21.3%-efficient n-type silicon solar cell with a full area rear TiOx/LiF/Al electron-selective contact

https://doi.org/10.1016/j.solmat.2019.110291Get rights and content

Highlights

  • Utilizing XPS measurement shows the mechanism of passivation while the usage of TEM confirms the effect of LiF.

  • The insertion of LiF reduces the contact resistivity considerably while keeping the good passivation.

  • It is the first time reaching a high efficiency of 21.3% for crystalline silicon solar cell with a full area contact.

Abstract

In this work, we investigate an efficient electron-selective passivating contact with TiOx/LiF/Al contact structure, which offers both low surface recombination and specific contact resistance. Optimized TiOx layer thickness of 4 nm provides high quality surface passivation, achieving minority carrier lifetime of 3.03 ms on 5 Ω cm n-type wafers, with a saturated current density J0 of 23 fA/cm2. In addition, inserting a 1 nm LiF between the 4 nm TiOx and Al reduces the contact resistivity to 18 mΩ cm2. The low contact resistivity of TiOx/LiF/Al contact is attributed to barrier reduction from the low work function of LiF/Al stack. A champion solar cell efficiency of 21.3% has been achieved for an n-type crystalline silicon device with a full-area rear TiOx/LiF/Al contact, demonstrating the excellent potential of this passivating contact for fabricating high-efficiency silicon solar cells.

Introduction

N-type crystalline silicon (n-c-Si) shows a higher tolerance to common metal impurities and no light-induced degradation, which results in higher bulk quality and greater stability compared to p-type crystalline silicon (p-c-Si) [1,2]. As such, many high efficiency silicon solar cells are fabricated on high-quality n-c-Si wafers. Two efficiency recorded n-c-Si solar cells of particular prominence are the 25.7% efficiency solar cell with tunneling oxide passivating contact [3], and 26.7%-efficient silicon heterojunction solar cells (SHJ) with the intrinsic and doped hydrogenated amorphous silicon layers [4]. One of the key factors in obtaining high efficiency is the application of carrier-selective contacts (CSC), which introduces excellent surface passivation and low contact resistivity. In recent years, metal oxide layers have been widely used in c-Si solar cells as CSCs on account of their appropriate band gap and suitable work function, which are able to transmit one type of carriers while blocking the other. For example, thin layers of high work function MoOx (~5.7 eV) have been used to substitute p-type hydrogenated amorphous silicon layers (p-a-Si:H) as hole selective layers on SHJ solar cells, achieving 22.5% efficiency [5]. Similarly, TiOx with relatively low work function (~4 eV) have been used as electron-selective layers on conventional silicon solar cells with tunneling oxides, achieving an efficiency of 22.1% [6]. These indicate that metal oxide layers show a great potential for achieving high performance c-Si solar cells.

A large part of the success of TiOx comes from its ability to provide high carrier selectivity, as shown in the work presented by Yang et al. in 2016 [7], obtaining the efficiency of 19.6% with TiOx/Al electron-selective contact. However, the TiOx/Al electron-selective contact behaves poorly in terms of contact resistivity. This limits the efficiency of solar cells, commonly attributed to a large barrier height. Allen et al. [8] investigated the research of replacing the Al with a relatively lower work function metal Ca (~2.9 eV) to acquire lower contact resistivity and realized an efficiency of 21.8% with a partial rear contact structure. As an alternative, a LiF/Al stack with a low work function (~2.8 eV) are widely used in organic photovoltaics to form effective electron transport layers and has recently attracted interest in silicon photovoltaics. It has been shown that the LiF/Al stack performs well in dopant-free asymmetric hetero-contact silicon solar cells, which achieved an efficiency of 19.4% [9]. Moreover, inserting a TiOx ultrathin layer between an i-a-Si:H passivating layer and LiF significantly enhanced the efficiency by 1.3% [10]. The latest work shows the TiOx/LiF/Al stack enhances the efficiency of cells, enabling over 23% with the partial rear contact structure, demonstrating the great potential of the TiOx/LiF/Al stack as electron-selective passivating contacts [11].

In this work, we investigated the combination of low work function LiF/Al stack with passivating TiOx layer to obtain both good surface passivation and low contact resistivity. The optimized electron-selective passivating contact was applied as full-area rear contact in c-Si solar cell, resulting in efficiency of 21.3%.

Section snippets

Experimental

N-type (100)-oriented Fz silicon wafers (400 μm, 5 Ω cm) were used for passivation measurement while n-type (100)-oriented Cz silicon wafers (290 μm, 0.1 Ω cm) were utilized for contact resistivity measurements. For surface passivation measurements, the wafers were damaged etched, RCA cleaned and dipped in dilute HF solution, prior to deposition of TiOx layers. Different thicknesses by atomic layer deposition (ALD, TFS 200, BENEQ, Finland) at 230 °C were deposited on both sides of wafers. The

Result and discussion

The interfacial layer features of TiOx/LiF/Al contact were characterized by TEM and XPS, as depicted in Fig. 1. In Fig. 1(a), different layers with measured thickness are visualized clearly by TEM due to different material structure. An interfacial layer with a thickness of about 0.25 nm could be observed between silicon substrate and TiOx, which has an impact on surface passivation and the corresponding contact resistance. In order to further study the interfacial layer between Si and TiOx,

Conclusion

This paper describes an electron-selective passivating contact TiOx/LiF/Al structure with both good passivation and low contact resistivity. Characterization of the interfacial layer of Si/TiOx shows high density of the mixture of SiOx and TiOx, which has a critical effect on the passivation and contact resistivity. The impact of the TiOx layer thickness, annealing times and annealing temperatures was investigated. It is clear that as-deposited 4 nm TiOx is the optimized thickness that results

Credit author statement

Wenjie Wang: Conceptualization, Investigation, Visualization, Writing - Original Draft, Writing - Review & Editing

Jian He: Investigation, Writing - Review & Editing.

Di Yan: Investigation, Writing - Review & Editing.

Chris Samundsett: Investigation, Writing - Review & Editing.

Sieu Pheng Phang: Investigation, Writing - Review & Editing.

Zengguang Huang: Investigation, Writing - Review & Editing.

Wenzhong Shen: Conceptualization, Supervision, Project administration, Funding acquisition, Writing -

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was supported by the Major State Basic Research Development Program of China (No. 2018YFB1500501), National Natural Science Foundation of China (Nos. 11834011, 11674225, 61774069 and 11974242), and China Scholarship Council (CSC) funding. J.H. and Y.W. acknowledge the support of the Australian Renewable Energy Agency (ARENA) Research and Development Program (2017/RND007).

References (17)

There are more references available in the full text version of this article.

Cited by (44)

  • Photon management in silicon photovoltaic cells: A critical review

    2024, Solar Energy Materials and Solar Cells
  • Characterization of sputtered nanocrystalline gallium nitride for electron selective contact of crystalline silicon solar cell

    2022, Thin Solid Films
    Citation Excerpt :

    On the other hand, zinc oxide (ZnO) [9], titanium oxide (TiOx) [10,11], lithium fluoride (LiF) [12], titanium nitride (TiNx) [13] were reported as the electron selective contact materials. In the case of electron selective contact layer, A SHJ solar cell with TiOx/LiF/Al stacking structure showed a conversion efficiency of over 21 % [14]. This stacking structure is important to obtain low contact resistance [15,16].

  • Towards low cost, industrially compatible silicon heterojunction solar cells using hybrid carrier selective passivating contacts

    2022, Optical Materials
    Citation Excerpt :

    The TiO2 layers inherently result in low contact resistivity in view of the reduced barrier height due to fermi level depinning. It has been experimentally shown in Ref. [29] that incorporation of TiO2 layers between the silicon and metal electrode reduces the contact resistivity by at least two orders of magnitude. A low work function difference between silicon and TiO2 further proves favorable for low contact resistivity and efficient front design.

  • Dopant-free materials for carrier-selective passivating contact solar cells: A review

    2022, Surfaces and Interfaces
    Citation Excerpt :

    There are many non-oxide materials which act as electron selective contacts with silicon [Table 9]. Among them there are alkali and alkaline earth metal fluorides such as LiF [46,49,129,133,137,138,193,197,198], MgF2 [48,232,73,167],KF [49], CaF2 [49], CsF [49] as well as other salts TiN [234,75], GaN [78], ZnSnN2 [77], GaP [79,236], ZnS [81] and In2S3 [243]. However Si solar cells featuring KF, CaF2, CsF as electron-selective contacts are yet to be fabricated.

View all citing articles on Scopus
View full text