Abstract

We have systematically investigated native point defect chemistry of Cu <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ZnSn(S,Se) <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> and the effects of selected extrinsic dopants by spray coating films from DMSO-thiourea inks. Over 6000 unique compositions of Cu <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ZnSn(S,Se) <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">4</sub> have been analyzed, along with the effects of Cd, Fe, and Na doping. Spectrally resolved absolute intensity photoluminescence is then used to map the optoelectronic properties. The data are analyzed using a full-spectrum fitting technique that allows us to extract the optical bandgap, quasi-Fermi level splitting (QFLS), as well as characteristics of the subbandgap absorption and emission. We find that changes in the optoelectronic properties primarily result from changes in the Cu-content. From stoichiometric, as the Cu-content is decreased, the PL peak shifts to higher energy, the bandgap increases, and the optolectronic quality (QFLS/QFLS <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">max</sub> ) increases. However, the full-width at half-maximum (FWHM) and the average subbandgap absorptivity remain nearly constant, indicating that the magnitude of electrostatic potential fluctuations does not change significantly. The most likely Shockley-Read-Hall (SRH)-active defects appear to be defects and defect clusters involving SnZn, CuSn, and SnCu antisite defects, particularly the cluster [2Cu <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Zn</sub> + Sn <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Zn</sub> ]. We further show the benign effect of Cd doping, the detrimental effect of Fe doping, and the ability to make 9.5% efficient devices with ink-based NaCl doping.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call