Oxychalcogenides for transparent p-type conductors

Abstract

The quinary layered oxychalcogenide, Sr₃Sc₂O₅Cu₂S₂, was used as a prototype to investigate the effect of composition and structure, on transparency and conductivity of analogous materials. A selection of analogues was computationally screened for predicted stability and optical transparency. The compounds Ba₃Sc₂O₅Cu₂S₂ and Ca₃Al₂O₅Cu₂S₂ were both expected to show band gap energies higher than 3.1 eV and conductivities higher than that of the prototype. The compound Ba₃Sc₂O₅Cu₂S₂ was successfully synthesised here by solid-state reaction at 800 ^oC for 12 h, apparently for the first time. Its measured band-gap energy was 3.24 eV. Synthesis of Ca₃Al₂O₅Cu₂S₂ was not achieved. It was concluded that a high-pressure synthetic route would be required to realise this compound.

The compounds intermediate between the prototype and new member BaxSr₃-xSc₂O₅Cu₂S₂ where x = 1 and 2, were also successfully synthesised. Refinement of their crystal structures showed a preference of barium for the ‘intra-layer’ site in the perovskite layer, as rationalised by the Goldschmidt tolerance factor. The perovskite A site mixing yielded extreme bond distances and angles in the conductive copper sulfide layer of x = 2, beyond the range spanned by the end-members. Williamson-Hall analysis revealed increased strain and reduced crystallite size for the intermediates and general anisotropic growth of these compounds favouring the direction of conduction, as confirmed by scanning-electron microscopy.

The successful syntheses of the analogous compounds Ba₃Sc₂O₅Cu₂S₂, Ba₃Sc₂O₅Cu₂Se₂, Ba₃Y₂O₅Cu₂Se₂, Ba₃In₂O₅Cu₂Se₂, Ba₃Sc₂O₅Ag₂Se₂ and Ba₃In₂O₅Ag₂Se₂ were also reported here, for the first time to the authors knowledge. It was found that the increased radius of the perovskite M cation, increased the basal lattice parameter and linked coinage metal-metal distance, expected to benefit transparency. However, observed reductions in band-gap energy for both the indium and yttrium analogues, relative to those of the scandium compounds, were rationalised by electronegativity and (n-1)d-orbital occupancy, respectively. The coinage metal-chalcogenide distance increased, and angle deviated from the ideal tetrahedral, with increasing M cation radius, in all cases. Barium and scandium proved to be the optimal candidate perovskite layer cations. Substitution of selenium and silver into the conductive layer was expected to increase conductivity at the expense of an observed decrease in band gap for the scandium analogues.

The compounds Ba₃Sc₂O₅Cu₂S₂ and Ba3Sc2O5Cu2Se2 showed promise as improved transparent p-type conductors. Their compositions and structures suggested improved conductivity compared to that of Sr₃Sc₂O₅Cu₂S. The 3.05 eV band gap of Ba3Sc2O5Cu2Se2 may be increased above the visible light threshold by p-type doping. A preliminary investigation saw the prototype successfully doped by 5 at. % sodium with an accompanied Moss-Burstein shift in band gap energy of + 0.5 eV to 3.2 eV.

More information

Identifiers

ORCID for Geoffrey Hyett: orcid.org/0000-0001-9302-9723

Catalogue record

Export record