Driven by sub-bandgap electric work and Peltier heat, thermoelectric-driven light-emitting diodes (TED-LEDs) not only offer much enhanced power-conversion-efficiency but also eliminate the waste heat generated during the operation of LEDs. However, cost-effective and high-efficiency TED-LEDs are not readily accessible for the epitaxially grown III-V LEDs due to the high chip cost and efficiency droop at low-medium brightness (current densities). Here we show that electroluminescence of colloidal quantum dots (QDs) LEDs (QLEDs) circumvents the deficiencies faced by conventional LEDs. The optimal red-emitting device fabricated by cost-effective solution processing technics exhibits external- and internal-power-conversion-efficiency of 21.5% and 93.5% at 100 cd/m², suited for high-efficiency solid-state lighting and high-resolution display. At this brightness, the electric driving voltage (V) of 1.89 V is lower than the photon voltage (V_p = hv/q = 1.96 V, q being the elemental charge). With typical V_p = 1.96 V, electroluminescence can be detected with the driving voltage as low as 1.0–1.2 V. Luminance of the thermoelectric-driven QLEDs (TED-QLEDs) remains ideally diffusion-dominated with the driving voltage lower than ~ 1.5 V, and further improvement on charge transport is expected to extend the linear ideality to all practical driving voltages.

A great variety of high-quality inorganic nanocrystals are synthesized solely in hydrocarbon solvents in both academic and industrial settings on a daily basis, which is largely complicated by lack of simple precursors containing inorganic element(s) yet soluble in the reaction solvents at ambient temperatures. Here, we introduce a new strategy for preparing the precursors, namely inorganic (or element-containing organic) molecules dispersed in hydrocarbon (Vaseline-octadecene) gel. This strategy not only greatly expands spectra of potential precursors and their concentration range, but also simplifies synthetic system, enables automated large-scale synthesis, and minimizes environmental concerns.

Alkanoate-coated CdSe/CdS core/shell quantum dots (QDs) with near-unity photoluminescence (PL) quantum yield and mono-exponential PL decay dynamics are applied for studying quasi-stationary charge transfer from photo-excited QDs to quinone derivatives physically-adsorbed within the ligand monolayer of a QD. Though PL quenching efficiency due to electron transfer can be up to > 80%, transient PL and transient absorption spectra reveal that the charge transfer rate ranges from single-digit nanoseconds to sub-nanoseconds, which is ~ 3 orders of magnitude slower than that of static charge transfer and ~ 2 orders of magnitude faster than that of collisional charge transfer. The physically-adsorbed acceptors can slowly (500–1, 000 min dependent on the size of the quinone derivatives) desorb from the ligand monolayer after removal of the free acceptors. Contrary to collisional charge transfer, the efficiency of quasi-stationary charge transfer increases as the ligand length increases by providing additional adsorption compartments in the elongated hydrocarbon chain region. Because ligand monolayer commonly exists for a typical colloidal nanocrystal, the quasi-stationary charge transfer uncovered here would likely play an important role when colloidal nanocrystals are involved in photocatalysis, photovoltaic devices, and other applications related to photo-excitation.

This work studies extinction properties of ZnSe quantum dots terminated with either Se-surface or Zn-surface (Se-ZnSe or Zn-ZnSe QDs). In addition to commonly observed photoluminescence quenching by anionic surface sites, Se-ZnSe QDs are found to show drastic signatures of Se-surface states in their UV-visible (Vis) absorption spectra. Similar to most QDs reported in literature, monodisperse Zn-ZnSe QDs show sharp absorption features and blue-shifted yet steep absorption edge respect to the bulk bandgap. However, for monodisperse Se-ZnSe QDs, all absorption features are smeared and a low-energy tail is identified to extend to an energy window below the bulk ZnSe bandgap. Along increasing their size, a cyclic growth of ZnSe QDs switches their surface from Zn-terminated to Se-terminated ones, which confirms that the specific absorption signatures are reproducibly repeated between those of two types of the QDs. Though the extinction coefficients per unit of Se-ZnSe QDs are always larger than those of Zn-ZnSe QDs with the same size, both of them approach the same bulk limit. In addition to contribution of the lattice, extinction coefficients per nanocrystal of Zn-ZnSe QDs show an exponential term against their sizes, which is expected for quantum-confinement enhancement of electron-hole wavefunction overlapping. For Se-ZnSe QDs, there is the third term identified for their extinction coefficients per nanocrystal, which is proportional to the square of size of the QDs and consistent with surface contribution.

The extinction coefficient of semiconductor nanocrystals is a key parameter for understanding both the quantum confinement and applications of the nanocrystals. The existing extinction coefficients of CdE (E = Se, S) nanocrystals were found to have an unacceptable deviation for the zinc-blende CdE quantum dots (QDs). The analysis reveals that, in addition to the interference of impurities, the commonly applied extinction coefficient per CdE nanocrystal is sensitive to the size, shape, and density of the surface ligands of nanocrystals. The extinction coefficient per CdE unit does not depend on accurate information of the size, shape, and number of surface ligands of the nanocrystals. A new three-step purification scheme was developed to investigate three classes of possible impurities for accurate determination of the extinction coefficient per CdE unit, including CdE clusters not considered previously. Given that the sole ligands of zinc-blende CdE nanocrystals are cadmium fatty acid salts (CdFa₂), a universal formula for the nanocrystals can be written as (CdE)_n(CdFa₂)_m. The n: m ratio was accurately determined for purified nanocrystals. The resulting extinction coefficients per unit for both CdSe and CdS QDs were found to decrease exponentially as the size of the QDs increases, with the corresponding bulk value as the large-size limit.

Against general wisdom in crystallization, the nucleation of InP and Ⅲ-Ⅴ quantum dots (QDs) often dominates their growth. Systematic studies on InP QDs identified the key reason for this: the dense and tight alkanoate-ligand shell around each nanocrystal. Different strategies were explored to enable necessary ligand dynamics—i.e., ligands rapidly switching between being bonded to and detached from a nanocrystal upon thermal agitation—on nanocrystals to simultaneously retain colloidal stability and allow appreciable growth. Among all the surface-activation reagents tested, 2, 4-diketones (such as acetylacetone) allowed the full growth of InP QDs with indium alkanoates and trimethylsilylphosphine as precursors. While small fatty acids (such as acetic acid) were partially active, common neutral ligands (such as fatty amines, organophosphines, and phosphine oxides) showed limited activation effects. The existing amine-based synthesis of InP QDs was activated by acetic acid formed in situ. Surface activation with common precursors enabled the growth of InP QDs with a distinguishable absorption peak between ~450 and 650 nm at mild temperatures (140–180 ℃). Furthermore, surface activation was generally applicable for InAs and Ⅲ-Ⅴ based core/shell QDs.

A one-pot/three-step synthetic scheme was developed for phase-pure epitaxy of CdS shells on zinc-blende CdSe nanocrystals to yield shells with up to sixteen monolayers. The key parameters for the epitaxy were identified, including the core nanocrystal concentration, solvent type/composition, quality of the core nanocrystals, epitaxial growth temperature, type/concentration of ligands, and composition of the precursors. Most of these key parameters were not influential when the synthetic goal was thin-shell CdSe/CdS core/shell nanocrystals. The finalized synthetic scheme was reproducible at an almost quantitative level in terms of the crystal structure, shell thickness, and optical properties.

Using CdSe/CdS core/shell nanocrystals with 1–10 monolayers of CdS shell as the model system, we studied effects of thiol ligands on optical properties of the nanocrystals. The core/shell nanocrystals with original ligands possessed near unity photoluminescence (PL) quantum yield and single-exponential PL decay dynamics. The effects of thiol ligands on optical properties were found to depend on the shell thickness, environment (with/without oxygen), and excitation power (single- or multi-exciton). Systematic and quantitative results reported in this work should provide necessary information for fundamental understanding and technical applications of quantum dots (QDs) coated with thiol ligands.

General application of "greener methods" to the synthesis of monodisperse colloidal nanocrystals introduces impurities, including metal carboxylate precursors, non-volatile solvents, free ligands, and non-nanocrystalline side products. These impurities seriously diminish the solution processability and potential applications of colloidal nanocrystals. A protocol was established for evaluating purification schemes. The results revealed that commonly applied purification schemes and their variants do not exhibit a high level of performance and may degrade the ligand surface coverage. A new scheme involving chloroform–acetonitrile precipitation quantitatively removed all impurities from colloidal solutions of CdSe and CdS nanocrystals coated with a variety of carboxylate ligands. The new scheme was benign to the surface structure of nanocrystal-ligands complexes and resulted in each nanocrystal bearing a close-packed monolayer of carboxylate ligands.

A suspension of fine selenium (Se) powder (100 or 200 mesh) in octadecene (Se-SUS) has proven to be a high-performance, versatile, convenient, reproducible, yet green Se precursor. The advantages of Se-SUS arise from its highly reactive chemical nature and flexibility. These two features made it possible to carry out the synthesis of high quality metal selenide nanocrystals with diverse compositions and structures, including binary, core/shell, transition metal doped, and complex composition nanocrystals. These successes further demonstrated that Se-SUS is a powerful Se precursor for solving a few long-standing challenges in the synthesis of high quality selenide nanocrystals. For instance, Se-SUS was successfully employed as a Se precursor for shell growth in high quality core/shell nanocrystals to replace expensive and highly toxic precursors, such as Se-phosphine and bis-trimethylsilyl selenide, with greatly lowered epitaxial temperatures (as low as 150 ℃) to avoid alloying. As another example, Se-SUS enabled "co-nucleation doping" as a means of preparing high quality Mn doped ZnSe nanocrystals with pure, stable, and highly efficient dopant fluorescence.