ABSTRACT
The design of new host materials for phosphorescent organic light emitting diodes (OLEDs) is challenging because several physical property requirements must be met simultaneously. A triplet energy ( ET) higher than that of the chosen emitting dopant, appropriate highest occupied molecular orbital/lowest unoccupied molecular orbital energy levels, good charge carrier transport, and high stability are all required. Here, computational methods were used to screen structures to find the most promising candidates for OLED hosts. The screening was carried out in three Tiers. The Tier 1 selection, based on density functional theory calculations, identified a set of eight molecular structures with ET > 2.9 eV, suitable for hosting blue phosphorescent dopants such as iridium(III)bis((4,6-di-fluorophenyl)-pyridinato-N,C2')picolinate. Phenanthro[9,10- d]imidazole was chosen as the starting point for the Tier 2 selection. Thirty-seven unique molecular structures were enumerated by isoelectronic nitrogen transmutation of up to two CH fragments of the phenanthrene. Three molecules, that is, imidazo[4,5- f]-phenanthrolines with nitrogens at the 1,10-, 3,8-, and 4,7-positions, were selected for Tier 3, which involved the use of molecular dynamics simulations and electron coupling calculations to predict differences in charge transport between the three materials. The three were explored experimentally through synthesis and device fabrication. The singlet, triplet, and frontier orbital energies computed using single-molecule density functional theory calculations ( Tiers 1 and 2) were consistent with the experimental values in a fluid solution, and the multiscale modeling scheme ( Tier 3) correctly predicted the poor device performance of one material. We conclude that screening host materials using only single-molecule quantum mechanical data was not sufficient to predict whether a given material would make a good OLED host with certainty; however, they can be used to screen out materials that are destined to fail due to low singlet/triplet energies or a poor match of the frontier orbital energies to the dopant or transport materials.
ABSTRACT
Extracellular matrices (ECM) derived from pluripotent stem cells (PSCs) provide a unique tissue microenvironment that can direct cellular differentiation and tissue regeneration, and rejuvenate aged progenitor cells. The unlimited growth capacity of PSCs allows for the scalable generation of PSC-secreted ECMs. Therefore, the derivation and characterization of PSC-derived ECMs is of critical importance in drug screening, disease modeling and tissue regeneration. In this study, 3-D ECMs were generated from decellularized undifferentiated embryonic stem cell (ESC) aggregates (AGG), spontaneously differentiated embryoid bodies (EB), and ESC-derived neural progenitor cell (NPC) aggregates. The capacities of different ECMs to direct proliferation and neural differentiation of the reseeded mouse ESCs and human induced pluripotent stem cells (iPSCs) were characterized. Proteomic analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS) revealed protein expression profiles that reflected distinct niche properties for each tested ECM group. The reseeded mouse ESCs and human iPSCs responded to different types of ECMs with different cellular phenotypes. Cells grown on the AGG-ECM displayed high levels of pluripotent markers Oct-4 and Nanog, while the cells grown on the NPC-ECM showed increased expression of neural marker ß-tubulin III. The expression levels of ß-catenin were high for cells grown on the AGG-ECM and the EB-ECM, but reduced in cells grown on the NPC-ECM, indicating a possible role of Wnt/ß-catenin signaling in the cell-matrix interactions. This study demonstrates that PSC-derived ECMs can influence stem cell fate decisions by providing a spectrum of stem cell niche microenvironments during tissue development.
