Academia.eduAcademia.edu

Porous silicon applications in solar cell technology

1997, Physica Scripta

AI-generated Abstract

This paper examines the role of porous silicon in advancing solar cell technology. It discusses the properties of porous silicon that enhance light absorption, improve charge carrier mobility, and facilitate better energy conversion efficiency. The analysis includes recent developments in the fabrication processes of porous silicon and its integration into various solar cell architectures, along with a review of current challenges and future directions for research in this area.

Home Search Collections Journals About Contact us My IOPscience Porous silicon applications in solar cell technology This content has been downloaded from IOPscience. Please scroll down to see the full text. 1997 Phys. Scr. 1997 255 (http://iopscience.iop.org/1402-4896/1997/T69/053) View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: 130.188.8.27 This content was downloaded on 30/06/2014 at 07:07 Please note that terms and conditions apply. Physica Scripta. Vol. T69, 255-258, 1997 Porous Silicon Applications in Solar Cell Technology V. Pacebutas, K. Grigoras and A. Krotkus Semiconductor Physics Institute, A. Gostauto 11,2600 Vilnius, Lithuania Received May 15,1996; accepted June 18,1996 Abstract Much less attention was paid to other potential applications of PS in the solar cell technology. Initial attempts to Two different applications of PS layers in the solar cell technology were demonstrated. In the h t case, microporous layers, which are formed on observe a photovoltaic effect at a PS/bulk silicon interface silicon surface after relatively long anodical etching, roughen surface and have been very discouraging. Schottky-type structures made enhance the optical con6nement in the wafer increasing quantum efficiency on the top of the porous layer had shown no photovoltage, in the spectral region close to 1pm. Such a light-trappingeffect was verified which was rather artificially explained by a complex, graded by spectral transmittivity and photoresponse measurements. In the second case, thin nanoporous layers formed by a short anodical treatment on bandgap structure of the porous layer [SI. Further investialready manufactured solar cells were investigated. The increase of the gation of this effect has shown that it is originating from short-circuit current and efficiency by nearly 30% has been observed. Such poor electrical transport properties of the topmost, nanoan improvement was explained by antireflective and surface passivating porous part of the PS layer [7]. actions of the porous silicon layer. In this paper, we present the results of the investigation on PS application in silicon photovoltaic cell technology. It will be demonstrated that the porous layer formation can be 1. Introduction beneficial in both the surface texturization and the antiPorous silicon (PS) layers obtained by anodical etching in a reflection coating. It will be demonstrated also that PS hydrofluoric acid electrolyte are potentially interesting for could further enhance the cell performance due to its surface photovoltaic applications from several points of view. It is a passivation during the anodical etching. sponge-like material which could efficiently distort the wavefront of the incoming radiation and cause the light 2. Porous layer formation trapping in a cell due to the textured surface. Because of its porous nature, the effective refractive index of PS is lower The PS layers were formed by using a standard anodical than that of bulk silicon thus it can be used for anti- etching in a HF : C,H,OH or HF :H,O (1:1) electrolytes reflection coating of silicon solar cells and for enhancing with a net from a platinum wire as a cathode. The layer their external quantum efficiencies. In addition, recent structure was investigated by SEM and TEM, its effect on results of the visible light emission from PS indicate that the photovoltaic performance was studied by spectral reflecthere is an apparent increase of the PS bandgap relative to tance and transmittance measurements and by measuring the bandgap of the bulk silicon due to the quantum size I-I/ characteristics of the solar cells before and after anodeffect [l], which makes PS a good candidate for the top ization. Generally, there are two types of a porous layer obtained layer in a silicon-based tandem-cell configuration. The majority of the research work which has been per- on the silicon wafer after its electrochemical etching in a formed until recently was dedicated to the investigation of HF-based electrolyte. The top part of the new structure is the second from the three mentioned above applications of made of h e , nanometer-size silicon particles surrounded by PS in the photovoltaic solar cell technology - its use as an a matrix of silicon oxide and other products of the chemical antireflection coating of a cell [2-51. Prasad et al. [2] had reaction. This nanoporous layer is penetrated by multiple observed significant enhancements of both the short-circuit cavities with average diameters of a few tens of nanometers, current (by 25%) and the open-circuit voltage after porous its electrical resistivity is high, and the absorption edge is layer formation as early as in 1982. However, the PS forma- strongly blue-shifted as compared with the optical absorption process used in this investigation was purely chemical, tion in bulk silicon. Remaining part of the PS layer, which is without any external electrical bias, which is rather slow lying beneath the surface nanoporous layer, becomes and, therefore, hardly applicable to already manufactured sponge-like with characteristic dimensions of the voids and cells with the top contact metallization. Electrochemical the silicon columns ranging (for silicon wafers of various etching has been used for PS formation on the top surface conductivity type and doping level and for different duraof silicon solar cells in [3-51. The authors of those investiga- tions of the anodical treatment) from several hundreds of tions had evidenced the antireflection action of the PS layer nanometers to several micrometers. Thicknesses of both the and the increase of the solar cell quantum efficiency but, nanoporous and the microporous layers are also strongly probably because of non-optimized cell structure, had failed dependent on the conditions and the duration of the anodto obtain an overall improvement in the solar cell per- ization. Short anodization (electrical charge passed through the electrolyte smaller than 0.3 C/cm2) results only in a thin formance. (-100nm) nanoporous layer formation. After a much longer anodical treatment with anodization charges exceeding 10C/cmZ, the nanoporous layer becomes 1-2 prn thick, e-mail: [email protected] Physic4 Scripta T69 256 V . Patebutas, K . Grigoras and A. Krotkus it is accompanied by a formation of a microporous silicon layer reaching as deeply as for several tens of micrometers into the substrate. 3. Surface texturization Illumination Two different anodization regimes described above are employed in two different photovoltaic PS applications. Rough, sponge-like topology of the microporous layer obtained after a long anodical etching can be used for the solar cell surface texturization if the inhomogenities in that layer have dimensions comparable to the light wavelength. We had demonstrated that such a PS layer topology is most easily obtained on n-type silicon wafers. The effect of surface texturization by anodical etching has been monitored by measuring the changes in spectral transmittance and spectral photosensitivity of the wafers induced by PS layer formation. In the first of these experiments, spectral dependencies of the ratio T between the transmittances of a wafer covered with a PS layer and of a polished silicon wafer of exactly the same thickness were measured. Standard monochromator-photomultiplier arrangement with an integrating sphere collecting all radiation transmitted through the samples was used. Figure 1 presents the I -0.2' ' ' ' ' results of such measurements performed on n-type Si wafers 0.9 1.o 1.1 1.2 1.3 0.8 with one of their surfaces anodically etched for different times. A significant reduction in T is observable over a wide b) A Y Pm spectral region even for the wafer which was electrochemi- Fig.2. Photoconductivity measurements of the silicon-electrolytejunction: cally treated for shortest time duration. This reduction is (a) experimental set-up; (b) photoresponse spectra - dots correspond to caused by the isotropic distribution of the photons inside a experimental results, fullline - theoretical calculation. wafer with a rough texturized surface. It has been shown in [SI that in such a wafer, due to enhancement in a total internal reflection, many of the photons become trapped, with the microporous layer, the other half remaining polished. Photoresponse spectra were measured for the radiincreasing odds to be absorbed. It is evident that the effect of the light-trapping should be ation entering the sample in the vicinity of the PS layer and more pronounced in the lower absorption region where only far away from that region. Results of both measurements a small part of the incoming radiation is absorbed during a are shown in Fig. 2(b). The spectra coincide at short wave~ single pass through the wafer. Therefore, the surface tex- length region, but at the wavelengths around 1 . 1 the turization should result in an increased photosensitivity in photoresponsitivity is substantially higher in the region with the bandedge spectral region. Figure 2(a) illustrates the a texturized back surface. A simple calculation according to scheme of our spectral photoconductivity experiment. The [8] of the difference between a fraction of the incident light sample was placed into a bath filled with H,O-KCl electro- absorbed from a "trapped" ray and such a fraction absorbed lyte which provided a large-area, Schottky-like junction during a single pass of the ray across the wafer [full line on with the silicon wafer. The sample was illuminated through Fig. 2(b)] shows that it has a maximum in the spectral range its polished front face. A half of the rear face was covered where the observed photosensitivity enhancement is the highest. Similar set of experiments has been performed also on 0.7I , I p-type silicon wafers, however in this case the effect of the 2.5 min PS layer on the transmittance and the photoresponsitivity 0 10.0 min was much smaller due to too fine dimensions of the cavities in the microporous layer. m? ' I 0.4 1 ' I I . ' I :i 4. Antireflection coatings In the second investigation we had studied the effects of a thin nanoporous layer on the performance of photovoltaic 0.0 OP 3C-D 0.95 1.oo bbobou, 1.05 1 1.10 115 I 1 20 A* pm Fig. 1 . Spectral dependencies of the relative transmittivity of silicon wafers covered with PS-layer for different anodical etching durations. Physica Scripta T69 cells made from multicrystalline silicon. The cells had a 0.4 pm deep n+-p junction, randomly textured surfaces, and screen-printed silver contacts on the front side. Anodical etching conditions were optimised on (1 x 1)cm2 area cells, then the optimum conditions were applied to larger ( 5 x 5 ) and (10 x 10)cm2area cells. Porous Silicon Applications in Solar Cell Technology 257 Fig. 5. The mapping of the relative increase in photocurrent after anodical treatment of the solar cell performed with a focused He-Ne laser beam. at electrical charges of 0.05-0.09 C/cm2. The effect of the PS layer on the performance of large area solar cells was even more pronounced. Average increase of the energy conversion efficiency after PS layer manufacturing, which was evidenced for 10 (5 x 5)cm2 area cells was 30%. N Fig. 3. Relative changes of four main solar cell parameters as functions of the electrical charge transmitted during the anodization. Changes in four different (1 x 1)cm2 area solar cell parameters studied as functions of the electrical charge forced through the electrolyte during anodization are shown in Fig. 3. These parameters were determined from I-V measurements under 100mW/cm2 AM1.5 illumination. The effect of the anodization is most significant in the case of the short-circuit current which increased by more than 20%. Less distinct are the influences of the anodical etching on the fill-factor and the open-circuit voltage. The photovoltaic energy conversion efficiency affected by all three above mentioned parameters was found to increase by more than 20% Fig. 4. TEM micrograph of nanoporous silicon layer. Fig. 6. Oscilloscope traces of photocurrent relaxation. Excitation wavelengths (a) 1.06 pm and @) 0.53 p.(1- before, 2 - after etching). Physica Scripta T69 258 V . Patebutas, K. Grigoras and A. Krotkus In separate experiments the thickness and the refractive index of the PS layer have been determined. The values of these parameters for the coatings were found to be 80100nm and 1.8-2.2, respectively. As it can be seen from TEM micrograph presented on Fig. 4,the PS layer has optically smooth surfaces and its porous structure has very fine dimensions. Spectral reflectance measurements had evidenced the antireflective action of the PS layers. However, the antireflective action alone cannot explain the observed large improvement of the cell parameters, Spectral photoresponse, which was measured on selected spots of a (1 x 1)cm2 cell before and after anodization, in some cases increased by more than two times. Figure 5 shows the results of the photoresponse mapping of a (1 x 1)cm2 area cell performed with a focused He-Ne laser beam (wavelength of 630 nm). Relative changes on photoresponse induced by the anodization are different in various parts of the cell, which can be attributed to different levels of surface texturization and surface passivation on separate blocks of the multicrystalline material. A possible explanation of these observations can be the reduction of the surface recombination rate by hydrogen coverage of the PS surface and the diffusion and capture of H at the grain boundaries. The role of the surface passivation by PS is further evidenced by the transient photoconductivity measurements on the samples cut from the multicrystalline solar cells. Figure 6 shows the photocurrent decay traces measured before and after PS application. The samples were excited by the first (1.06 pm) and the second (0.53 pm) harmonics of a modelocked neodymium laser with a pulse duration of 3 ps. Transient photocurrent pulses are practically not affected by the anodization in the case of the infrared excitation with a large absorption length in silicon, but show a signscant Physica Scripta T69 View publication stats increase of a characteristic decay time after anodization when the sample is excited by a green radiation which is absorbed mainly at the surface region. 5. Conclusions In conclusion, we had demonstrated two different applications of PS layers in the solar cell technology. Microporous layers, which are formed on silicon surfaces after relatively long anodical etching, cause surface roughening enhancing the optical confinement in the cells and increasing quantum efficiency in the region of the silicon bandedge absorption. Thin nanoporous layers formed on already manufactured solar cells after a short anodical treatment provide antireflective coatings and additionally passivate the surfaces of the cells. References 1. Canham, L., Appl. Phys. Lett. 57, 1046 (1990). 2. Prasad, A., Balakrishnan, S., Jain, S. K. and Jain, G. C., J. Electrochem. Soc. 129,596 (1982). 3. Wei, G. P., Zheng, Y.M., Huang, 2.J., Li, Y.,Feng, J. and MO,Y.W., Solar Energy Materials and Solar Cells 35,3 19 (1994). 4. Menna, P., Di Francia, G. and La Ferrara, V., Solar Energy Materials and Solar Cells 37, 13 (1995). 5. Bastide, S., Cuniot, M., Williams,P., Le Quang, N., Sarti, D. and LevyClement, C., Proc. 12th European Photovoltaic Solar Energy Conference, Amsterdam, The Netherlands, 1994, p. 780. 6. Maruska, H. P., Namavar, F. and Kalkhoran, N. M., Appl. Phys. Lett. 61, 1338 (1992). 7. Pabbutas, V., Krotkus, A., Simkient, I. and Viselga, R., J. Appl. Phys. 77,2501 (1995). 8. Yablonovitch, E. and Cody, G. D., IEEE Transact. on Electr. Devices ED-29,300(1982).