Elsevier

Thin Solid Films

Volume 660, 30 August 2018, Pages 247-252
Thin Solid Films

Fabrication of high band gap kesterite solar cell absorber materials for tandem applications

https://doi.org/10.1016/j.tsf.2018.06.038Get rights and content

Highlights

  • We created CZGSSe absorber layers with sulfur inclusion going from 30% to 100%.

  • These absorbers have a respective band gap of 1.45 eV to 2.0 eV.

  • The results from these annealing procedures were very well reproducible.

  • The sulfur inclusion is very slow. The first 30 min both CZGSe and CZGSSe coexist.

  • It is impossible to have a uniform absorber layer with less than 30% S inclusion.

Abstract

Using the thermal annealing of evaporated metallic precursors in successive H2Se and H2S atmospheres, it was possible to reproducibly manufacture kesterite absorber material for solar cell applications with a sulfur content varying from 30% to 100%. Respective band gaps for these sulfur inclusions were measured at approximately 1.45 eV and 2.0 eV. A recipe was devised for which results could be reproduced within an error margin of ±5% and the influence of the H2S pressure during the post sulfurization was negligible on all measurable and observable parameters. The evolution of the S/Se ratio in the sample was observed to be linearly dependent on the annealing time. It was also observed that at very early stages of the post-sulfurization, both the original Cu2(Zn,Ge)Se4 (CZGSe) and a primary Cu2(Zn,Ge)(S,Se)4 (CZGSSe) phase with a sulfur inclusion of ~30% coexist in the sample. The (112) x-ray diffraction (XRD) reflection of the CZGSe phase progressively disappears in favor for the first mixed CZGSSe phase. Using grazing incidence-XRD, the S/Se ratio was shown to be inhomogeneous. Indeed, the XRD measurement of the top layers led to the calculation of higher sulfur inclusions than was the case when measuring the bulk material. Top-scanning electron microscopy (SEM) as well as cross-SEM measurements were taken in order to determine the impact of the sulfur inclusion on the crystal growth and the overall quality of the produced absorber layers. The obtained images revealed a reduction in crystal size and the appearance of numerous holes in the layer as the S/Se ratio is increased.

Introduction

The total amount of energy generated by solar energy has been very rapidly increasing these past few years. The global capacity has been multiplied by 4 in the last five years [1] and is expected to keep increasing in the coming years with the emergence of renewable energy favorable policies in many countries on the planet. It will be necessary, in the years to come, to be able to consistently produce large amounts of highly efficient solar cells to keep up with the demand. Traditional Si-based photovoltaic (PV) setups might still occupy an important place on the global market but could very well be replaced by Thin Film (TF) PV technologies with the emergence of building-integrated (BI) PV solutions. Indeed, TFPV can be implemented on a wide variety of solid or flexible substrates, which makes them interesting for BIPV applications. Classical TFPV materials like Cu(In,Ga)(S,Se)2 (CIGS) and CdTe, on the other hand, could be limited in the long run, due to scarcity of elements like In and Ga or the toxicity of Cd.

Due to this, in recent years, kesterite absorber materials, most prominently represented by Cu2ZnSn(SxSe1-x)4 (CZTS) [[2], [3], [4], [5], [6]], have been largely studied as a replacement thin film absorber for solar cell applications. The main advantage this type of materials provides is the non-toxic and earth abundant nature of the elements involved in their production. However, no matter how good a certain material might be, the efficiency of the subsequently produced solar cell is intrinsically limited by what is commonly known as the Shockley-Queiser limit, which predicts the theoretical maximal values of the various solar cell parameters [7]. One way to outperform this theoretical limit is to create cells using multiple p-n junctions, commonly known as tandem cells. In order for this type of architecture to work, it is necessary to have a top-cell with a higher band gap than the one of the regular bottom cell. In addition to the commonly used S/Se replacement [8, 9], elemental replacement of Sn by Ge in the CZTS structure has been investigated as a possible solution to increase the band gap of the absorber material [8]. In the case of the selenium phase of the tin kesterite, Cu2ZnSnSe4, the band gap has been shown to vary from 1.1 eV to 1.4 eV when varying the Ge/(Sn + Ge) ratio from 0 to 1 [8]. Subsequently, Cu2ZnGe(SxSe1-x)4 (CZGS(e) [10], with a band gap of 1.4 eV to 2.0 eV [8, 11] is a promising candidate to produce a high band gap kesterite material to be used in the top cell of a tandem thin film solar cell. The ideal band gap for the top cell is situated between 1.6 eV and 1.8 eV [12].

The most predominantly used methods to produce CZGS are wet processes (e.g. [13, 14]) or annealing of metallic precursors using elemental Se and S sources [[15], [16], [17], [18]]. These methods allow for a high throughput. In the present case, it was decided to produce samples through annealing of metallic precursors, but to perform the annealing in gaseous atmosphere (H2Se and H2S). This approach was chosen in the hope that it would allow for an optimal control of the concentration of elements in the annealing reactor.

Section snippets

Experimental methodology

The metallic precursors are deposited on a 5 cm × 5 cm Soda-Lime Glass substrates coated with a 150 nm Si(O,N) diffusion barrier and a 400 nm Mo back contact. The deposition is performed by e-beam evaporation in a Pfeiffer E-Gun PLS 580. A stack of a 200 nm layer of Ge, a 125 nm layer of Zn, a 170 nm thick layer of Cu is deposited by evaporation from solid, 99.999% pure, metal pellet crucibles in vacuum.

Before the samples are processed any further, they are split into two 2.5 cm × 5 cm samples.

Results and discussion

Before developing the recipe used to obtain the samples presented in this work and the results that are linked to them, an annealing recipe with one single filling of the reaction chamber with H2S gas was used. It appeared that reducing the sulfur inclusion after a 120 min annealing period was possible by reducing the amount of H2S pressure injected into the reactor prior to annealing, as shown in (Fig. 1a). In this figure, the sulfur inclusion, as determined from XRD measurements, is plotted

Conclusions

Using an annealing process in which evaporated metallic precursors are successively submitted to a short selenization step and a long sulfurization step, it was possible to create CZGSSe absorber layers with sulfur inclusion going from 30% to 100% that have respective band gap of 1.45 eV to 2.0 eV. The results of these annealing procedures were reproducible and brought forth an almost pure kesterite with a controllable band gap. In addition to the linear dependence between sulfurization time

Acknowledgements

This work was performed within the framework of the SWInG project funded by the European Union's Horizon 2020 research and innovation program, grant agreement No. 640868.

M. Neuwirth gratefully acknowledges financial support by the German Federal Ministry of Education and Research (BMBF, FKZ: 03SF0530B) and the Karlsruhe School of Optics and Photonics (KSOP).

References (23)

  • M.P. Suryawanshi et al.

    CZTS based thin film solar cells: a status review

    Mater. Technol.

    (2013)
  • Cited by (12)

    • Wide band gap Cu<inf>2</inf>ZnGe(S,Se)<inf>4</inf> thin films and solar cells: Influence of Na content and incorporation method

      2021, Solar Energy
      Citation Excerpt :

      In order to improve the device performance, it is necessary to optimize the sulphur content and its distribution through the absorber layer, both of which strongly affect the properties of the material. Kohl et al. (2018) observed an inhomogeneous distribution of the [S]/([S] + [Se]) atomic ratio through the CZGSSe kesterite layer when metallic precursors were annealed with H2Se and H2S, detecting a sulphur inclusion at the surface up to 20 % higher than in the bulk material. It has been observed in literature that sulfurization of the surface of the absorber leads to a higher Eg on the surface, increasing the VOC and device efficiency (Cai et al., 2017).

    • Correlation and optimization of the optoelectrical properties of DC magnetron-sputtered Cu<inf>2</inf>ZnSnS<inf>4</inf> absorber layer as a function of the material composition

      2021, Materials Science in Semiconductor Processing
      Citation Excerpt :

      This tendency is less pronounced for samples after RTP 4, where bandgap varies from 1.50 to 1.53eV from S to L. The S2 sample in the S group shows the highest bandgap for the CZTS compound and could be interesting to consider in the cases of ultrathin [8] or tandem structures [50]. The samples of M and L groups are suitable for single junction solar cell application due to their near optimal bandgap range.

    • Study of role of different defects on the performance of CZTSe solar cells using SCAPS

      2020, Optik
      Citation Excerpt :

      It exhibits characteristics for instance it is environmentally benign and its component elements are earth abundant, it has optimum direct bandgap and also it has high absorption coefficient. These properties are dependent upon choice of material and an alternate to the present solar photovoltaic thin film technologies [1,2]. It has been found to be a potential substitute to other contemporary thin film technology such as CIGS and CdTe technology, with an efficiency of 11.1% shown already [3].

    View all citing articles on Scopus
    View full text