Noble-metal-free CoMn2O4/Sr3Fe2O7-s/alpha-Al2O3 catalysts were developed and evaluated for catalytic methane oxidation. The incorporation of CoMn2O4 as the activator and Sr3Fe2O7-s as the promoter effectively promoted surface oxygen activation, redox cycling, and oxygen vacancy formation. The use of gamma-Al2O3 as the support precursor resulted in a relatively high specific surface area after transformation to alpha-Al2O3 and the phase transition stabilized the Sr3Fe2O7-s structure during preparation. The highest catalytic activity was obtained for 7 wt%CoMn2O4/Sr3Fe2O7-s/alpha-Al2O3, which achieved 100% CH4 conversion at 450 degrees C. These results demonstrate the potential of spinel-perovskite composite catalysts as promising noble-metal-free systems for low-temperature methane oxidation.
Two-dimensional (2D) materials integrating photoelectrochemical (PEC) and thermoelectric (TE) functionalities hold great promise for sustainable energy technologies. This work presents a comprehensive first-principles study of alpha- and (1-GeP monolayers, demonstrating their unprecedented coupling of strain-adaptive PEC and TE performance. Both phases exhibit dynamic and thermal stability, with (1-GeP showing superior mechanical strength. Under strain, alpha-GeP retains an indirect bandgap, while beta-GeP undergoes a semiconductor-to-metal transition. Optical absorption, especially in (1-GeP, is highly responsive to biaxial strain. PEC analysis shows that alpha-GeP sustains band-edge alignment up to 16 % strain, while (1-GeP achieves a solar-to-hydrogen efficiency of 17.14 %, surpassing typical 2D material benchmarks. Thermoelectric performance is equally notable, with Seebeck coefficients exceeding 2100 mu V/K and near-unity ZT for alpha-GeP. These findings demonstrate the synergistic control of structure and strain, positioning GeP monolayers, particularly alpha-GeP, as leading multifunctional platforms for next-generation, strain-adaptive PEC-TE energy harvesting applications.
Natural gypsum in ordinary Portland cement was partially substituted with ferrous sulfate hydrates synthesized from Linz-Donawitz steelmaking sludge and waste sulfuric acid through a cyclic reaction filtration process. Replacement levels of 25%, 50%, 75%, and 100% were evaluated, and the 50% replacement mixture was identified as the optimal level based on mechanical and microstructural performance. The optimal mixture achieved compressive strengths of 37.1, 40.9, 44.2, and 47.5 MPa at 3, 7, 14, and 28 days, representing a 12.3% improvement relative to the control. Setting-time measurements indicated a progressive but manageable delay, increasing from 111/152 min (initial/final) in control to 120/161 min at 25% replacement and 154/191 min at 50% replacement. X-ray diffraction and thermogravimetric analyses confirmed accelerated ettringite formation, moderated AFm development, and sustained portlandite content, resulting in a refined pore structure. The ferrous sulfate also reduced hexavalent chromium to trivalent chromium, providing additional occupational and environmental benefits. A cradle-to-gate life-cycle assessment revealed that substituting natural gypsum with waste-derived ferrous sulfate hydrates reduced the global warming potential by approximately 52%, primarily due to avoided gypsum extraction and lower upstream neutralization burdens. Overall, up to 50% of natural gypsum in Portland cement can be replaced with waste-derived ferrous sulfate hydrates while maintaining acceptable setting characteristics, improving mechanical performance, and significantly reducing environmental impacts.
The structural stability, electronic response, and thermal reusability of a hexagonal zinc oxide (ZnO) nanosheet for the detection of dimethyl methylphosphonate (DMMP), a nerve-agent simulant, were investigated using density functional theory (DFT) calculations and molecular dynamics (MD) simulations with machine-learning interatomic potentials (MLIPs). Strong chemisorption of DMMP on the ZnO surface is observed via a monodentate configuration, in which the phosphoryl oxygen atom (O_DMMP) directly bonds to a surface Zn atom. This interaction is highly exothermic, with a short O_DMMP-Zn bond length of 1.97 angstrom and an adsorption energy of -1.06 eV. Electronic structure analysis, including electron localization function (ELF) and projected density of states (PDOS), reveals that hybridization between O p orbitals of DMMP and Zn d orbitals leads to the formation of a polar covalent dative bond. Despite the strong adsorption, the ZnO nanosheet exhibits limited electrical sensitivity for chemi-resistive sensing, as indicated by a modest Fermi level shift (Delta EF = 0.43 eV) and a negligible band gap variation (Delta Eg approximate to -1.21%). Recovery time analysis suggests that mild thermal treatment (similar to 121 degrees C) can enable rapid sensor regeneration, although strong binding results in a prolonged recovery time (similar to 40 h) at 300 K under the worst-case assumption (E-des = vertical bar E-ads vertical bar). Furthermore, MLIPs-based MD simulations using the CHGNet framework were employed to investigate competitive adsorption between DMMP and HBO on the ZnO monolayer. The results demonstrate that DMMP preferentially occupies Zn Lewis acid sites at 300 K due to its higher kinetic sticking probability and thermodynamic favorability compared to water. MD trajectories reveal a hopping diffusion mechanism, in which DMMP migrates between adjacent Zn sites across the surface. A stepwise thermal decomposition pathway is identified during temperature ramping from 300 to 1000 K, initiated by hydrogen dissociation at similar to 800 K (51.43 ps), followed by C-O bond anchoring at similar to 56.70 ps. In contrast, water molecules desorb molecularly without dissociation over the temperature range of 700-1000 K. This study provides important insights into the sensing performance, thermal limitations, and potential poisoning mechanisms of ZnO-based gas sensors, while also demonstrating the capability of MLIPs to accurately capture complex organophosphorus-metal oxide interactions.
In the era of artificial intelligence (AI), machine learning interatomic potentials (MLIPs) have revolutionized materials science and engineering, enabling large-scale and accurate atomistic modeling in various chemical systems. The performance of these potentials hinges on two key components: the model architecture, which defines the mathematical representation of atomic interactions, and the training data set, which supplies foundational knowledge. To scrutinize recent advances and chart the course for future development, this review presents a comprehensive analysis of reactive MLIPs, focusing on these two pillars. We first explore the evolution of model architectures, from descriptor-based models to state-of-the-art equivariant graph neural networks, critically assessing the essential requisites for achieving both physical accuracy and computational efficiency. Concurrently, we examine the sophisticated data acquisition strategies required to capture the complexities of chemical reactions, with a particular focus on uncertainty-driven active learning for sampling high-energy transition states and reaction pathways. Finally, beyond reviewing current methodologies, we provide an outlook on domain-specific challenges and discuss emerging opportunities in next-generation technologies, such as generative AI and cognitive autonomous agents. We believe this review will serve as a valuable guideline for practitioners in selecting and constructing MLIPs tailored for the study of chemical reactivity.
