No abstract

Photodetectors based on reduced graphene oxide (rGO) have attracted much attention owing to their simple and low‐cost fabrication process. However, the aggregation and defects of rGO flakes still limit the performance of rGO photodetectors. Controlling the composition of rGO has become a vital factor for its prospective applications. For example, the interconnection between rGO and polymers for modified morphologies of rGO films leads to an enhanced performance of devices. In this work, a practical approach to engineer surface uniformity and enhance the performance of a photodetector by modifying the rGO film with hydrophilic polymers poly(vinyl alcohol) (PVA) is reported. Compared with the rGO photodetector, the on/off ratio for the PVA/rGO photodetector shows 3.5 times improvement, and the detectivity shows 53% enhancement even when the photodetector is operated at a low bias of 0.3 V. This study provides an effective route to realize PVA/rGO photodetectors with a low‐power operation which shows promising opportunities for the future development of green systems.

In this work, Ga2O3 nanorods were converted from GaOOH nanorods grown using the hydrothermal synthesis method as the sensing membranes of NO2 gas sensors. Since a sensing membrane with a high surface-to-volume ratio is a very important issue for gas sensors, the thickness of the seed layer and the concentrations of the hydrothermal precursor gallium nitrate nonahydrate (Ga(NO3)3·9H2O) and hexamethylenetetramine (HMT) were optimized to achieve a high surface-to-volume ratio in the GaOOH nanorods. The results showed that the largest surface-to-volume ratio of the GaOOH nanorods could be obtained using the 50-nm-thick SnO2 seed layer and the Ga(NO3)3·9H2O/HMT concentration of 12 mM/10 mM. In addition, the GaOOH nanorods were converted to Ga2O3 nanorods by thermal annealing in a pure N2 ambient atmosphere for 2 h at various temperatures of 300 °C, 400 °C, and 500 °C, respectively. Compared with the Ga2O3 nanorod sensing membranes annealed at 300 °C and 500 °C, the NO2 gas sensors using the 400 °C-annealed Ga2O3 nanorod sensing membrane exhibited optimal responsivity of 1184.6%, a response time of 63.6 s, and a recovery time of 135.7 s at a NO2 concentration of 10 ppm. The low NO2 concentration of 100 ppb could be detected by the Ga2O3 nanorod-structured NO2 gas sensors and the achieved responsivity was 34.2%.

Recently, researchers utilize Knowledge Graph (KG) as side information in recommendation system to address cold start and sparsity issue and improve the recommendation performance. Existing KG-aware recommendation model use the feature of neighboring entities and structural information to update the embedding of currently located entity. Although the fruitful information is benecial to the following task, the cost of exploring the entire graph is massive and impractical. In order to reduce the computational cost and maintain the pa ern of extracting features, KG-aware recommendation model usually utilize xed-size and random set of neighbors rather than complete information in KG. Nonetheless, there are two critical issues in these approaches: First of all, xedsize and randomly selected neighbors restrict the view of graph. In addition, as the order of graph feature increases, the growth of parameter dimensionality of the model may lead the training process hard to converge. To solve the aforementioned limitations, we propose GraphSW, a strategy based on stage-wise training framework which would only access to a subset of the entities in KG in every stage. During the following stages, the learned embedding from previous stages is provided to the network in the next stage and the model can learn the information gradually from the KG. We apply stage-wise training on two SOTA recommendation models, RippleNet and Knowledge Graph Convolutional Networks (KGCN). Moreover, we evaluate the performance on six real world datasets, Last.FM 2011, Book-Crossing,movie, LFM-1b 2015, Amazon-book and Yelp 2018. e result of our experiments shows that proposed strategy can help both models to collect more information from the KG and improve the performance. Furthermore, it is observed that GraphSW can assist KGCN to converge e ectively in high-order graph feature.

No abstract