Spectroscopic and Magneto-Optical Signatures of Cu1+ and Cu2+ Defects in Copper Indium Sulfide Quantum Dots

Colloidal quantum dots (QDs) of I-III-VI ternary compounds such as copper indium sulfide (CIS) and copper indium selenide (CISe) have been under intense investigation due to both their unusual photophysical properties and considerable technological utility. These materials feature a toxic-element-free composition, a tunable bandgap that covers near-infrared and visible spectral energies, and a highly efficient photoluminescence (PL) whose spectrum is located in the reabsorption-free intragap region. These properties make them attractive for light-emission and light-harvesting applications including photovoltaics and luminescent solar concentrators. Despite a large body of literature on device-related studies of CISe(S) QDs, the understanding of their fundamental photophysical properties is surprisingly poor. Two particular subjects that are still heavily debated in the literature include the mechanism(s) for strong intragap emission and the reason(s) for a poorly defined (featureless) absorption edge, which often "tails" below the nominal bandgap. Here, we address these questions by conducting comprehensive spectroscopic studies of CIS QD samples with varied Cu-to-In ratios using resonant PL and PL excitation, femtosecond transient absorption, and magnetic circular dichroism measurements. These studies reveal a strong effect of stoichiometry on the concentration of Cu(1+)vs Cu2+ defects (occurring as Cu-In '' and CuCu center dot species, respectively), and their effects on QD optical properties. In particular, we demonstrate that the increase in the relative amount of Cu(2+)vs Cu1+ centers suppresses intragap absorption associated with Cu1+ states and sharpens band-edge absorption. In addition, we show that both Cu1+ and Cu2+ centers are emissive but are characterized by distinct activation mechanisms and slightly different emission energies due to different crystal lattice environments. An important overall conclusion of this study is that the relative importance of the Cu(2+)vs Cu1+ emission/absorption channels can be controlled by tuning the Cu-to-In ratio, suggesting that the control of sample stoichiometry represents a powerful tool for achieving functionalities (e.g., strong intragap emission) that are not accessible with ideal, defect-free materials.