DFT service costs uplifted for first time in a decade
Related Articles
First-principles and machine-learning approaches for interpreting and predicting the properties of MXenes
MXenes are a versatile family of 2D inorganic materials with applications in energy storage, shielding, sensing, and catalysis. This review highlights computational studies using density functional theory and machine-learning approaches to explore their structure (stacking, functionalization, doping), properties (electronic, mechanical, magnetic), and application potential. Key advances and challenges are critically examined, offering insights into applying computational research to transition these materials from the lab to practical use.
Water and wastewater infrastructure inequity in unincorporated communities
Uneven access to water and wastewater infrastructure is shaped by local governance. A substantial number of U.S. households lack adequate access and the U.S. is one of the few countries with large populations living outside of city bounds, in unincorporated areas. Few studies address how infrastructure services and local governance are intertwined at a regional scale. We examine the connection between incorporation status and access to centralized infrastructure, using negative binomial regression. A novel dataset informs this analysis, comprised of 31,383 Census block groups located in nine states representing over 25% of the national population. We find evidence that inequities in access are associated with unincorporated status and poverty rates. Sewer coverage rates are significantly lower for unincorporated communities in close proximity to municipal boundaries. Infrastructure equity could be improved by targeting high-poverty unincorporated communities, addressing challenges with noncontiguous service areas, and strengthening regional water planning and participatory governance.
Accelerating crystal structure search through active learning with neural networks for rapid relaxations
Global optimization of crystal compositions is a significant yet computationally intensive method to identify stable structures within chemical space. The specific physical properties linked to a three-dimensional atomic arrangement make this an essential task in the development of new materials. We present a method that efficiently uses active learning of neural network force fields for structure relaxation, minimizing the required number of steps in the process. This is achieved by neural network force fields equipped with uncertainty estimation, which iteratively guide a pool of randomly generated candidates toward their respective local minima. Using this approach, we are able to effectively identify the most promising candidates for further evaluation using density functional theory (DFT). Our method not only reliably reduces computational costs by up to two orders of magnitude across the benchmark systems Si16, Na8Cl8, Ga8As8 and Al4O6 but also excels in finding the most stable minimum for the unseen, more complex systems Si46 and Al16O24. Moreover, we demonstrate at the example of Si16 that our method can find multiple relevant local minima while only adding minor computational effort.
Opportunities and challenges in modelling ligand adsorption on semiconductor nanocrystals
Semiconductor nanocrystals, including their superstructures and hybridized systems, have opened up a new realm to design next-generation functional materials creatively. Their great success and unlimited potential should be largely attributed to surface-adsorbed ligands. However, due to a lack of means to probe and understand their roles in experiments, only a handful of effective ligands have been identified through trial-and-error processes. Alternatively, computational and theoretical methods are ideal for providing physical insights and further guidance. Still, their applications in ligand-coated semiconductor nanocrystals are relatively scarce compared to those of other systems, such as biological chemistry. In this perspective, we first highlight the success of ab initio methods in modeling ligand adsorption. Then, we discuss the opportunities of molecular dynamics and theory in accommodating complex colloidal nature, where we unfold the challenges therein. Finally, we emphasize the need for high-quality force fields to resolve these challenges and look forward to simulation-guided inverse design.
Revisiting the origin of non-volatile resistive switching in MoS2 atomristor
Recently, Non-Volatile Resistive Switching (NVRS) has been demonstrated in Metal-monolayer MoS2-Metal atomristors. While experiments based on Au metal report the origin of NVRS to be extrinsic, caused by the Au atom adsorption into sulfur vacancies, however, more recently molecular dynamics based on reactive forcefield (ReaxFF) suggest that both monolayer and multilayer MoS2 can also host intrinsic non-volatile resistive states whereby an S atom at a monosulfur vacancy (parent state) pops into the molybdenum plane (popped state) under applied out-of-plane electric field. Our rigorous computations based on Density Functional Theory (DFT) and M3GNet (deep learned forcefield) to carry out structural relaxations and molecular dynamics reveal that such a popped state is unstable and does not represent any intrinsic non-volatile resistive state. This is in contrast with the ReaxFF used in previous studies which inaccurately describes the Potential Energy Surface (PES) of MoS2 around the popped state. More importantly, Au atom adsorbed at a sulfur vacancy in MoS2 atomristors represents a stable non-volatile resistive state which is in excellent agreement with earlier experiment. Furthermore, it is observed that the local heating generated around the adsorbed Au atom in low resistive state leads to cycle-to-cycle variability in MoS2 atomristors.
Responses