Elastocaloric cooling using shape memory alloys (SMAs) offers a green alternative to conventional vapor compression-based technology. However, extending this novel technology to subzero Celsius temperatures is challenging due to the weak caloric effect (typical adiabatic temperature change Delta T <10 K) and severe functional degradation of existing low-temperature superelastic SMAs. Here, we achieved an ultrastable large elastocaloric effect at low temperatures in Ni51.5Ti48.5 SMA by introducing hyperdense Ti3Ni4 nanoprecipitates with average spacing of only 8.2 nm into the phase-transition (PT) matrix. The formation of hyperdense nanoprecipitates reduces the Ni/Ti ratio of the PT matrix to improve the stress-induced PT latent heat and simultaneously creates a strong coherent strain field to significantly suppress the thermally induced PT. These dual benefits enable a considerable caloric effect with Delta T of 9.4-23.7 K in the temperature window of 213-295 K. Furthermore, these nanoprecipitates and the associated strain field significantly strengthen the PT matrix and improve austenite-martensite compatibility. This effectively suppresses dislocation accumulation and residual martensite formation during cyclic deformation, enabling near-zero functional degradation over 10(5) PT cycles in the entire temperature window mentioned above. Our alloy overcomes the key limitations of existing low-temperature superelastic SMAs, paving the way for subzero Celsius elastocaloric refrigeration.
This study designs, fabricates, and evaluates both conventional AC-dispersive inorganic electroluminescent (EL) devices and novel planar electrode-type EL devices. The devices utilize a copper- and chlorine-doped zinc sulfide (ZnS:Cu,Cl) luminescent layer and comb-shaped electrode structures made of ITO, Au, and Pt. The influences of the electrode material conductivity and transparency on the luminescence characteristics and brightness were systematically investigated through comprehensive electrical characterization including impedance analysis, I-V characteristics, and voltage retention measurements. Optical pathway differences between device architectures were analyzed, considering light extraction efficiency and surface plasmon effects. Basic electromagnetic field simulations were performed to visualize field patterns. We discovered that the luminescence characteristics and brightness of the devices were primarily affected by the transparency of the comb electrodes rather than by the conductivity of the electrode materials. The absence of a dielectric layer in comb-electrode devices resulted in reduced brightness due to limited charge accumulation and optical losses. This research provides valuable insights into inorganic EL device design and evaluation methodologies.
A combined ultrasonic method for characterizing micrometer-scale thin films on solid substrates in liquid environments is presented. The method integrates acoustic resonance spectroscopy with pulse-echo measurements to obtain two independent observables: the resonance frequency from frequency-domain analysis and the time difference between echoes reflected from coated and uncoated regions. Using these quantities within a single configuration, film thickness and ultrasonic velocity are simultaneously determined without assuming prior knowledge of acoustic properties. The approach is based on normal-incidence reflection and requires neither dispersion-curve inversion nor calibration using reference specimens. Experiments on polymer and epoxy thin films with thicknesses of approximately 10-25 mu m under water showed good agreement with optical reference measurements. These results show that combining resonance-based frequency information with time-domain echo analysis enables the thickness and ultrasonic velocity of micrometer-scale thin films to be decoupled and identified.
We studied the fundamental phenomena of flow stress variations with deformation temperature and plastic strain during high-temperature deformation of copper. Oxygen-free copper specimens were compressed at room temperature, 573, 673, and 773 K. Dislocation density and texture evolution were evaluated by in-situ neutron diffraction measurements during the compressive deformation. The higher the deformation temperature, the lower the dislocation density during deformation. This was induced by dynamic recovery and recrystallization, which are more likely to occur at higher temperatures. The flow stresses estimated from the Bailey-Hirsch equation based on dislocation densities determined from neutron diffraction measurements reproduced the experimental temperature dependence of the flow stress, including stress oscillations. The dislocation strengthening factor, which determines flow stress from dislocation density, decreased at higher temperatures, suggesting that the frequency of forest interaction between dislocations decreased with increasing deformation temperatures. This may be due to the recovery caused by the dislocation climb. {110} texture developed with the compressive deformation. The higher the deformation temperature, the weaker the texture evolved. Texture evolution in high-temperature deformation proceeded similarly to that in room-temperature deformation until stress oscillations occurred. On the other hand, once stress oscillations began, the texture evolution was retarded. [doi:10.2320/matertrans.MT-D2025009]
Abstract In this study, we propose a knowledge-selective transfer reinforcement learning method that simultaneously achieves heterogeneous domain transfer and knowledge selection in autonomous robots. In recent years, autonomous robots capable of recognition, decision, and action in their environments have been developed, and their utilization is advancing in a wide range of fields such as disaster response and logistics support. Reinforcement learning (RL) enables adaptive learning in unknown environments and is instrumental in realizing technologies such as autonomous robots. However, RL requires extensive exploration, leading to the issue of long training times. To address this, transfer reinforcement learning (TRL) has been introduced to reduce the training time by reusing previously learned knowledge. Nevertheless, the transfer effectiveness depends on the knowledge reused, creating a need for appropriate knowledge selection. Therefore, we propose heterogeneous domain SAP-net (HDSAP-net), which enables knowledge transfer by utilizing heterogeneous-domain knowledge sets acquired from various agents and tasks, thus extending the existing Spreading Activation Policy Network (SAP-net). In its algorithm design, HDSAP-net incorporates inter-task mapping using linear interpolation to bridge discrepancies between the heterogeneous domains. Furthermore, by analyzing the behavior of HDSAP-net, we formulated a network design method using optimal transport cost and implemented a new activation-value control method, thereby improving its performance. This enables TRL, wherein knowledge sets derived from various robots are shared among various agents, which was challenging with the conventional SAP-net. Verification experiments using the physics simulator Webots involving a mobile robot, robotic arm, and drone demonstrated that HDSAP-net significantly improves the learning efficiency when compared with that of conventional RL. The proposed method reduced the learning time by 32.1% to 82.0%, confirming its capability of autonomously discovering and utilizing effective knowledge across various domains.
