Research Article
Front and Back contact engineering for high-efficient and low-cost hydrothermal derived Sb2(S, Se)3 solar cells by using FTO/SnO2 and carbon

https://doi.org/10.1016/j.jmst.2020.03.049Get rights and content

Abstract

Antimony chalcogenide Sb2(S, Se)3 is attracting a lot of attention as photovoltaic absorber owing to its rewarding photoelectric properties, low toxicity, and earth abundance. However, its device efficiency is still limited by the absorber material quality and device interface recombination. In this work, a fluorine-doped tin oxide (FTO) substrate with ultra-thin SnO2 layer and a low-cost stabilized carbon paste are introduced as a front and back contact layer respectively in Sb2(S, Se)3 based planar solar cells. Over 5.2 % efficiency is demonstrated in the structure of FTO/SnO2/CdS/Sb2(S, Se)3/Carbon/Ag, where the Sb2(S, Se)3 is prepared by hydrothermal technique. The complementary device physics characterizations reveal that the interfacial recombination between TCO and CdS is significantly suppressed by the introduction of ultra-thin SnO2 layer, which is profited from the leakage protection and bandgap offset engineering by its high resistivity and suitable conduction band minimum. Meanwhile, the successful adoption of the low-cost stabilized carbon as a back contact here shows an enormous potential to replace the conventional organic hole transport materials and noble metal. We hope this work can provide positive guidance to optimize Sb2(S, Se)3 based planar solar cells in the future.

Introduction

Recently Antimony chalcogenide Sb2(SxSe1-x)3 (0 ≤ x ≤ 1) including Sb2Se3, Sb2S3 and Sb2(S, Se)3 have received increasing attention as earth-abundant, non-toxic, and stable absorber materials. [1,2], The rewarding properties, such as tunable bandgap (1.10∼1.88 eV), high absorption coefficient (> 105 cm−1), intrinsic p-type doping and so on, make it potential to replace copper indium gallium selenide (CIGS), cadmium telluride (CdTe) thin films which contain rare or toxic elements. [[3], [4], [5]] To date, the power conversion efficiency (PCE) of superstrated Sb2Se3 device (analogous CdTe configuration) has achieved 7.6 %. The substrated device (analogous CIGS configuration) was boosted to 9.2 %, and the sensitized Sb2S3 device has obtained 7.5 %. [[6], [7], [8]] While the PCE of mixed sulfoselenide Sb2(S, Se)3, possessing the tunable bandgap for ideal bandgap predicted by Shockley-Queisser theory, is lagging compared with that of Sb2Se3 and Sb2S3 solar cells. In 2014, Seok group had obtained a sensitized Sb2(S, Se)3 device by the diffusion-reaction of Sb2S3/Sb2Se3 stack, which delivered a record PCE of 6.6 % for Sb2(S, Se)3 solar cells. [9] Since that, many efforts have been put into elevating the device PCE for years. These mainly focus on high-quality Sb2(S, Se)3 growth for improving the lifetime of photo-generated charge carriers. For instance, Chen et al. have developed a facile one-step solution deposition of Sb2(S, Se)3 film for device application. [10] Tang and Song et al. adopted in-situ sulfurization of evaporated Sb2Se3 and rapid thermal evaporation (RTE) routes to form Sb2(S, Se)3 solar cells. [11] Nair et al. also produced Sb2(S, Se)3 thin film by thermal evaporation of Sb2S3 and Sb2Se3 powders. [12] Our group has introduced a low-cost hydrothermal synthesis combined with post selenization for the preparation of Sb2(S, Se)3 thin film, where the in-situ deposition of Sb2S3 precursor occurred on CdS buffer layer. The resulted high-quality PN junction delivered a PCE of 6.14 % when the organic hole transport layer (HTL, Spiro-OMeTA) was used [13].

To further boost the Sb2(S, Se)3 device efficiency, another crucially important interface issue should be brought to the forefront. Because the device interface determines whether high-quality absorber materials can transfer to final energy conversion efficiency. While the improved properties of absorber materials have been intensely studied, the interface engineering in Sb2(S, Se)3 solar cells are still in its initial stage. For example, Chen and Tang's groups have modified the back contact interface of Sb2(S,Se)3/Au by inserting an organic HTL, in which the HTL such as Spiro-OMeTA, CuSCN or CZ-TA can suppress the back surface recombination and assist carrier collection. [10,14,15] Notably, all of these materials are too much expensive to extend in industry. Another interface engineering, namely the engineering of Sb2(S, Se)3/buffer layer interface, is commonly realized through composition regulation or introducing a double buffer layer. Mai and Chen's groups have reported that the interface engineering of absorber/buffer layer via doping CdS with oxygen and indium respectively, which could passivate interface defects and lead to enhanced device performance. [16,17] Besides, the double buffer layers, such as TiO2/CdS, ZnO/CdS and SnO2/CdS, have also been proven to effectively optimize bandgap alignment for antimony chalcogenide solar cells. [[18], [19], [20]] But the preparation of double buffer layer will increase the complexity of device fabrication. Furthermore, the common preparation method of TiO2, ZnO and SnO2 buffer layers is the spin coating, which has been demonstrated not suitable for industrialization when compared with the proven vacuum deposition process. For example, the commercial sputtering or chemical vapor deposition (CVD) routes are capable of producing higher quality thin film with pure composition, compact morphology and strong adhesion compared with the spin coating method. [21]

To date, an unnoticeable interface, conductive front contact/buffer layer, is still studied scarcely. In the Sb2(S, Se)3 solar cells with facile superstrate configuration, tin-doped indium oxide (ITO) and fluorine-doped tin oxide (FTO) have been traditionally used as transparent conducting oxide (TCO). However, little advice has been given about how to passivate the recombination of TCO/buffer layer interface, which remains an open question in Sb2(S, Se)3 solar cells. In this work, the front contact engineering has been executed by using FTO substrate with an ultra-thin SnO2 layer. The high resistivity of SnO2 layer can passivate the TCO/CdS interface and suspend the interface recombination. Thus, the Sb2(S, Se)3 device with modified FTO possesses a high built-in voltage and lower recombination interface, which contributes to the increased VOC. Meanwhile, the low-cost stabilized carbon was examined as a potential electrode for Sb2(S, Se)3 solar cells, which achieved a PCE of 5.2 %.

Section snippets

Device fabrication process

The illustrative diagram of Sb2(S, Se)3 thin-film solar cells can be seen in Fig. 1. Firstly, as shown in Fig. 1a, a CVD assisted fabrication method of the SnO2 layer on the FTO substrate was introduced and the processes were executed by HuaiAn Yaoke Optoelectronics Co. Ltd. To engineer the front contact, a ∼30 nm SnO2 layer was sequentially coated on FTO substrate during the CVD process. Then, before the deposition of CdS layer, the substrate was cleaned using detergent, acetone, alcohol and

Photovoltaic performance

Fig. 2a shows J-V curves of the champion devices of Sb2(S, Se)3 with FTO and FTO/SnO2 TCO tested under AM1.5 illumination. The detail performance metrics of the devices are labeled in the inset of Fig. 2a. It can be clearly observed that solar cells with SnO2 coated TCO exhibit superior performance compared to that of the solar cells with FTO, which exhibits a PCE of 3.1 % for the FTO-based device and 5.2 % for the FTO/SnO2-based device, respectively. The improvement in the power conversion

Conclusion

In summary, the interface engineering between TCO and buffer layer was effectively realized by the commercial FTO with ultra-thin SnO2 coating. The SnO2 layer relieves not only the leakage path but also the bandgap offset. The complementary device physics characterizations disclose that the front contact engineering can suspend the interfacial recombination significantly, which finally enhances the device performance. In addition, a low-cost stabilized carbon paste was painted as a back contact

Acknowledgements

This work was supported by National Natural Science Foundation of China (Grant No. 61974028), Fujian Normal University (FNU) Training Program of Innovation and Enterpreneurship for Undergraduates (cxxl-2019135, 2019140, 2019143). We also acknowledge the support from HuaiAn Yaoke Optoelectronics Co. Ltd.

References (35)

  • Y. Yin et al.

    Sci. Bull.

    (2019)
  • G. Liang et al.

    Sol. Energy Mater. Sol. Cells

    (2017)
  • Z. Li et al.

    Nat. Commun.

    (2019)
  • D. Li et al.

    Nano Energy

    (2018)
  • C. Ou et al.

    Sol. Energy Mater. Sol. Cells

    (2019)
  • J. Tao et al.

    Nano Energy

    (2019)
  • G.S. Park et al.

    Thin Solid Films

    (2000)
  • T.T. Nguyen et al.

    Ceram. Int.

    (2019)
  • J. Szanyi

    Appl. Surf. Sci.

    (2002)
  • H.H. Güllü et al.

    J. Alloys Compd.

    (2013)
  • W. Wang et al.

    Electrochim. Acta

    (2018)
  • W. Melitz et al.

    Surf. Sci. Rep.

    (2011)
  • Y. Zhou et al.

    Nat. Photon.

    (2015)
  • V. Steinmann et al.

    Nat. Photon.

    (2015)
  • X. Chen et al.

    J. Mater. Chem. C

    (2017)
  • X. Wen et al.

    Nat. Commun.

    (2018)
  • Y.C. Choi et al.

    Adv. Funct. Mater.

    (2014)
  • Cited by (30)

    • Construction of CdS/Sb<inf>2</inf>Se<inf>3</inf> planar heterojunction by full vapor transport deposition

      2022, Vacuum
      Citation Excerpt :

      In addition, the band gap was calculated by combining UV data of CdS (Fig. S2) and EQE data of Sb2Se3 (1.2 eV, from Fig. S4 and Fig. S5), Sb2Se3 grown on different substrates had similar valence band maximum positions of 4.91 eV and 4.95 eV, respectively. Finally, the energy level data of ITO/TiO2/Carbon/Ag were referred to the literatures [30,31]. As shown in the energy level sketch of devices of Fig. 5i, CdS-VTD and Sb2Se3 own a more appropriate levels alignment compared with CdS-CBD, which will reduce the contact barrier and then facilitate the photo-generated carrier collection.

    • Thermal evaporation–deposited hexagonal CdS buffer layer with improved quality, enlarged band gap, and reduced band gap offset to boost performance of Sb<inf>2</inf>(S,Se)<inf>3</inf> solar cells

      2022, Journal of Alloys and Compounds
      Citation Excerpt :

      According to the calculated energy levels (Fig. 7d), the CBM of the Sb2(S,Se)3 and those of both CdS films form a cascade alignment. However, it is noteworthy that the CBM offset value between FTO and CBD-CdS (0.76 eV) may promote recombination and reduce VOC [8,16]. This problem can be partly resolved by applying a TED-CdS buffer layer because the corresponding CBM offset value decreases to 0.60 eV.

    • A facile Ion-Exchange assisted chemical bath deposition of CuSbS<inf>2</inf> for thin film solar cells

      2022, Solar Energy
      Citation Excerpt :

      Then we obtain a complete band diagram of the CuSbS2 device as shown in the Fig. 12c. The band levels of other functional layer are referred to the previous work (Yao et al., 2020). It can be seen that CuSbS2 energy levels can form excellent matching energy levels with TiO2/CdS double buffer layer and low-cost stable carbon, which makes it easier for the generated carriers to be transported to the electrodes (Wu et al., 2022).

    • Solution-processed antimony chalcogenides based thin film solar cells: A brief overview of recent developments

      2022, Solar Energy
      Citation Excerpt :

      The improvement in PCE was associated with the position of conduction band maximum of CdS which could ease the carrier transport by improving VOC (Wang et al., 2019). To enhance the carrier extraction at the two terminals of antimony sulfoselenide solar cell, front and back contact engineering was carried out by Yao et al. (Yao et al., 2020). In the device with structure FTO/SnO2/CdS/Sb2(S,Se)3/Carbon/Ag, a thin SnO2 layer (30 nm) was deposited by chemical vapor deposition (CVD) which helped in reducing the bandgap offset and leakage path.

    View all citing articles on Scopus
    View full text