The research on metal halide perovskite light‐emitting diodes (PeLEDs) with green and infrared emission has demonstrated significant progress in achieving higher functional performance. However, the realization of stable pure‐blue (≈470 nm wavelength) PeLEDs with increased brightness and efficiency still constitutes a considerable challenge. Here, a novel acid etching‐driven ligand exchange strategy is devised for achieving pure‐blue emitting small‐sized (≈4 nm) CsPbBr3 perovskite quantum dots (QDs) with ultralow trap density and excellent stability. The acid, hydrogen bromide (HBr), is employed to etch imperfect [PbBr6]4− octahedrons, thereby removing surface defects and excessive carboxylate ligands. Subsequently, didodecylamine and phenethylamine are successively introduced to bond the residual uncoordinated sites of the QDs and attain in situ exchange with the original long‐chain organic ligands, resulting in near‐unity quantum yield (97%) and remarkable stability. The QD‐based PeLEDs exhibit pure‐blue electroluminescence at 470 nm (corresponding to the Commission Internationale del'Eclairage (CIE) (0.13, 0.11) coordinates), an external quantum efficiency of 4.7%, and a remarkable luminance of 3850 cd m−2, which is the highest brightness reported so far for pure‐blue PeLEDs. Furthermore, the PeLEDs exhibit robust durability, with a half‐lifetime exceeding 12 h under continuous operation, representing a record performance value for blue PeLEDs.

Silicon photonics has offered a versatile platform for the recent development of integrated optomechanical circuits. However, silicon is limited to wavelengths above 1.1 µm and does not allow device operation in the visible spectrum range where low-noise lasers are conveniently available. The narrow bandgap of silicon also makes silicon optomechanical devices susceptible to strong two-photon absorption and free carrier absorption, which often introduce strong thermal effects that limit the devices' stability and cooling performance. Further, silicon also does not provide the desired lowest order optical nonlinearity for interfacing with other active electrical components on a chip. On the other hand, aluminum nitride (AlN) is a wide-band semiconductor widely used in micromechanical resonators due to its low mechanical loss and high electromechanical coupling strength. In this paper, we report the development of AlN-on-silicon platform for low loss, wide-band optical guiding, as well as its use for achieving simultaneously high-optical-qualityfactor and high-mechanical-quality-factor optomechanical devices. Exploiting AlN's inherent second-order nonlinearity we further demonstrate electro-optic modulation and efficient second harmonic generation in AlN photonic circuits. 2Our results suggest that low-cost AlN-on-silicon photonic circuits are excellent substitutes for complementary metal-oxide-semiconductor-compatible photonic circuits for building new functional optomechanical devices that are free from carrier effects. Contents

All inorganic halide perovskites in the form of colloidal quantum dots (QDs) have come into people’s view as one of the potential materials for the high-efficiency solar cells; nevertheless, the high surface trap density and poor stability of QDs restrict the performance improvement and application. Here, we obtain colloidal inorganic perovskite CsPb1–x Zn x I3 QDs by the hot-injection synthesis process with the addition of ZnCl2. Synchrotron-based X-ray absorption fine structures demonstrate that the guest Zn2+ ions are doped into the CsPbI3 structure to improve the local ordering of the lattice of the perovskite, reducing the octahedral distortions. The increase of the Goldschmidt tolerance factor and the Pb–I bond energy also enhance the stability of the perovskite structure. Furthermore, the Cl– ions from ZnCl2 occupy the iodide vacancies of the perovskite to decrease the nonradiative recombination. The synergistic effect of doping and defect passivation makes for stable colloidal CsPb0.97Zn0.03I3 QDs with ultralow density of trap states. The champion solar cell based on the QDs shows a power conversion efficiency of 14.8% and a largely improved stability under ambient conditions.

By analyzing the propagating behavior of the supermodes in a coupled-waveguide system, we have derived a universal criterion for designing adiabatic mode transformers. The criterion relates , the fraction of power scattered into the unwanted mode, to waveguide design parameters and gives the shortest possible length of an adiabatic mode transformer, which is approximately 2/ 1/2 times the distance of maximal power transfer between the waveguides. The results from numerical calculations based on a transfer-matrix formalism support this theory very well. [7,8]. Two schemes have been implemented to realize the mode transformationresonant coupling and adiabatic coupling. In a resonant coupler, by designing the coupling region to be of a half beat length, the light transfers from one waveguide to the other [8]. The coupler length can be made very short in this manner; however, it is practically difficult to determine the exact beat length, rendering the efficiency of power transfer uncertain and the resulting devices of dubious value. An adiabatic coupler, on the other hand, does not require a precise definition of power-transfer length [1,9,10], but it has to be sufficiently long to satisfy the adiabatic condition to reduce the coupling of power into other unwanted modes. Clearly a longer coupler not only reduces the component density but also suffers from higher transmission losses and higher probability of material defects and fabrication imperfections. The optimal design procedure of adiabatic mode transformers has been proposed in different ways. Love et al. first studied the fiber tapers, suggesting that for a given taper length the optimal delineating curve should have the local taper angle inversely proportional to the local beat length [1,11]. This design principle was also employed in experiments [10,12]. Another design concept is based on equalization of the "single-step loss" (defined as the overlap integral of the modes in two adjacent segments) along the taper [13,14]. Based on the stationary field distributions rather than the wave propagation behavior, those analyses did not point out the shortest taper length with which a certain coupling efficiency could be achieved. In this Letter we derive a universal criterion for the adiabatic mode transition in a coupledwaveguide system and suggest the shortest length of an adiabatic mode transformer for a given maximal tolerated scattering from the wanted mode into other modes during power transfer.The mode transformer to be analyzed here is based on a coupled-waveguide system shown in Fig. 1. It consists of two waveguides, waveguide 1 and waveguide 2, placed in close proximity to each other. The refractive index or geometry of at least one waveguide is gradually varied along the propagation direction z. Light is coupled into this transformer at input plane z = z i ͑=0͒ and out at output plane z = z f . The normalized local modes (or "supermodes"), denoted as e e for the even mode and e o for the odd mode, of this coupled-waveguide system are expressed as col...