Extrinsic p-Type Doping of Sb₂Se₃ Single Crystals and Solar Cells

Antimony selenide (Sb₂Se₃) has garnered significant attention as a photovoltaic absorber material due to its promising characteristics, such as a suitable band gap (~1.2 eV), high absorption coefficient, and abundance of constituent elements. One notable feature of Sb₂Se₃ is its low intrinsic carrier concentration at room temperature, with experimental values as low as ~10¹⁰ cm^-3 [1]. Inspection of native defects using density functional theory [2] reveals that the Fermi level remains close to the middle of the band gap irrespective of stoichiometry, and single crystals grown from high purity material is found experimentally to be highly insulating regardless of whether they are grown selenium-rich or selenium-poor [3]. This suggests that intrinsic doping is insufficient for achieving the conductivity required for efficient device performance.<br/>Therefore, the necessary conductivity for functional Sb₂Se₃ devices requires extrinsic dopants, most often in the form of unintentional impurities. These impurities can significantly impact device performance, yet they are seldom identified, never mind studied in detail. In standard device architectures, a p-type absorber layer is typically preferred. However, the reliance on unknown, unintentional impurities frequently results in n-type conductivity. This n-type conductivity, although not always explicitly reported, can be inferred from photoemission measurements and often leads to charge separation via an n-n+ heterojunction [3], which is not optimal for device efficiency.<br/>Efforts to achieve p-type doping in Sb₂Se₃ have been explored in the literature, primarily focusing on group 4A elements such as tin (Sn) and lead (Pb). Tin in particular has received considerable attention due to its position in the periodic table, with one less electron than antimony. However, tin tends is often found in the 4+ oxidation state, whereas the 2+ state is required to increase hole density when incorporated substitutionally onto the Sb site (which is in the 3+ state). At present, there has been little consideration of group 2B transition metal elements, which are more likely to be found in the 2+ state, as potential p-type dopants.<br/>In this work, we investigate the potential of cadmium (Cd) and zinc (Zn) to induce p-type doping in Sb₂Se₃. Our approach involves fabricating sample sets of both highly pure single crystals and full solar cell devices in parallel. This dual-faceted strategy allows us to correlate the electrical properties of the doped material in an idealized material system with the performance in a functioning device stack. After investigating a wide range of dopant incorporation levels, we find that both Cd and Zn can contribute effective p-type doping. Cd appears to be a more promising choice of dopant, producing crystals with low resistivity and apparent doping densities in excess of 10¹⁶cm^-3, and can therefore comfortably reach the carrier concentrations required in an absorber layer. Initial device tests have shown a significant improvement in the V_oc of thin film solar cells compared to baseline devices prepared using commercially available Sb₂Se₃ which includes Cl as an unintentional impurity, as well as intentionally doped Sn doped Sb₂Se₃ source material. Through this study, we aim to gain a deeper understanding of the impact of Cd and Zn doping on the electrical properties and overall performance of Sb₂Se₃ solar cells, providing insights that could lead to more efficient photovoltaic devices.<br/><br/>[1] N. Cifuentes et al, “Electronic Conduction Mechanisms and Defects in Polycrystalline Antimony Selenide”<i>J. Phys. Chem. C</i> 2020, 124, 14, 7677–7682<br/>[2] C.N Savory & D. O. Scanlon, “The complex defect chemistry of antimony selenide” J. Mater. Chem. A, 2019, 7, 10739-10744<br/>[3] T D. C. Hobson et al, “Isotype Heterojunction Solar Cells Using n-Type Sb2Se3 Thin Films”, Chemistry of Materials 2020, 32 (6), 2621-2630