应用材料

Proton-exchange membrane (PEM) water electrolysis is a critical technology for hydrogen production. The oxygen evolution reaction (OER) kinetics at the anode significantly determines the electrolysis performance, requiring the development of active and stable catalysts for high conversion rates. Despite extensive experimental studies, it is still difficult to fully understand how the catalyst state, i.e., the structure, morphology, and oxidation state, which vary by synthesis conditions, affect the OER kinetics and free energies. In this study, we delve into the influence of catalyst calcination on the catalyst state and its relationship with the OER kinetics by a combination of experimental analysis and microkinetic modeling. Our results show that the increasing degree of crystallinity upon calcination and, thus, the reduced number of active sites are the main reason for the decreasing performance of Ir-oxide nanoparticles. Additionally, the water adsorption step becomes thermodynamically more favorable, CUS-mediated PCET and O2 release are modestly hindered, and the bridge-site redox contribution declines with increasing crystallinity. These subtle, systematic shifts help explain the nonlinear structure-activity relationships reported in the literature. This understanding of the interplay between catalyst synthesis conditions and the OER performance facilitates the tailored design and optimization of high-performance catalysts for more efficient electrocatalytic water electrolysis.

The present study focuses on the analysis of elastic scattering in the $\alpha \!-\!d$ system for low-energy data up to 14 $\mathrm{MeV}$. This system holds particular significance due to its direct connection to the $<^>6\mathrm{Li}$ production reaction, which is an important process in nuclear astrophysics and light nucleus synthesis. The primary objective of this investigation is to explore the low-lying excited states of positive parity with isospin $T=0$, which are crucial for accurately describing the resonance behavior and underlying nuclear dynamics of the system. The central idea of this paper is to obtain the interaction potential using the inverse scattering method: we employ the genetic algorithm approach. In this framework, the genetic algorithm is guided by the underlying physical laws, enabling accurate extraction of the inverse scattering potential from the elastic scattering data. In our methodology, as a reference potential, a combination of three smoothly joined Morse functions is utilized, characterized by 10 model parameters. These parameters are optimized in an iterative fashion to ensure the best fit to the phase shifts extracted from the experimental scattering data. The process of optimization is guided by the computed scattering phase shifts by solving the phase equation using the fifth-order Runge-Kutta method for the reference potential in each iteration. Our approach yields inverse potentials for both single- and multichannel scattering. Using the scattering phase shifts obtained from these inverse potentials, we calculate the partial cross section to determine the resonance energies and decay widths. The obtained values for the resonance energies and decay widths for the $<^>3D_1$, $<^>3D_2$, and $<^>3D_3$ states of $\alpha \!-\!d$ are in correspondence with the experimental results.

Moving towards a more holistic approach to disaster risk management, in which a multi-hazard and multi-risk approach is central, offers many opportunities to increase society's resilience. In 2022, we presented a research agenda of six points that could contribute towards this paradigm shift. In this perspective paper we synthesise key learnings from the MYRIAD-EU project – which ran from September 2021 to December 2025 – reflecting on progress and challenges faced in pursuing this research agenda, and share perspectives that may help to further improve multi-hazard and multi-risk assessment and management. Going forward, we point to several avenues for continued scientific research that we feel would benefit the field: continue the mainstreaming and mutual understanding of concepts and definitions; continue developing a strong evidence base of how dynamics in hazard, exposure, and vulnerability in space and time shape multi-risk; further developing methods for providing both current and future multi-hazard and multi-risk scenarios; increasing the availability of appropriate, solutions-oriented, usable tools; more explicitly including equity issues and equitable disaster risk reduction and adaptation; continue extensively testing and coproducing multi-hazard and multi-risk knowledge in in-depth case studies; supporting the development of Multi-Hazard Early Warning Systems; and strengthening opportunities for Early Career Researcher leadership and empowerment within project structures. We suggest concrete ways in which we believe these topics can be addressed in future years and decades.

In this paper, we investigate Row Hammer (RH) failure mechanism in gate-all-around (GAA) nanosheet-based 3D DRAM and analyze both ‘0’ and ‘1' fail disturbances in lateral and vertical neighboring cells. Two primary leakage mechanisms are identified: discharging leakage current from the storage node and capacitive coupling leakage between word lines. Mixed-mode TCAD simulations are employed to analyze the electron current density (ECD) and voltage variations in the victim cells under aggressive pulsing conditions. The comparative study with saddle fin recessed channel access transistor (SRCAT) based DRAM reveals that nanosheet access transistors significantly reduced RH-induced leakage. This is mainly due to the device geometry of the nanosheet which inherently blocks the dominant leakage path between aggressor and victim cells by a common bit line and offers RH induced leakage reductions of 62% in lateral and 71%, 66%, and 64% in vertical neighboring cell spacings of 100 nm, 50 nm and 30 nm respectively, resulting in more effective Row Hammer suppression than SRCAT based DRAM.

The integration of advanced System-on-Chip (SoC) and High Bandwidth Memory (HBM) architectures demands ultra-high interconnect density to meet the bandwidth, latency and energy efficiency requirement for high-performance computing (HPC) applications. Increasing redistribution layer (RDL) density involves not only scaling of line width and space but also minimizing via landing pad critical dimensions (CD). For instance, interconnect density can be increased by more than 4x by shrinking RDL line/space from 2/2 µm to 1/1 µm and landing pad size from 15 µm to 3 µm.This work demonstrates Digital Lithography Technology™ (DLT™) as a production viable solution to achieve 1 µm RDL line width/spacing, 2 µm via CD, and 3 µm landing pad CD with overlay accuracy < 0.25 µm on 300mm wafer. These advancements significantly extend the scaling roadmap for high-density interconnects. By enabling precise scaling of via and landing pad dimensions, DLT™ enables higher interconnect density without compromising packaging process yield and reliability. This breakthrough supports scalable SoC/HBM architectures with improved bandwidth efficiency, reduced power consumption, and enhanced design flexibility, paving the way for next-generation heterogeneous integration.