Strain regulation has become an important strategy to tune the surface chemistry and optimize the catalytic performance of nanocatalysts. Herein, the construction of atomic‐layer IrOx on IrCo nanodendrites with tunable IrO bond length by compressive strain effect for oxygen evolution reaction (OER) in acidic environment is demonstrated. Evidenced from in situ extended X‐ray absorption fine structure, it is shown that the compressive strain of the IrOx layer on the IrCo nanodendrites decreases gradually from 2.51% to the unstrained state with atomic layer growth (from ≈2 to ≈9 atomic layers of IrOx), resulting in the variation of the IrO bond length from shortened 1.94 Å to normal 1.99 Å. The ≈3 atomic‐layer IrOx on IrCo nanodendrites with an IrO bond length of 1.96 Å (1.51% strain) exhibits the optimal OER activity compared to the higher‐strained (2.51%, ≈2 atomic‐layer IrOx) and unstrained (>6 atomic‐layer IrOx) counterparts, with an overpotential of only 247 mV to achieve a current density of 10 mA cm−2. Density functional theory calculations reveal that the precisely tuned compressive strain effect balances the adsorbate–substrate interaction and facilitates the rate‐determining step to form HOO*, thus assuring the best performance of the three atomic‐layer IrOx for OER.

The synthesis and bioimaging of multicolor carbon dots from citric acid and urea.

Different from conventional microimaging techniques, polarization imaging can generate multiple polarization images in a single perspective by changing the polarization angle. However, how to efficiently fuse the information in these multiple polarization images by a convolutional neural network (CNN) is still a challenging problem. In this paper, we propose a hybrid 3D–2D convolutional neural network called MuellerNet, to classify biological cells with Mueller matrix images (MMIs). The MuellerNet includes a normal stream and a polarimetric stream, in which the first Mueller matrix image is taken as the input of normal stream, and the rest MMIs are stacked to form the input of a polarimetric stream. The normal stream is mainly constructed with a backbone network and, in the polarimetric stream, the attention mechanism is used to adaptively assign weights to different convolutional maps. To improve the network’s discrimination, a loss function is introduced to simultaneously optimize parameters of the two streams. Two Mueller matrix image datasets are built, which include four types of breast cancer cells and three types of algal cells, respectively. Experiments are conducted on these two datasets with many well-known and recent networks. Results show that the proposed network efficiently improves the classification accuracy and helps to find discriminative features in MMIs.