We report herein a new molecular catalyst for efficient water splitting, aluminum porphyrins (tetra-methylpyridiniumylporphyrinatealuminum: AlTMPyP), containing earth's most abundant metal as the central ion. One-electron oxidation of the aluminum porphyrin initiates the two-electron oxidation of water to form hydrogen peroxide as the primary reaction product with the lowest known overpotential (97 mV). The aluminum-peroxo complex was detected by a cold-spray ionization mass-spectrometry in high-resolution MS (HRMS) mode and the structure of the intermediate species was further confirmed using laser Raman spectroscopy, indicating the hydroperoxy complex of AlTMPyP to be the key intermediate in the reaction. The two-electron oxidation of water to form hydrogen peroxide was essentially quantitative, with a Faradaic efficiency of 99 %. The catalytic reaction was found to be highly efficient, with a turnover frequency up to ∼2×10 s . A reaction mechanism is proposed involving oxygen-oxygen bond formation by the attack of a hydroxide ion on the oxyl-radical-like axial ligand oxygen atom in the one-electron-oxidized form of AlTMPyP(O ) , followed by a second electron transfer to the electrode.

Artificial photosynthesis, which splits water into H2/O2 or reduces CO2 and N2 by visible light, catalyzed by molecular catalysts (MCs) with/without being coupled with a solar cell should serve as one of the most promising renewable energy systems. There still, however, exist bottleneck subjects to be resolved in the MCs’ approach, despite the recent progress. The key subject is the “photon-flux-density problem” of the rarefied sunlight radiation, which leads to a difficulty for the stepwise four-photon excitation of MCs to induce oxygen evolution from water. On the basis of our recent challenges on the two-electron oxidation of water by one-photon visible light excitation of an MC, where the MC does not need to wait for the next photon’s arrival, here we report an artificial photosynthesis system that produces H2 and H2O2 simultaneously on highly earth-abundant element-based aluminum-porphyrins by only one-photon excitation of visible light as the first exemplum to overcome the bottleneck.

In domain adaptation, when there is a large distance between the source and target domains, the prediction performance will degrade. Gradual domain adaptation is one of the solutions to such an issue, assuming that we have access to intermediate domains, which shift gradually from the source to target domains. In previous works, it was assumed that the number of samples in the intermediate domains is sufficiently large; hence, self-training was possible without the need for labeled data. If access to an intermediate domain is restricted, self-training will fail. Practically, the cost of samples in intermediate domains will vary, and it is natural to consider that the closer an intermediate domain is to the target domain, the higher the cost of obtaining samples from the intermediate domain is. To solve the trade-off between cost and accuracy, we propose a framework that combines multifidelity and active domain adaptation. The effectiveness of the proposed method is evaluated by experiments with both artificial and real-world datasets. Codes are available at https://github.com/ssgw320/gdamf.

The Cover Picture shows the reaction mechanism of two‐electron oxidation of water to form hydrogen peroxide initiated by one‐electron oxidation of a new class of molecular catalysts, aluminum porphyrins, containing the Earth's most abundant metal as the central ion. One‐electron oxidation of an aluminum porphyrin initiates the two‐electron oxidation of water to form hydrogen peroxide as the primary reaction product with the lowest known overpotential (97 mV) and a turnover frequency up to ∼2×104 s−1. The finding thus best fits with the Bauxite mine (Al2O3) and water as the background. The picture can be found under, which is licensed under the Creative Commons Attribution‐Share Alike 2.0 Generic license. More details can be found in the Full Paper by Kuttassery et al. on page 1909 in Issue 9, 2017 (DOI: 10.1002/cssc.201700322).

Conventional domain adaptation methods do not work well when a large gap exists between the source and the target domain. Gradual domain adaptation is one of the approaches to address the problem by leveraging the intermediate domain, which gradually shifts from the source to the target domain. The previous work assumed that the number of the intermediate domains is large and the distance of the adjacent domains is small; hence, the gradual domain adaptation algorithm by self-training with unlabeled datasets was applicable. In practice, however, gradual self-training will fail because the number of the intermediate domains is limited, and the distance of the adjacent domains is large. We propose using normalizing flows to mitigate this problem while maintaining the framework of unsupervised domain adaptation. We generate pseudo intermediate domains from normalizing flows and then use them for gradual domain adaptation. We evaluate our method by experiments with real-world datasets and confirm that our proposed method mitigates the above explained problem and improves the classification performance.