nitride and porous silicon dioxide. Prog. Photovoltaics 1999;7: 245-60. Soppe W, Rieffe H, Weeber A. Bulk and surface passivation of silicon solar cells accomplished by silicon nitride deposited on industrial scale by microwave PECVD. Prog. Photovoltaics 2005;13: 551-69. Doshi P, Jellison GE, Rohatgi A. Characterization and optimization of absorbing plasma-enhanced chemical vapor deposited antireflection coatings for silicon photovoltaics. Appl. Opt.1997;36: 7826-37. Winderbaum S, Leo AJ, Shea SP, Koval TD, Kumar B. Characterisation of industrial-scale remote PECVD SiN depositions. Proc. 3rd World Conf. on Photovoltaic Energy Conversion, Osaka, Japan; 2003, p. 1128. Wright DN, Marstein ES, Rognmo A, Holt A. Plasma-enhanced chemical vapour-deposited silicon nitride films; The effect of annealing on optical properties and etch rates. Sol. Energy Mat. Sol. Cells 2008; 92:1091-8. Swatowska B, Stapinski T. Amorphous hydrogenated silicon-nitride films for applications in solar cells. Vacuum 2008; 82: 942-6. Jellison GE, Modine FA. Parameterization of the optical functions of amorphous materials in the interband region. Appl. Phys. Lett. 1996;69: 371-3. Duttagupta S, Hoex B, Lin F, Mueller T, Aberle AG. High-quality surface passivation of low-resistivity p-type c-Si by hydrogenated amorphous silicon nitride deposited by industrial-scale microwave PECVD. Proc. 37th IEEE Photovoltaic Specialist Conf., Seattle, USA; 2011. Kern W. The evolution of silicon wafer cleaning technology. J. Electrochem. Soc. 1990;137:1887-92. Sinton RA, Cuevas A. Contactless determination of current-voltage characteristics and minority-carrier lifetimes in semiconductors from quasi-steady-state photoconductance data. Appl. Phys. Lett. 1996;69: 2510-2. Kane DE, Swanson RM. Measurement of the emitter saturation current by a contactless photoconductivity decay method. Proc. 18th IEEE Photovoltaic Specialist Conf., Las Vegas, USA; 1985, p. 578-81. Hoex B. Functional thin films for high-efficiency solar cells [PhD thesis]. Eindhoven University of Technology, The Netherlands; 2008. Sentaurus, Manual Version 2010.03, Synopsys Inc., Mountain View, CA, USA, 2008.

83