Search for: All records

Abstract The concept of employing highly concentrated electrolytes has been widely incorporated into electrolyte design, due to their enhanced Li‐metal passivation and oxidative stability compared to their diluted counterparts. However, issues such as high viscosity and sub‐optimal wettability, compromise their suitability for commercialization. In this study, we present a highly concentrated dimethyl ether‐based electrolyte that appears as a liquid phase at ambient conditions via Li⁺‐ solvents ion‐dipole interactions (Coulombic condensation). Unlike conventional high salt concentration ether‐based electrolytes, it demonstrates enhanced transport properties and fluidity. The anion‐rich solvation structure also contributes to the formation of a LiF‐rich salt‐derived solid electrolyte interphase, facilitating stable Li metal cycling for over 1000 cycles at 0.5 mA cm⁻², 1 mAh cm⁻²condition. When combined with a sulfurized polyacrylonitrile (SPAN) electrode, the electrolyte effectively reduces the polysulfide shuttling effect and ensures stable performance across a range of charging currents, up to 6 mA cm⁻². This research underscores a promising strategy for developing an anion‐rich, high concentration ether electrolyte with decreased viscosity, which supports a Li metal anode with exceptional temperature durability and rapid charging capabilities. 

Despite their tremendous success in a range of domains, deep learning systems are inherently susceptible to two types of manipulations: adversarial inputs -- maliciously crafted samples that deceive target deep neural network (DNN) models, and poisoned models -- adversely forged DNNs that misbehave on pre-defined inputs. While prior work has intensively studied the two attack vectors in parallel, there is still a lack of understanding about their fundamental connections: what are the dynamic interactions between the two attack vectors? what are the implications of such interactions for optimizing existing attacks? what are the potential countermeasures against the enhanced attacks? Answering these key questions is crucial for assessing and mitigating the holistic vulnerabilities of DNNs deployed in realistic settings. Here we take a solid step towards this goal by conducting the first systematic study of the two attack vectors within a unified framework. Specifically, (i) we develop a new attack model that jointly optimizes adversarial inputs and poisoned models; (ii) with both analytical and empirical evidence, we reveal that there exist intriguing "mutual reinforcement" effects between the two attack vectors -- leveraging one vector significantly amplifies the effectiveness of the other; (iii) we demonstrate that such effects enable a large design spectrum for the adversary to enhance the existing attacks that exploit both vectors (e.g., backdoor attacks), such as maximizing the attack evasiveness with respect to various detection methods; (iv) finally, we discuss potential countermeasures against such optimized attacks and their technical challenges, pointing to several promising research directions.