A facile one-step strategy for anchoring defective CoN single clusters on partly reduced graphene oxide (RGO) is constructed to significantly improve the catalytic performance of non-noble metal complexes toward oxygen reduction reaction (ORR). Sequent loading with trace amounts of metal-free porphyrin and Co in RGO can dramatically enhance both the half-wave potential and the peak current density. Intriguingly, the RGO/P/2Co single cluster exhibits the best ORR catalytic performance with the half-wave potential of 0.834 V, extremely approaching that of commercial Pt/C (0.836 V). This half-wave potential surpasses most of the reported half-wave potentials of RGO supported non-noble metal ORR catalysts through low-temperature synthesis. Furthermore, the as-prepared RGO/P/2Co delivers a peak current density of 1.3 times higher than that of Pt/C at the same loading, together with a high mass activity of 2.76 A mg. During the durability test, a cathodic current loss less than 10% is recorded after 8000 continuous potential cycles. Insights into this successful example will be conducive to the development of elegant routes for constructing metal nitrogen (MN)-based ORR catalysts with high efficiency, outstanding stability, and excellent selectivity.

No abstract

No abstract

Water-soluble and biocompatible protein-protected gold nanoclusters (Au NCs) hold great promise for numerous applications. However, design and precise regulation of their structure at an atomic level remain challenging. Herein, we have engineered and constructed a gold clustering site at the 4-fold symmetric axis channel of the apo-ferritin cage. Using a series of X-ray crystal structures, we evaluated the stepwise accumulation process of Au ions into the cage and the formation of a multinuclear Au cluster in our designed cavity. We also disclosed the role of key residues in the metal accumulation process. X-ray crystal structures in combination with quantum chemical (QC) calculation revealed a unique Au clustering site with up to 12 Au atoms positions in the cavity. Moreover, the structure of the gold nanocluster was precisely tuned by the dosage of the Au precursor. As the gold concentration increases, the number of Au atoms position at the clustering site increases from 8 to 12, and a structural rearrangement was observed at a higher Au concentration. Furthermore, the binding affinity order of the four Au binding sites on apo-ferritin was unveiled with a stepwise increase of Au precursor concentration.

During the past decades, due to the lack of sufficient labeled data, most studies on crossdomain parsing focus on unsupervised domain adaptation, assuming there is no targetdomain training data. However, unsupervised approaches make limited progress so far due to the intrinsic difficulty of both domain adaptation and parsing. This paper tackles the semi-supervised domain adaptation problem for Chinese dependency parsing, based on two newly-annotated large-scale domain-specific datasets. 1 We propose a simple domain embedding approach to merge the sourceand target-domain training data, which is shown to be more effective than both direct corpus concatenation and multi-task learning. In order to utilize unlabeled target-domain data, we employ the recent contextualized word representations and show that a simple fine-tuning procedure can further boost cross-domain parsing accuracy by large margins. * Corresponding author 1 The two domain-specific datasets, plus another one for product comment texts, are also used in the NLPCC-2019 shared task (http://hlt.suda.edu.cn/index. php/Nlpcc-2019-shared-task) on cross-domain Chinese dependency parsing. Please note that the settings for the source-domain training data are different between this work and NLPCC-2019 shared task.