Plasmonic effects in amorphous silicon thin film solar cells with randomly textured metal back contact were investigated experimentally and numerically. The influence of different metal back contacts with and without ZnO interlayer was studied and losses in the individual layers of the solar cell were quantified. The amorphous silicon thin film solar cells were prepared on randomly textured substrates using large area production equipment and exhibit conversion efficiencies approaching 10%. The optical wave propagation within the solar cells was studied by Finite Difference Time Domain simulations. The quantum efficiency of solar cells with and without ZnO interlayer was simulated and the interplay between the reflection, quantum efficiency and absorption in the back contact will be discussed.
©2012 Optical Society of America
Light-trapping or photon management in thin film solar cells allows for both a reduction of the reflection losses and increased optical path length of the incident light inside the absorber layer [1, 2]. Randomly textured transparent conductive oxides (TCO) are widely used to improve the incoupling of light in the solar cell and diffract/scatter the incident light [1–6]. Since an amorphous silicon solar cell is very thin, the roughness of the front contact propagates through the layer stack . Therefore, the roughness of the metal back contact of the solar cell is comparable to the roughness of the front contact. The combination of the front and back surface texture allows for enhanced light trapping compared to solar cells with surface textures only in the front or the back [3, 4]. However, the interaction of light with rough metal interfaces can result in enhanced light scattering and diffraction [7–9], but also in an enhanced optical absorption within the metal/dielectric interface. Hence, light trapping and plasmonic losses have to be considered when determining the optimal surface texture. Possibly a trade-off exists between light trapping and plasmonic induced absorption of light in the back contact.
The plasmonic absorption in the back contact can be reduced by introducing a zinc-oxide (ZnO) interlayer between metal back contact and silicon layers . Schematic sketches of a thin film solar cell with and without ZnO interlayer are shown in Fig. 1(a) and 1(b). The influence of plasmonic absorption for different back metal contacts with and without ZnO interlayer was investigated in this study experimentally and by numerical simulation. A finite difference time domain (FDTD) tool was used to numerically simulate the wave propagation in the solar cell structures and to determine plasmonic losses in the metal back contact. The experimental results are described in Section 2, while simulation results are presented in Section 3.
2. Experimental results
The amorphous p-i-n silicon solar cells were prepared by plasma enhanced chemical vapor deposition (PECVD) on 1.43 m2 glass substrates. Due to the high defect density and low carrier diffusion lengths of amorphous silicon  the solar cell is very thin with a total thickness of 320 nm. The glass substrates were coated with a randomly textured tin oxide (SnO2) layer. Different metal back contacts were prepared with and without a ZnO interlayer. The silver (Ag) and aluminum (Al) metal contacts and the aluminum doped ZnO interlayer were prepared by DC magnetron sputtering. The samples without the ZnO interlayer will be used as reference.
The quantum efficiency and the total absorption of the amorphous silicon solar cells with Si/Al, ZnO/Al, Si/Ag and ZnO/Ag back contact are shown in Fig. 2 . The total absorption was determined from measurement of the solar cell total reflection (1-R). In order to obtain the total reflection, we measured both specular and diffused reflection of the solar cell samples by UV-VIS spectrometer with an integrating sphere. The total reflection of the solar cell is the sum of the specular and diffused reflection. For wavelengths shorter than 500 nm the incoming light gets absorbed in the front of the amorphous silicon solar cell and does not reach the back contact. Consequently, the quantum efficiency and the total absorption are similar for all samples. For longer wavelengths the incoming light reaches the back contact, and the performance of the solar cells depends on the back contact design. The samples without ZnO interlayer exhibit higher total absorption for wavelengths longer than 650 nm, but their quantum efficiency is lower compared to samples with ZnO interlayer. The higher total absorption in combination with the reduced quantum efficiency can be attributed to higher absorption of the light in the back contact [12, 13]. The higher absorption is caused by incoupling of the light into localized plasmon resonances in the nanotextured back contact . The additional ZnO interlayer between the silicon diode and the metal back contact causes a shift of the localized plasmon resonances to shorter wavelengths. Hence less light can be coupled into them, since for shorter wavelength almost no light reaches the back contact. Consequently, the optical losses in the back contact are reduced and the quantum efficiency and the short circuit current are increased [12, 14]. The quantum efficiencies in Fig. 2 were measured under short circuit conditions. Measurements at reverse bias result in almost identical quantum efficiencies underlining the good electrical properties of the solar cells. Therefore, poor charge extraction can be ruled out as an origin of the drop of the quantum efficiency for longer wavelengths.The electrical parameters of the amorphous silicon solar cell with different back contact configurations are summarized in Table 1 . The short circuit current increases by 1.2 mA/cm2 and 0.9 mA/cm2 for solar cells with ZnO/Al and ZnO/Ag back contact compared to Si/Al and Si/Ag back contact respectively. The increase in the short circuit current is attributed to a drop of the optical loss in the back contact. However, the direct measurement of optical losses in the back contact of a solar cell is not possible. Using optical simulation tools, the absorption in the back contact of the solar cell can be determined.
3. Simulation results
To analyze the wave propagation in the solar cell structure and determine the absorption for the individual layers of the solar cell various models were developed [15–19]. In this study finite difference time domain (FDTD) methods were used to numerically solve Maxwell’s equations in 3-dimensions . Analyzing the optical wave propagation for thin film solar cells on randomly textured substrates is complex, and in the following analysis, it was assumed that the randomly textured substrate can be approximated by a periodic surface texture. The periodic surface texture can be described by a unit cell, which provides all information of the periodically textured solar cell.
The schematic cross section of the solar cell with integrated pyramidal surface texture is shown in Fig. 3(a) and 3(d). The front contact consists of a 420 nm thick flat tin oxide (SnO2) layer with pyramidal texture on top. The base of the pyramid is assumed to be square shaped and the dimensions of the pyramidal texture are described by the period and the height. Conformal propagation of the surface feature was assumed in subsequent layers. The surface analysis of nanotextured tin oxide substrates shows that the pyramidal texture exhibits an average period and height of 320 nm and 80 nm. The average opening angle of the pyramids is approximately equal to 126°. In the optical simulation we assumed the period and the height of the pyramid texture to be 320 nm and 80 nm respectively. The amorphous hydrogenated silicon (a-Si:H) solar cell structure consists of a 10 nm p-doped a-Si:H, 300 nm intrinsic (i-layer) a-Si:H and 10 nm n-doped a-Si:H silicon. Back contact configurations Si/Al, ZnO/Al, Si/Ag, and ZnO/Ag were used in the simulation study. The ZnO interlayer thickness was kept constant to 100 nm. The input wave for the simulations was assumed to be circularly polarized, and wavelength range for input wave was between 300 nm and 800 nm.
The time average power loss, Q(x,y,z), was calculated using the equation:Figures 3(b), 3(c), 3(e) and 3(f) presents the power loss profile of the amorphous silicon solar cell with different back contact configurations for an incident wavelength of a 680 nm.
Figures 3(b) and 3(c) exhibit a slice through the center of a 3D power loss profile for solar cells with silver and aluminum back contact, respectively. The simulations show that the light penetrates the metal contacts by more than 10 nm. As a consequence the reflection from the back contact drops and the absorption in the active region of the p-i-n solar cell is reduced. Optical absorption in the Al back contact is distinctly higher compared to the Ag back contact. Introducing a ZnO interlayer leads to a distinct drop of the absorption in the Ag and Al back contact as depicted in Figs. 3(e) and 3(f) respectively. As a consequence, more light is being reflected back into solar cell, which results in an increased absorption in the p-i-n diode. The simulated quantum efficiency (QE) of solar cell with and without ZnO interlayer is shown in Fig. 4 . The quantum efficiency is defined as the ratio of the collected charge carriers and the number of input photons of a given energy. In our simulations we assume that all the generated charge carriers are collected. With other words, we assume the internal quantum efficiency to be 100%. Therefore, the simulated quantum efficiency is calculated as power absorbed in the intrinsic layer with respect to the total power, Popt, incident on the unit cell:Eq. (1). Figure 4 shows the simulated quantum efficiency and optical losses in the solar cell for different back contact configurations. The calculated optical losses in the back of the solar cell are presented together, since plasmonic losses are present at interface between metal and dielectric. A clear drop of the simulated quantum efficiency is observed for the solar cell without a ZnO interlayer, similar to experimental results. The periodic surface texture used for the simulation provides a good approximation of the randomly textured substrate. Numerical simulations show that absorption in the back contact is reduced when ZnO interlayer is present. Due to the optical properties of the ZnO the plasmon resonance wavelengths are shifted to shorter wavelengths. In this wavelength range the incoming light do not reach back contact and cannot be coupled into plasmon resonances. However, the optical losses are not completely eliminated by introducing the ZnO interlayer. A comparison of the optical losses in the Ag and Al back contact shows that the losses are higher for the Al back contact. Therefore, the optical simulations are consistent with the experimental results. Front contact losses (losses in front TCO layer and p-layer of p-i-n diode) are similar for all simulated back contact designs.
In the next step the short circuit current was determined. Based on the simulated quantum efficiency the short circuit current was calculated by:Table 2 . The measured short circuit current is also listed in the Table 2 and simulated results are in a good agreement with experimental values.
The optical simulations show that the plasmonic losses in the back contact are reduced when a ZnO interlayer is introduced. As a result, both the reflection losses and the quantum efficiency are being increased. The same behavior was observed for the experimental results. Due to the shift of the plasmonic resonances more light is reflected from the back contact into the solar cell compared to solar cells with Si/Ag and Si/Al back contact designs. The reflected light can be divided into specular and diffused reflection. The specular reflected light might leave the solar cell after passing thought the solar cell. Most of the diffuse reflected light is absorbed in the i-layer of the solar cell so that the quantum efficiency and short circuit current of the solar cell are increased. Whether the reflected light in the solar cells with ZnO/metal back contact contributes to the enhanced reflection or quantum efficiency depends on the interface texture. For surface textures with periods smaller than 200 nm, the light will not be effectively diffracted or scattered. The surface texture acts as refractive index gradient and the reflection losses will increase. For surface texture with a period of 200 nm to 500 nm the effect depends on the height of surface texture. For small heights (<50 nm) of the surface texture the back contact acts almost as flat back contact. Consequently the reflection losses are increased. For higher surface textures (50 - 250 nm) more light is diffracted/scattered leading to a distinct increase of the quantum efficiency. For periods larger than 500 nm the quantum efficiency drops again. The periods are too large to effectively scatter longer wavelength light (650 – 800 nm). Consequently the reflection losses will distinctly increase.When comparing different back contact designs the short circuit current and the conversion efficiency is not the only parameter to be considered. The material cost of the different contact materials is another factor which has to be considered. Using aluminum can reduce manufacturing costs for creating solar cell back contact by 30% compared to silver. Therefore, the ZnO interlayer in combination with an aluminum contact allows for an increase of the short circuit current while reducing the manufacturing cost of the solar cell back contact.
The plasmonic losses in metal back contacts of amorphous silicon thin film solar cell were investigated experimentally and numerically. Numerical simulations allow for quantification of optical losses within the solar cell. Experimental results and numerical simulations both confirmed that plasmonic losses in back contact of the solar cell are reduced when ZnO interlayer is present. Reducing the optical losses in the back contact of solar cells leads to increase of quantum efficiency and reflection of the solar cell, while front contact losses remain the same. When ZnO interlayer is present between the p-i-n solar cell and the silver/aluminum back contact conversion efficiency increases from 8.62% to 9.37% for aluminum and from 7.87% to 9.59% for silver back contact, compared to the corresponding cells without ZnO interlayer. The increase in the conversion efficiency is related to lower plasmonic losses in the ZnO/metal back contact compared to the Si/Metal back contact. Numerical simulation of the wave propagation in such structures shows a good agreement with the experimental results. The absorption in the back contact corresponds to a short circuit current of 6 mA/cm2 for aluminum and 5 mA/cm2 for silver. By introducing the ZnO interlayer the corresponding losses can be reduced to 3 mA/cm2 for aluminum and 2 mA/cm2 for silver. The short circuit current of the solar cell with a Zn/Ag and ZnO/Al back contacts are increased by 0.9 mA/cm2 and 1.2 mA/cm2. The ZnO interlayer in combination with an aluminum contact allows for an increase of the short circuit current while keeping the manufacturing cost of the solar cell back contact low.
References and links
1. A. V. Shah, H. Schade, M. Vanecek, J. Meier, E. Vallat-Sauvain, N. Wyrsch, U. Kroll, C. Droz, and J. Bailat, “Thin-film silicon solar cell technology,” Prog. Photovoltaics 12(23), 113–142 (2004). [CrossRef]
2. M. Konagai, “Present status and future prospects of silicon thin-film solar cells,” Jpn. J. Appl. Phys. 50, 030001 (2011). [CrossRef]
3. H. Sai, H. Jia, and M. Kondo, “Impact of front and rear texture of thin-film microcrystalline silicon solar cells on their light trapping properties,” J. Appl. Phys. 108(4), 044505 (2010). [CrossRef]
4. D. Madzharov, R. Dewan, and D. Knipp, “Influence of front and back grating on light trapping in microcrystalline thin-film silicon solar cells,” Opt. Express 19(S2Suppl 2), A95–A107 (2011). [CrossRef] [PubMed]
5. J. Müller, B. Rech, J. Springer, and M. Vanecek, “TCO and light trapping in silicon thin film solar cells,” Sol. Energy 77(6), 917–930 (2004). [CrossRef]
6. J. Meier, S. Dubail, S. Golay, U. Kroll, S. Faÿ, E. Vallat-Sauvain, L. Feitknecht, J. Dubail, and A. Shah, “Microcrystalline silicon and the impact on micromorph tandem solar cells,” Sol. Energy Mater. Sol. Cells 74(1-4), 457–467 (2002). [CrossRef]
7. K. Yamamoto, M. Yoshimi, Y. Tawada, Y. Okamoto, A. Nakajima, and S. Igari, “Thin-film poly-Si solar cells on glass substrate fabricated at low temperature,” Appl. Phys., A Mater. Sci. Process. 69(2), 179–185 (1999). [CrossRef]
9. V. E. Ferry, M. A. Verschuuren, H. B. T. Li, E. Verhagen, R. J. Walters, R. E. I. Schropp, H. A. Atwater, and A. Polman, “Light trapping in ultrathin plasmonic solar cells,” Opt. Express 18(S2Suppl 2), A237–A245 (2010). [CrossRef] [PubMed]
10. J. Morris, R. R. Arya, J. G. O’Dowd, and S. Wiedeman, “Absorption enhancement in hydrogenated amorphous silicon (a-Si:H) based solar cells,” J. Appl. Phys. 67(2), 1079–1087 (1990). [CrossRef]
11. R. A. Street, “Hydrogenated amorphous silicon,” Cambridge University Press, Cambridge (1991).
12. U. W. Paetzold, F. Hallermann, B. E. Pieters, U. Rau, R. Carius, and G. von Plessen, “Localized plasmonic losses at metal back contact of thin film silicon solar cells,” Proc. SPIE 7725, 772517, 772517-9 (2010). [CrossRef]
13. J. Hüpkes, T. Wätjen, R. V. Aubel, R. Schmitz, W. Reetz, and A. Gordijn, “Material study on Zno/Ag back reflectors for silicon thin film solar cells,” 23rd EU PVSEC Proc, Session 3AV(2.24), 2419–2421 (2008).
14. J. Springer, A. Poruba, L. Müllerova, M. Vanecek, O. Kluth, and B. Rech, “Absorption loss at nanorough silver back reflector of thin-film silicon solar cells,” J. Appl. Phys. 95(3), 1427–1429 (2004). [CrossRef]
15. M. Zeman, R. A. C. M. M. van Swaaij, J. W. Metselaar, and R. E. I. Schropp, “Optical modeling of a-Si:H solar cells with rough interfaces: effect of back contact and interface roughness,” J. Appl. Phys. 88(11), 6436–6443 (2000). [CrossRef]
16. J. Springer, A. Poruba, and M. Vanecek, “Improved three-dimensional optical model for thin-film silicon solar cells,” J. Appl. Phys. 96(9), 5329–5337 (2004). [CrossRef]
17. M. Zeman and J. Krc, “Optical and electrical modeling of thin-film silicon solar cells,” J. Mater. Res. 23(04), 889–898 (2008). [CrossRef]
18. R. Dewan, I. Vasilev, V. Jovanov, and D. Knipp, “Optical enhancement and losses of pyramid textured thin- film silicon solar cells,” J. Appl. Phys. 110(1), 013101 (2011). [CrossRef]
19. S. Fahr, T. Kirchartz, C. Rockstuhl, and F. Lederer, “Approaching the Lambertian limit in randomly textured thin-film solar cells,” Opt. Express 19(S4Suppl 4), A865–A874 (2011). [CrossRef] [PubMed]