Search

Experiments have shown that crystalline–amorphous nanolayered composites show a strong size dependence in strength and plasticity. However, the underlying mechanism remains unknown. Here, the layer thickness (h)-dependent compressive strength and plastic deformation of Nb/amorphous CuNb nanolayered composites with h ranging from 2.8 to 20 nm have been studied by molecular dynamics simulations. It is found that the strength increases monotonically with the decrease of h, which can be well captured by the refined confined layer slip model. Furthermore, the shear banding-induced strain localization was analyzed in detail based on the evolution of the von Mises strain distribution with the applied strain and a strain localization parameter that represents the deviation of the specific atomic Mises strain from the average one of all atoms. The results show that the strain localization of the composite shows a significant size dependence and its magnitude increases with the decrease in the layer thickness. The prevention of shear band propagation in thicker layer samples is attributed to that the amorphous phase is capable of forming an atomic vortex to alleviate the strain concentration caused by dislocation absorption. The thicker amorphous layers accommodate enhanced homogeneous plasticity than the thinner ones by forming a larger vortex. Note that the above size dependence of strength and strain localization in the composite agrees well with existing experimental measurements and observations in the layer thickness range considered. As a result, the present work gives a deeper insight into the understanding of the size-dependent strengthening and strain localization mechanism in the amorphous/crystalline composites. [ABSTRACT FROM AUTHOR]

Aiming at the discharge problem of inductively coupled plasma (ICP), a closed wave-transmitting cavi-ty configuration is designed, the fluid mechanics model of ICP is established, and the multi physical simulation software COMSOL is used for three-dimensional simulation research. The variation laws of typical parameters such as electron density, electron temperature and plasma potential with time in the discharge process are analyzed. On this basis, the power of radio frequency power supply, the air pressure in the cavity and other conditions are changed to obtain the parameter range and spatial gradient distribution of ICP. The results show that the discharge power mainly affects the numerical range of plasma parameters, and the change of air pressure has an effect on the numerical value and spatial distribution of plasma parameters. When the power of radio frequency power supply increases from 100 W to 300 W, the peak electron density increases from 2.54×1017 m-3 to 5.68×1017 m-3, and the electric potential induced in the heating zone and the corresponding electron temperature decrease slightly after the reaction is stable. Increasing the air pressure in the cavity within a certain range will significantly increase the electron density. The air pressure increases from 10 Pa to 30 Pa, and the electron density increases from 9.12×1017 m-3 to 3.62×1018 m-3. However, the uniformity of plasma parameter distribution will deteriorate if the air pressure is too high.

Summary: Dragonfly topology is a commonly utilized design for interconnection networks in parallel and distributed systems. A classical dragonfly can be denoted as dragonfly(k$$ k $$,m$$ m $$,l$$ l $$), wherem$$ m $$ is the number of routers in a group,l$$ l $$ is the number of links per router connected to other groups, andk$$ k $$ is the number of links per router connected to compute nodes. Each router has otherm−1$$ m-1 $$ links fully connected to otherm−1$$ m-1 $$ routers within a group. Each group hasml$$ ml $$ links connected to other groups. The groups are also fully connected, therefore there areml+1$$ ml+1 $$ groups in total. The router radix in a dragonfly(k$$ k $$,m$$ m $$,l$$ l $$) isl+k+m−1$$ l+k+m-1 $$. Building a large dragonfly system requires a large number of high‐radix routers, increasing hardware costs. To reduce hardware costs, this paper proposes a more flexible topology called dimension‐extended dragonfly (DED). Rather than routers in a group being fully connected, each router in a group is arranged in ann$$ n $$‐dimensional matrix, and routers of the same dimension are fully connected. We usen$$ n $$ to denote the dimension such that each group in the DED hasmn$$ {m}^n $$ routers. This study comprehensively evaluates DED in terms of cost, performance, fault tolerance, and packet latency. The findings show that DED provides a more economical hardware solution compared to traditional Dragonfly and Cascade topologies, especially forn≥3$$ n\ge 3 $$. Beyond cost‐efficiency, DED enhances system design flexibility. It offers diverse possibilities for system scaling through different combinations of diameter and radix, giving system architects more adaptable options. To further enhance the versatility of DED, three disjoint path routing algorithms are proposed and their fault tolerance is evaluated through simulation. The simulation results also show that the packet latency of DED is lower than dragonfly and cascade. [ABSTRACT FROM AUTHOR]