TCDformer: A transformer framework for non-stationary time series forecasting based on trend and change-point detection.

Although time series prediction models based on Transformer architecture have achieved significant advances, concerns have arisen regarding their performance with non-stationary real-world data. Traditional methods often use stabilization techniques to boost predictability, but this often results in the loss of non-stationarity, notably underperforming when tackling major events in practical applications. To address this challenge, this research introduces an innovative method named TCDformer (Trend and Change-point Detection Transformer). TCDformer employs a unique strategy, initially encoding abrupt changes in non-stationary time series using the local linear scaling approximation (LLSA) module. The reconstructed contextual time series is then decomposed into trend and seasonal components. The final prediction results are derived from the additive combination of a multilayer perceptron (MLP) for predicting trend components and wavelet attention mechanisms for seasonal components. Comprehensive experimental results show that on standard time series prediction datasets, TCDformer significantly surpasses existing benchmark models in terms of performance, reducing MSE by 47.36% and MAE by 31.12%. This approach offers an effective framework for managing non-stationary time series, achieving a balance between performance and interpretability, making it especially suitable for addressing non-stationarity challenges in real-world scenarios.

Water-assisted hydrogen spillover in Pt nanoparticle-based metal-organic framework composites.

Hydrogen spillover is the migration of activated hydrogen atoms from a metal particle onto the surface of catalyst support, which has made significant progress in heterogeneous catalysis. The phenomenon has been well researched on oxide supports, yet its occurrence, detection method and mechanism on non-oxide supports such as metal-organic frameworks (MOFs) remain controversial. Herein, we develop a facile strategy for efficiency enhancement of hydrogen spillover on various MOFs with the aid of water molecules. By encapsulating platinum (Pt) nanoparticles in MOF-801 for activating hydrogen and hydrogenation of C=C in the MOF ligand as activated hydrogen detector, a research platform is built with Pt@MOF-801 to measure the hydrogenation region for quantifying the efficiency and spatial extent of hydrogen spillover. A water-assisted hydrogen spillover path is found with lower migration energy barrier than the traditional spillover path via ligand. The synergy of the two paths explains a significant boost of hydrogen spillover in MOF-801 from imperceptible existence to spanning at least 100-nm-diameter region. Moreover, such strategy shows universality in different MOF and covalent organic framework materials for efficiency promotion of hydrogen spillover and improvement of catalytic activity and antitoxicity, opening up new horizons for catalyst design in porous crystalline materials.

Catalytic Modification of Porous Two-Dimensional Ni-MOFs on Portable Electrochemical Paper-Based Sensors for Glucose and Hydrogen Peroxide Detection.

Rapid and accurate detection of changes in glucose (Glu) and hydrogen peroxide (H2O2) concentrations is essential for the predictive diagnosis of diseases. Electrochemical biosensors exhibiting high sensitivity, reliable selectivity, and rapid response provide an advantageous and promising solution. A porous two-dimensional conductive metal-organic framework (cMOF), Ni-HHTP (HHTP = 2,3,6,7,10,11-hexahydroxytriphenylene), was prepared by using a one-pot method. Subsequently, it was employed to construct enzyme-free paper-based electrochemical sensors by applying mass-producing screen-printing and inkjet-printing techniques. These sensors effectively determined Glu and H2O2 concentrations, achieving low limits of detection of 1.30 µM and 2.13 µM, and high sensitivities of 5573.21 µA µM-1 cm-2 and 179.85 µA µM-1 cm-2, respectively. More importantly, the Ni-HHTP-based electrochemical sensors showed an ability to analyze real biological samples by successfully distinguishing human serum from artificial sweat samples. This work provides a new perspective for the use of cMOFs in the field of enzyme-free electrochemical sensing, highlighting their potential for future applications in the design and development of new multifunctional and high-performance flexible electronic sensors.

Fabrication of Two-Dimensional Metal-Organic Framework Nanosheets through Crystal Dissolution-Growth Kinetics.

Controlling the morphology of the metal-organic framework (MOF) for nanosheets is beneficial for understanding their crystal growth kinetics and useful for extending these MOF nanosheets to advanced applications, in particular for gas separation and device integration. However, synthesizing MOF nanosheets with uniform thickness or desirable size still remains challenging. Herein, we provide a crystal dissolution-growth strategy for fabricating dispersible porphyrin MOF nanosheets with lateral dimensions and nanometer thickness. A morphological transition (bulk crystals-nanosheets-bulk crystals) in Zn-TCPP was observed when controlling the crystal growth kinetics by adjusting the reaction parameters (temperature and acidity). These findings encouraged the synthesis of other types of nanosheets (Cu-TCPP, Zn-TCPP (Pd), and Cu-BDC nanosheets). Zn-TCPP (Pd) nanosheets were applied in field-effect transistors and exhibited photoresponse characteristics. This work demonstrates a new strategy for obtaining MOF nanosheets and casts a new light upon fabricating two-dimensional inorganic-organic hybrid materials with controlled thickness.