No abstract

No abstract

Low‐cost solution‐processed CdTe nanocrystal (NC) solar cells always suffer from a high interface energy barrier and unbalanced hole/electron transport as well as anisotropic atom diffusion on the CdTe surface due to the limited amount of hole/electron interface materials or the difficulty in interface processing. In this work, a novel strategy is first adopted with gradient electron transport layer (CdS/CdSe) modification in the cathode and a new crosslinkable hole transport polymer (P‐TPA) implantation in the anode. The carrier recombination at interfaces is greatly decreased and thus the carrier collection is increased. Moreover, the light harvesting is improved both in short and long wavelength regions, making Jsc and Voc increase simultaneously. A champion solar cell shows a very high power conversion efficiency of 9.2% and an outstanding Jsc of 25.31 mA cm−2, which are among the highest values for all solution‐processed CdTe NC solar cells with a superstrate structure, and the latter value is even higher than that of traditional thick CdTe thin‐film solar cells (2 µm) via the high temperature close space sublimation method. This work demonstrates that facile surface modifications in both the cathode and anode with stepped extraction and organic–inorganic hybridization are very promising in constructing next‐generation highly efficient NC photovoltaic devices.

Solution-processed CdTe nanocrystals solar cells have attracted much attention due to their low cost, low material consumption, and potential for roll-to-roll production. Among all kinds of semiconductor materials, CdS exhibits the lowest lattice mismatch with CdTe, which permits high junction quality and high device performance. In this study, high quality CdS nanocrystals were prepared by a non-injection technique with tetraethylthiuram disufide and 2,2 -dithiobisbenzothiazole as the stabilizers. Based on the CdTe and CdS nanocrystals, devices with the architecture of ITO/ZnO/CdS/CdTe/MoO x /Au were fabricated successfully by a solution process under ambient condition. The effects of annealing conditions, film thickness, and detailed device structure on the CdTe/CdS nanocrystal solar cells were investigated and discussed in detail. We demonstrate that high junction quality can be obtained by using CdS nanocrystal thin film compared to traditional CdS film via chemical bath deposition (CBD). The best device had short circuit current density (J sc ), open circuit voltage (V oc ) and fill factor (FF) of 17.26 mA/cm 2 , 0.56 V, and 52.84%, respectively, resulting in a power conversion efficiency (PCE) of 5.14%, which is significantly higher than that reported using CBD CdS as the window layer. This work provides important suggestions for the further improvement of efficiency in CdTe nanocrystal solar cells.

We propose Sb-doped TiO2 as electron acceptor material for depleted CdTe nanocrystal (NC) hetero-junction solar cells. Novel devices with the architecture of FTO/ZnO/Sb:TiO2/CdTe/Au based on CdTe NC and TiO2 precursor are fabricated by rational ambient solution process. By introducing TiO2 with dopant concentration, we are able to tailor the optoelectronic properties of NC solar cells. Our novel devices demonstrate a very high open circuit voltage of 0.74 V, which is the highest Voc reported for any CdTe NC based solar cells. The power conversion efficiency (PCE) of solar cells increases with the increase of Sb-doped content from 1% to 3%, then decreases almost linearly with further increase of Sb content due to the recombination effect. The champion device shows Jsc, Voc, FF, and PCE of 14.65 mA/cm2, 0.70 V, 34.44, and 3.53% respectively, which is prospective for solution processed NC solar cells with high Voc.

Considering the scarcity of medical data, most datasets in medical image analysis are an order of magnitude smaller than those of natural images. However, most Network Architecture Search (NAS) approaches in medical images focused on specific datasets and did not take into account the generalization ability of the learned architectures on unseen datasets as well as different domains. In this paper, we address this point by proposing to search for generalizable U-shape architectures on a composited dataset that mixes medical images from multiple segmentation tasks and domains creatively, which is named "MixSearch". Specifically, we propose a novel approach to mix multiple small-scale datasets from multiple domains and segmentation tasks to produce a large-scale dataset. Then, a novel weaved encoder-decoder structure is designed to search for a generalized segmentation network in both cell-level and network-level. The network produced by the proposed MixSearch framework achieves state-of-the-art results compared with advanced encoderdecoder networks across various datasets. Moreover, we also evaluate the learned network architectures on three additional datasets, which are unseen in the searching process. Extensive experiments show that the architectures automatically learned by our proposed MixSearch surpass U-Net and its variants by a significant margin, verifying the generalization ability and practicability of our proposed method. We make our code and learned network architectures available at: https://github.com/lswzjuer/NAS-WDAN/.

This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of ×4 based on pairs of low and corresponding high resolution images. The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29.00dB on DIV2K validation set. IMDN is set as the baseline for efficiency measurement. The challenge had 3 tracks including the main track (runtime), sub-track one (model complexity),

No abstract