Physical criteria for the interface passivation layer in hydrogenated amorphouscrystalline silicon heterojunction solar cell-Lei Zhao-中国科学院.pdf

Journal of Physics D: Applied Physics ACCEPTED MANUSCRIPT Physical criteria for the interface passivation layer in hydrogenated amorphous/crystalline silicon heterojunction solar cell To cite this article before publication: Lei Zhao et al 2017 J. Phys. D: Appl. Phys. in press https://doi.org/10.1088/1361-6463/aa9ecd Manuscript version: Accepted Manuscript Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process, and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors” This Accepted Manuscript is © 2017 IOP Publishing Ltd. During the embargo period (the 12 month period from the publication of the Version of Record of this article), the Accepted Manuscript is fully protected by copyright and cannot be reused or reposted elsewhere. As the Version of Record of this article is going to be / has been published on a subscription basis, this Accepted Manuscript is available for reuse under a CC BY-NC-ND 3.0 licence after the 12 month embargo period. After the embargo period, everyone is permitted to use copy and redistribute this article for non-commercial purposes only, provided that they adhere to all the terms of the licence https://creativecommons.org/licences/by-nc-nd/3.0 Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions will likely be required. All third party content is fully copyright protected, unless specifically stated otherwise in the figure caption in the Version of Record. View the article online for updates and enhancements. This content was downloaded from IP address 131.170.21.110 on 06/12/2017 at 03:19Physical criteria for the interface passivation layer in hydrogenated amorphous/crystalline silicon heterojunction solar cell Lei Zhao *,1,2 , Guanghong Wang 1,2 , Hongwei Diao 1 , Wenjing Wang 1,2 1 Key Laboratory of Solar Thermal Energy and Photovoltaic System of Chinese Academy of Sciences, Institute of Electrical Engineering, the Chinese Academy of Sciences, Beijing 100190, China 2 University of Chinese Academy of Sciences, Beijing 100049, China Corresponding author: e-mail zhaolei@mail.iee.ac.cn, Phone: +86 10 82547042, Fax: +86 10 82547041 Abstract: AFORS-HET simulation was utilized to explore the physical criteria for the passivation layer in hydrogenated amorphous/crystalline silicon heterojunction (SHJ) solar cell by systematically investigating the solar cell current density–voltage (J–V) performance as a function of the interface defect density (Dit) at the passivation layer/c-Si hetero-interface, the thickness (t) of the passivation layer, the bandgap (Eg) of the passivation layer, and the density of dangling bond states (Ddb)/band tail states (Dbt) in the band gap of the passivation layer. The corresponding impact regulations were presented clearly. Except for Dit, the impacts of Ddb, Dbt and Eg are strongly dependent on the passivation layer thickness t. While t is smaller than 4–5 nm, the solar cell performance is less sensitive to the variation of Ddb, Dbt and Eg. Low Dit at the a-Si:H/c-Si interface and small thickness t are the critical criteria for the passivation layer in such a case. However, if t has to be relatively larger, the microstructure, i.e. the material quality, including Ddb, Dbt and Eg, of the passivation layer should be controlled carefully. The involved mechanisms were analyzed and some applicable methods to prepare the passivation layer were further proposed. Keywords: passivation layer; amorphous/crystalline silicon heterojunction; solar cell; simulation 1. Introduction Recently, more and more photovoltaic practitioners pay much attention to the hydrogenated amorphous/crystalline silicon (a-Si:H/c-Si) heterojunction (SHJ) solar cell due to its high efficiency achievable in practice and very simple fabrication process [1–4]. The solar cell is basically produced by low temperature (~200 ºC) deposition of thin a-Si:H emitter and back surface field (BSF) layers onto both sides of the c-Si wafer-base. An intrinsic a-Si:H thin layer is inserted between the doped a-Si:H layer and the c-Si wafer to passivate the a-Si:H/c-Si hetero-interface. By utilizing this architecture, Panasonic achieved a conversion efficiency of 24.7% with the open-circuit voltage (VOC) of up to 750 mV [2]. Kaneka further combined SHJ and interdigitated back contact (IBC) schemes together and thus increased the conversion efficiency to be higher than 26% [4]. How to make the passivation layer becomes a critical key for achieving the high-performance SHJ solar cell. Usually, thin film silicon layers are deposited by two representative methods: hot wire chemical vapor deposition (HWCVD) and plasma enhanced chemical vapor deposition (PECVD) [5–8]. By changing the deposition conditions, thin film silicon layers with different microstructure can be prepared, and result in different passivation effects. Compared with microcrystalline silicon (μc-Si) or nanocrystalline silicon (nc-Si), it was found that a-Si:H had better passivation performance due to its relatively higher hydrogen content (CH). Hydrogen atoms in the form of monohydride (SiH) were more preferred than those in the form of dihydride (SiH2) for the passivation [9]. For excellent passivation layer, the crystallinity should be lower than 10% [10] and CH be 5.0 - 8.5% [11]. Many Page 1 of 18 AUTHOR SUBMITTED MANUSCRIPT - JPhysD-114942.R2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Accepted Manuscriptresearchers achieved excellent passivation performance when the passivation layer was deposited just close to the onset of the amorphous-to-crystalline transition phase [12, 13]. The silane depletion fraction was proposed as the indicator for the passivation. Good passivation could be obtained from highly depleted silane plasma [14]. Further, the passivation effect was found to be largely related to the passivation layer thickness. Generally, the thin film silicon layer formation started with nucleation and separate island growth. Then the islands coalesced into a closed film gradually, followed by a steady state growth [15, 16]. The initial island growth could induce a SiH2-rich interface layer with poor network and high density of interface states (Dit). During the subsequent film growth, the disordered interface layer would relax towards equilibrium and hydrogen atoms from the deposition could also diffuse onto the interface. Thus Dit could be reduced gradually until the film growth reached the steady state [17]. Besides, it was found that the recombination at the external surface of the passivation layer also played an important role for the thickness dependence of the passivation effect. While the passivation layer became quite thick, the passivation effect was predominantly determined by the bulk quality of the passivation layer [18]. The thickness reduction of the transition-phase a-Si:H passivation layer could result in a continuous VOC drop for the solar cell. However, an underdense passivation layer demonstrated an opposite trend. Higher VOC was obtained when it was ultrathin [19]. The passivation layer thickness could also induce the variation of band offsets at the passivation layer/c-Si hetero-interface [20]. So the passivation is not a simple interface behavior. The microscopic structural configuration of the thin film silicon bulk also needs be considered. Several groups proposed a-Si:H dual layers to improve the passivation with an interfacial layer deposited at relatively low hydrogen dilution rate (R) and a dense capping layer deposited at relatively high R. R was defined as the gas flow rate ratio of hydrogen to silane (R = [H2]/[SiH4]) [21, 22]. Some assistance treatments were also proposed to improve the passivation effect of the as-deposited layers. Multiple hydrogen plasma treatments during the film deposition could improve the hydrogenation of the a-Si:H/c-Si hetero-interface greatly. Hydrogen atoms from the hydrogen plasma not only promoted the crystallization of the as-deposited film but also diffused onto the interface to saturate the dangling bonds. Thus, the initial deposited film was allowed to involve more amorphous phase or be amorphous completely, which was helpful to avoid the defective epitaxial growth [23]. The passivation effect improvement by post-deposition hydrogen plasma treatment was also demonstrated [24]. Post-deposition annealing was another effective treatment to improve the passivation performance, particularly for low temperature-grown a-Si:H layers. The inhomogeneity of films grown at low temperature could become more homogeneous by film relaxation after post-annealing [25, 26]. The as-deposited Dit at the a-Si:H/c-Si hetero-interface was determined by the local network structure in a nonequilibrium state. After enough annealing, the final Dit was defined by the relaxed bulk a-Si:H network strain [27]. By combining post-deposition hydrogen plasma and thermal annealing treatments together, very high minority carrier lifetime over 15 ms on n-type float zone c-Si wafer was demonstrated [28]. It was further found that the passivation improvement by thermal annealing was at a metastable state, which could go back to the initial state after some time. Hydrogen plasma treatment resulted in a more stable state, which could be kept for a long time [29]. In brief, so many aspects should be concerned for the passivation layer. It seems that to achieve excellent passivation layer is quite complicated. Someone may be confused what the best passivation layer should be like and whether there is a simple method for its preparation. Since the passivation Page 2 of 18 AUTHOR SUBMITTED MANUSCRIPT - JPhysD-114942.R2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Accepted Manuscriptlayer is fully inserted in the p-n heterojunction of the SHJ solar cell, besides Dit, the material quality of the passivation layer together with the doped emitter also affects the solar cell performance. So, the optimization of the passivation layer should be carried out in the complete SHJ solar cell. Only measuring the lifetime obtained by single passivation layer is obviously not enough. Hence, the influence of Dit at the hetero-interface, the thickness (t) and the bandgap (Eg) of the passivation layer, as well as the dangling bond state density (Ddb) / band tail state density (Dbt) in the passivation layer on the SHJ solar cell performance were studied systematically here by utilizing AFORS-HET simulation. As a result, some physical criteria for the passivation layer were explored. The involved mechanisms were also analyzed elaborately. Based on the findings, some applicable methods for the passivation layer preparation were proposed. AFORS-HET is an open source on demand program, provided by Helmholtz-Zentrum Berlin für Materialien und Energie [30]. It solves the one dimensional semiconductor equations (Poisson′s equation, the transport and continuity equation for electrons and holes) with the help of finite differences. By utilizing suitable optical model, recombination model, carrier transport model, and metallic contact model, a variety of the solar cell characteristics can be simulated. AFORS-HET has been utilized by many researchers to optimize the solar cell performance and the results are well-accepted [31–34]. 2. Solar cell structure and simulation As shown in Fig. 1(a), the simulated SHJ solar cell is based on a pyramid-textured c-Si(n) wafer. The main electrical components are a-Si:H(p)/the passivation layer(i)/c-Si(n)/a-Si:H(i)/a-Si:H(n + ), i.e., the investigation is focused on the front passivation layer due to the solar cell performance is mainly determined by the p/i/n heterojunction. The front contact is composed of metal gridlines and transparent conductive oxide (TCO). The back contact is metal. AFORS-HET version 2.5 was utilized in the simulation. Both front and back contacts were assumed as flatband with the electron/hole surface recombination velocity of 10 7 cm/s. The carrier generation rate was calculated by using Lambert-Beer optical model with the path length factor as one. The optical loss from front surface reflection, grid shadowing, and TCO absorption was set as 8%. Fig. 1 (a) The simulated solar cell structure, (b) the gap defect state distributions in different deposited layers and the c-Si(n) wafer-base. Page 3 of 18 AUTHOR SUBMITTED MANUSCRIPT - JPhysD-114942.R2 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Accepted ManuscriptAs shown in Fig. 1(b), the band gap defects in the a-Si:H(p) emitter, the c-Si(n) base, a-Si:H(i) layer, and the a-Si:H(n + ) back surface field (BSF) were set as the default states in AFORS-HET. For the concerned passivation layer (i) between the emitter and the base, the same band structure distribution as that of a-Si:H was assumed. When different Eg was considered for the passivation layer, its electron affinity was adjusted correspondingly to keep its band alignment with the c-Si base constant, i.e., the ratio of the conduction band offset to the valance band offset was unchanged. At the same time, the gap defects in the passivation layer were also assumed to have the same distribution as that in a-Si:H(i) initially. Except that the defect in c-Si(n) was at a single level, the defects in all the deposited layers contained conduction/valence band tail states with exponential distribution and dangling bond states with Gauss distribution as donor/acceptor-like in the lower/upper part of the band gap. For the passivation layer (i)/c-Si(n) and c-Si(n)/a-Si:H(i) interfaces, continuous type defects were assumed for Dit, distributing in the layer gap as donor/acceptor-like in the lower/upper part of the c-Si gap. The initial Dit on both interfaces was set as 10 11 cm -2 /eV, w hich could be obtained via excellent chemical cleaning and passivation [35, 36]. Numerical model for the interface transport was set as thermionic emission. At the same time, the options to activate tunneling at hetero-interface and to calculate Richardson constant from effective mass were selected