Elsevier

Thin Solid Films

Volumes 361–362, 21 February 2000, Pages 383-387
Thin Solid Films

Chemical stability of Sb2Te3 back contacts to CdS/CdTe solar cells

https://doi.org/10.1016/S0040-6090(99)00807-XGet rights and content

Abstract

The chemical stability of back contact materials and metallisations like Sb2Te3 and Ni:V alloys to CdS/CdTe solar cells was investigated. Ternary and quaternary (Cd–Sb–Te–O, Sb–Ni–Te–O) phase diagrams were calculated using thermodynamic data. By applying these diagrams to CdTe/Sb2Te3 or Sb2Te3/Ni–V interfaces used as back contacts in CdS/CdTe solar cells, predictions could be made about chemical stability and possible interface reactions. The phase diagrams and predicted chemical behaviour were proven by some test reactions. Products and reactions were characterised by XRD and DSC. It was found that the CdTe/Sb2Te3 interface is in thermodynamic equilibrium and no reaction occurs. Sb2Te3/Ni and Sb2Te3/V interfaces are not in thermodynamic equilibrium and reaction products like NiTe, NiTe2 or V2Te3 are found in test reactions at low temperatures (200°C) after very short times (1 h), and in the case of V even at room temperature. It is therefore expected that chemical reactions at these interfaces will lead to both degradation of the Sb2Te3 and the Ni–V layer and may influence the efficiency of CdS/CdTe solar cells.

Introduction

Research in recent years has shown that CdTe/CdS thin layer polycrystalline solar cells offer a good possibility to produce cheap photovoltaic modules [1]. Maximum efficiencies reported for laboratory solar cells are about 16% [2], [3]. In theory it should be possible to construct CdTe/CdS solar cells with much higher efficiencies up to 30% [4]. In practice electrical barriers within the multi-layer device may lead to efficiency losses. To date it is still not well understood what materials, especially at the back of the cell (CdTe side), lead to good and stable contacts and therefore efficient and reliable solar cells. Electrical contacts to CdTe are recently reviewed elsewhere [5]. Long-term stability of the back contact may be achieved by using materials which are in thermodynamic equilibrium to CdTe and to their metallisation. No chemical reaction at the interfaces would be expected for these kinds of materials. To determine the relative stabilities of the applied semiconductors and metals to each other and to their native oxides, the construction of phase equilibrium diagrams is necessary. With these phase diagrams predictions of chemical stability and chemical reactions could be made, and possibly degradation of the thin layers and electronic barrier losses could be understood. In this work the stability of CdTe/Sb2Te3 contacts and Sb2Te3JNi:V metallisations are investigated (V is included in Ni targets for magnetron sputtering).

Section snippets

Experimental details

Experimental verification of the calculated phase diagrams (see Section 3) was carried out as follows.

CdTe, Sb2Te3, 5b203, Ni, NiO, and V were used as purchased from Aldrich in a minimum of 99.5+% grade. Reaction mixtures were prepared by thoroughly mixing the powder compounds in an agate mortar in an appropriate ratio, e.g. 0.12 g (0.5 mmol) CdTe and 0.31 g (0.5 mmol) Sb2Te3, 0.12 g (0.5 mmol) CdTe and 0.15 g (0.5 mmol) 5b2O3, 0.235 g (4 mmol) Ni and 0.626 g (1 mmol) Sb2Te3, 0.448 g (6 mmol)

Calculated phase diagrams

A typical composition of a CdS/CdTe solar cell structure investigated in this work is shown in Fig. 1. The back contact consists of two thin layers. The narrow band gap semiconductor Sb2Te3 contacts the p-CdTe layer and is metallised with a Ni/V alloy (7% V). Although it was shown that these materials form a good electrical contact with low barrier losses [6], the question of long time stability of theses contacts still arises.

To investigate the chemical stability of the back contact ternary

Conclusion

The object of this work was to investigate the fundamental chemical stability of the Sb2Te3/Ni back contact to CdS/CdTe solar cells. Theoretical calculated phase diagrams allowed us to predict chemical reactions at the interfaces of the back contact. The predicted reactions were proven by test reactions performed with mixtures of powder substances.

In general, the reactions predicted from room temperature thermodynamic data were in good agreement with the experimentally observed products at

Acknowledgements

The authors would like to thank the EU for financial support (Joule-Thermie Cadback, Project J0R39802 18) and Dr. M.G. Davidson from the Chemistry Department, University of Durham.

References (13)

  • K Durose et al.

    J. Cryst. Growth

    (1999)
  • A.M Azad et al.

    J. Chem. Thermodyn.

    (1986)
  • K Ishida et al.

    J. Less-Common Metals

    (1985)
  • T Aramoto et al.

    Jpn. J. Appl. Phys.

    (1997)
  • J Britt et al.

    Appl. Phys. Lett.

    (1993)
  • A De Vos et al.
There are more references available in the full text version of this article.

Cited by (0)

View full text