• Corpus ID: 235313555

Abnormal Staebler-Wronski effect of amorphous silicon

@inproceedings{Liu2021AbnormalSE,
  title={Abnormal Staebler-Wronski effect of amorphous silicon},
  author={Wenzhu Liu and Jianhua Shi and Liping Zhang and Anjun Han and Shenglei Huang and Xiaodong Li and Jun Peng and Yuhao Yang and Ya-Jie Gao and Jianguo Yu and Kaiyun Jiang and Xinbo Yang and Zhenfei Li and Junlin Du and Xin Song and Youlin Yu and Zhixin Ma and Yubo Yao and Haichuan Zhang and Lujia Xu and Jingxuan Kang and Yichao Xie and Hanyuan Liu and Fanying Meng and Fr{\'e}d{\'e}ric Laquai and Zengfeng Di and Zhengxin Liu},
  year={2021}
}

• Wenzhu Liu, Jianhua Shi, +24 authors Zhengxin Liu
• Published 3 June 2021
• Materials Science

Great achievements in last five years, such as record-efficient amorphous/crystalline silicon heterojunction (SHJ) solar cells and cutting-edge perovskite/SHJ tandem solar cells, place hydrogenated amorphous silicon (a-Si:H) at the forefront of emerging photovoltaics. Due to the extremely low doping efficiency of trivalent boron (B) in amorphous tetravalent silicon, light harvesting of aforementioned devices are limited by their fill factors (FF), which is a direct metric of the charge carrier transport. It is challenging but crucial to develop highly conductive doped a-Si:H for minimizing the FF losses. Here we report intensive light soaking can efficiently boost the dark conductance (dark) of B-doped a-Si:H “thin” films, which is an abnormal Staebler-Wronski effect. By implementing this abnormal effect to SHJ solar cells, we achieve a certificated power conversion efficiency (PCE) of 25.18% (26.05% on designated area) with FF of 85.42% on a 244.63-cm 2 wafer. This PCE is one of the highest reported values for total-area “top/rear” contact silicon solar cells. The FF reaches 98.30 per cent of its Shockley-Queisser limit. Hydrogenated amorphous silicon (a-Si:H) is a technologically important semiconductor for transistors, batteries, hydrogen production and solar cells (1-4). It has a long history of use in photovoltaic applications because it offers low defect density and tunable conduction type (5-7). These optoelectronic advantages strongly rely on configurations of H and Si in the three-dimensional space, described by the radial distribution function (8), so precise control of its microscopic structure (9-11) is a critical factor to achieving good devices. As B is a trivalent element, it is challenging to establish four-coordinated BSi4 compounds in the disordered a-Si:H matrix, reported approaches focusing on eliminating invalid SixBHy doping configurations (Fig. S1) include optimizing the B2H6 flow rate and post-deposition annealing. However, a lack of understanding of the complicated conductive mechanism of B-doped a-Si:H (p-a-Si:H) has prevented to develop the full potential of relevant devices. Here we report intensive light soaking is a “fast” means to improving dark of p-a-Si:H thin films that enables highly efficient SHJ solar cells. Since 1977, light soaking on “thick” a-Si:H films has been widely studied in the research field of a-Si:H thin-film solar cells, but only a small number of works pay attention to its effect on “thin” a-Si:H films, particularly in the research field of SHJ solar cells. Although a few groups report light soaking improves FF of SHJ solar cells by a magnitude of ~0.7%abs (12), the hidden mechanisms are still very unclear. We use in-situ methods to monitor the time-dependent changes of p-a-Si:H thin films during illuminations. Thin p-a-Si:H films are deposited on silica glasses, followed by evaporating silver strips to form the transmission-line-model structure. The in-situ current-voltage data (Fig. 1A) shows that dark of the p-a-Si:H thin film steadily increases during one-sun illumination, reaching dark/dark0 ~ 4.71 after 30 minutes. This phenomenon is strikingly contrast to the light-induced degradation of dark observed in “thick” intrinsic, p-type and n-type a-Si:H films (13-15), supporting the perspective that accumulated stress in “thick” films plays an important role in the Staebler-Wronski effect (16), because the maximum stress is roughly proportional to the thickness of the film. After turning off the illumination, the dark gradually decays (close) to its initial value after more than 1000 minutes (Fig. 1B). Such a decay behavior well fits to a combination of Debye model and Williams-Watts model (Fig. 1C),       WW D dark D D WW WW exp exp t t t                      , (1) D, WW and D, WW are constant coefficients and characteristic time constants of the Debye model and the Williams-Watts model respectively. Detailed parameters are summarized in Table S1. The Debye model with D = 1 describes free diffusion, while the Williams-Watts model with 0 < WW < 1 describes a continuous-time random walk composed of an alternation of “steps” and “pauses” (17). Examples of the Williams-Watts model include spin-correlation in Cu-Mn and Ag-Mn, spin-glass transition in BaFe12O19, dielectric relaxation in K0.3MoO3, spin-lattice relaxation in κ-(ET)2Cu[N(CN)2]Br, specific heat in FexZr1-x and H relaxation in a-Si (18-23). The good fitting in Fig. 1C suggests the abnormal Staebler-Wronski effect is mediated by two independent mechanisms that respectively control the “fast” Debye relaxation and the “slow” Williams-Watts relaxation. To determine the implicit mechanisms, we investigate the H configurations in p-a-Si:H thin films by time-of-flight secondary ion mass spectrometry (ToF-SIMS). The H spectra (Fig. 2A) show that 30-minute annealing at 180°C only slightly changes the H content in intrinsic a-Si:H (i-a-Si:H), in contrast, the same annealing process expels at least ~21.3% of the H content out of the p-a-Si:H. As shown in Fig. S2, ToF-SIMS spectra also reveal that room-temperature oxidation of an i/p-a-Si:H stack hardly changes the H content in the i-a-Si:H film, however, the same oxidation process getters ~17.1% of the H content in p-a-Si:H from inside to surface. Based on these findings, we conclude that the B doping plays a crucial part in the formation of metastable H configurations in p-a-Si:H. Next we combine molecular dynamics and ab initio dynamics to search the possible binding configurations of aforementioned metastable H. Ab initio dynamics observes displacement of four-coordinated Si atoms by B atoms shortens the bonds from ~2.35Å to ~2.07Å (Fig. S3), well consistent with the results of Pandey et al (24). Further simulations demonstrate these BSi4 sites have a large probability to capture H atoms to form metastable BHSi configurations when diffusive H atoms pass by (Fig. S4), which is in agreement with the reported NMR signal (25) and relevant simulations (26). As a consequence, conductance of p-a-Si:H is expected to decline due to reduction in the quantity of BSi4 (24). Fortunately, transition-state surveys (Fig. 2B) prove that the hopping barriers of H from BHSi and SiHSi to adjacent SiHSi are merely 0.99  0.22 eV and 0.40  0.11 eV respectively, and the energy required to emit H from SiHSi is only ~0.97  0.26 eV. These reactions show noticeably lower barriers than that required to emit H from SiH or BH single bonds (Fig. 2C), thus account for why there exists much more metastable H configurations in p-a-Si:H than that in i-a-Si:H (inferred from Fig. 2A). When intensive light soaking provides photons with energy over ~1.0 eV, huge amounts of Si3BH should be converted to BSi4 via the H hopping illustrated in Fig. 2B, resulting in improvement of dark as has been confirmed in Fig. 1A. This microscopic dynamics is supported by the evidence of light-induced formation of SiHSi configurations from low-temperature infrared spectroscopy (27). This mechanistic understanding is also evident in optoelectronic analysis. Olibet el al. demonstrates that effective minority carrier lifetime (eff) of passivated c-Si is dominated by field passivation and chemical passivation respectively at low injection level and high injection level (28). We prepare symmetric passivated structures of p-a-Si:H/i-a-Si:H/n-c-Si/i-a-Si:H/p-a-Si:H and i-a-Si:H/n-c-Si/i-a-Si:H, and measure their injection-dependent eff before and after two-hour light soaking under one-sun illumination, as well as that after 15-minute annealing at 180°C. Interestingly, the right graph in Fig. 2D shows the eff of p-a-Si:H/i-a-Si:H/n-c-Si/i-a-Si:H/p-a-Si:H increases substantially at low injection level, indicating the field passivation is improved with the greatest contribution from better B doping, also intriguing is that the low-injection-level eff return to their initial values after a 15-minute brief annealing at 180°C; in contrast, the eff remain unchanged at the high injection level, suggesting the light soaking has negligible impact on the chemical passivation. As a comparison, the left graph in Fig. 2D shows the eff of i-a-Si:H/n-c-Si/i-a-Si:H remains constant after either light soaking or annealing. Moreover, ultrafast and broadband transient absorption (TA) signals (Fig. 2E) indicate the light soaking also increases the mobility of photon-generated carriers from 7.10×10 cm·V·s to 1.81×10 cm·V·s. According to Sinton and other researcher (29), pseudo FF (pFF) of silicon solar cells includes the effect of space charge recombination, but excludes the influence of resistance. We fabricate devices with structure of Ag/IWO/p-a-Si:H/i-a-Si:H/n-c-Si/i-a-Si:H/n-a-Si:H/IWO/Ag, then probe their pFF before and after two-hour light soaking under one-sun illumination, as well as that after 15-minute annealing at 180°C. Fig. 2F shows the pFF maintains ~86.4% regardless of the light soaking and annealing. This again definitely demonstrates the light soaking makes no difference to the chemical passivation. Consideration of the dark (Fig. 1), eff (Fig. 2D), carrier mobility (Fig. 2E) and pFF (Fig. 2F) leads to the conclusion that the abnormal Staebler-Wronski effect stems from activation of B doping rather than reduction of defects in the p-a-Si:H. In this regard, we further ascribe the dark decay of dark in Fig. 1B to the detrimental reconstruction of BHSi configurations, because binding energy of H in BHSi shows ~0.59 eV higher over that in SiHSi. In accordance, the “fast” Debye relaxation and the “slow” Williams-Watts relaxation (Fig. 1C) are attributed to incorporation of “fast” diffusive H and “slow” hopping H into the BSi bonds respectively, forming invalid B doping that negatively effects on the dark as has been confirmed in Fig. 1B. Next we use this abnormal Staebler-Wronski effect to improve performance of SHJ solar cells. We fabricate devi