Influence of the Processing Environment on the Surface Composition and Electronic Structure of Size-Quantized CdSe Quantum Dots

Abstract

We show here how differences in surface composition for monolayer (ML)-tethered pyridine (Py)-treated CdSe quantum dots (QDs) affect band edge energetics, as a function of QD diameter. We compare UV−vis spectroscopy of QD solutions and X-ray (XPS) and ultraviolet (UPS) photoelectron spectroscopy of QDs tethered to 1,6hexanedithiol-modified Au surfaces. Differences in QD surface composition are brought about by differences in solvent composition during displacement of the native, X-type octadecylphosphonate (ODP − ) ligands with Py and by differences in exposure to trace H 2 O and O 2 during QD processing. Prior to surface tethering, exchange of the ODP − ligands with Py ligands under ambient (A) conditions leads to some surface passivation of the QD and no appreciable reduction in the spectroscopically estimated QD diameter. QDs processed in completely inert (I) environments (O 2 and H 2 O below detection limits in the solvents, processing environments, and during analysis) undergo a small decrease in diameter during Py exchange and an increase in size dispersity. These simple differences in processing conditions lead to important differences in QD band edge energetics which will impact their use as photocatalysts and photovoltaic or light-emitting diode active layers. XPS characterization of Py-capped QD-tethered MLs, compared with freshly etched CdSe(0001) single-crystal surfaces, indicates that I QDs show a higher Se surface coverage relative to A QDs. For A QDs, we propose that trace H 2 O present during processing provides a proton source that facilitates ODP − desorption and replacement with charge-compensating HO − ligands, which inhibits subsequent changes in QD surface composition. UPS-derived size-quantized ionization potentials (IPs) and electron affinities (EAs) for I QDs, corrected for shifts in the local vacuum level, closely track the energetic shifts predicted by the effective mass approximation (EMA). The vacuum level-corrected IP/EA values for A QDs show sizeable deviations from the EMA. We also describe an approach for characterization of UPS data for these QD MLs, which greatly enhances the sensitivity to mid-gap states above the VBM (seen for I QDs which are slightly enriched in surface Se) and offers a general approach to all semiconductor materials with low density of states in the valence band region. This study confirms the influence of ligand modification and processing environment on QD surface composition, which in turn impacts the CdSe QD energy levels that are important to applications in photocatalysis and optoelectronic device platforms.

Near-Surface Composition, Structure, and Energetics of TiO2 Thin Films: Characterization of Stress-Induced Defect States in Oxides Prepared via Chemical Vapor Deposition versus Solution Deposition from Sol–Gel Precursors

“…41 This approach has been shown to be effective in determining low concentrations of midgap states in metal oxides, perovskites, and semiconductor nanocrystalline thin films. 17,18,69 The measurement science approaches we show here can be extrapolated to a wide range of both n-type and p-type oxide films proposed as interlayers in thin film solar cells and other optoelectronic devices. 17,18,35 We posit that an increased understanding of the defects in these oxide films as they pertain to their use as electrical contacts in printed solar cell technologies will also provide insights into the use of these oxides as catalytic surfaces in both photoelectrochemical water and air-purification technologies where the photoactivated oxide layers need to be optimized to enhance both adsorption of key contaminants and biomolecules and their degradation into environmentally benign products.…”

“…Relative atomic concentrations are quantified from background-corrected and deconvoluted CL spectra, which are corrected for the orbital cross sections and KE-dependent analyzer transfer function. 69 UPS VB spectra and the low kinetic energy secondary electron cutoff (SECO) are measured with a He discharge lamp operating at 20 mA. The UHP He is passed through a liquid N 2 trap filled with molecular sieves.…”

“…As with many of our prior studies using this approach, we ensured first of all that the data acquisition times needed to provide this level of S/N did not introduce detectable damage or chemical changes to surface composition (as verified by XPS) that would impact on the conclusions shown here. 17,18,69 Using this approach, we observe photoemission from the He Iα excitation line at 21.22 eV and satellite emission lines (He Iβ = 23.09 eV and He Iγ = 23.74 eV) with relative intensities of ca. 1.4−1.8% and 0.5− 0.7%, respectively, which must be removed to characterize lowintensity photoemission peaks from electron-rich states in the oxide surface above VBM.…”

“…This is an approach that we have found produces physically realistic E VBM values for several different semiconductor materials, even for systems with low DOS near or above VBM, and in some instances spectral features seen above VBM with our conventional UPS spectrometer and approach to data treatment have been confirmed exactly with UPS sources that are monochromatized, i.e., producing valence band spectra with no need for further corrections. 17,18,69 ■ RESULTS AND DISCUSSION Structure and Morphology. GI-XRD and AFM demonstrate a clear difference in structural transformations for CVD versus SG TiO 2 thin films, on both ITO and FTO substrates, that were annealed after deposition by using protocols commonly employed for optoelectronic devices using these films as electron-selective contacts.…”

Predicting Indium Phosphide Quantum Dot Properties from Synthetic Procedures Using Machine Learning

“…As shown in Figure S1 and discussed in Section 2.1, many publications did not report details in the synthesis, which significantly decreases the size of the dataset despite the large number of published reports of InP QD synthesis in the past decades. Other synthetic parameters that also affect the synthetic outcomes but are often not mentioned include injection rate 55 , solvents and precursors purity 56,57 , and QD purification status and methods [58][59][60] . Moreover, heterogeneity of experimental condition in different labs also impacts the synthetic outcomes.…”