Marie Juramy

Lithium metal batteries (LMBs) show strong potential for next-generation energy storage, because of their high theoretical energy density. However, their practical deployment is limited by microstructure and dendrite formation, which raises serious safety and efficiency concerns. A key factor influencing dendrite growth is the passivating film, or solid electrolyte interphase (SEI), that forms at the lithium metal surface. Gaining insight into the SEI is essential for controlling dendrite formation, yet this remains challenging due to its complex composition and disorder. Lithium metal endogenous Dynamic Nuclear Polarization (DNP) offers opportunities to probe the SEI with enhanced sensitivity. In this work, we investigate how dendrite morphology influences both DNP enhancement and polarization transfer mechanisms. We develop a protocol to produce lithium dendrites with controlled morphologies on copper electrodes, and find that DNP efficiency is strongly affected by dendrite morphology and the substate they grow on, primarily through variations in the electron spin relaxation times. For Li dendrites grown on a copper substrate, extremely short relaxation times are seen for some morphologies, which we suggest is due largely to the Cu impurities present at the Li/SEI interface since the effect is more pronounced for the thin dendrites, where electrons travel for shorter distances before encountering a surface impurity. The enhancements also depend on the frequency of the applied microwave radio frequency (rf) irradiation, and the extent of rf penetration into the lithium microstructures. Finally, we explore polarization transfer from lithium metal to surrounding salt species in the SEI, providing insight into the underlying mechanisms.

Study of nucleation by solid-state nuclear magnetic resonance and dynamic nuclear polarization: La cristallisation est un processus connu depuis des siècles et joue un rôle important dans de nombreux domaines, malheureusement il reste encore mal compris. L'identification des phases solides produites pendant le processus de cristallisation ainsi que la détection de la formation des premiers cristaux sont des étapes cruciales pour améliorer notre compréhension de ce processus. Cependant, l'étude de la formation des premiers cristaux nécessite à la fois la détection des phases solides transitoires, ce qui est extrêmement difficile en raison de leur petite taille et de leur courte durée de vie, mais aussi le suivi de l'évolution des phases liquides et solides en fonction du temps. Nous avons développé une méthodologie utilisant la spectroscopie RMN DNP pour identifier les phases solides produites pendant le processus de cristallisation, et dans laquelle le processus est rapidement refroidis à basse température à un moment spécifique de la cristallisation. La combinaison de cette méthodologie avec les avantages de la cristallisation en milieu confiné a ensuite été utilisée pour accéder aux premières étapes de la cristallisation. Les résultats montrent que l'amélioration de la sensibilité de la RMN fournie par la DNP permet la détection de phases transitoires qui ne seraient pas observables avec des expériences traditionnelles de RMN du solide. L'efficacité de ces stratégies a été démontré sur un composé modèle. Par la suite, la généralisation de ces stratégies présente un fort potentiel pour fournir un outil permettant de détecter les propriétés polymorphiques d'une large gamme de molécules ainsi que d'étudier les premières étapes de leur cristallisation

NMR spectroscopy reveals LiNO ₃ -induced SEI restructuring, though increasing its concentration yields minimal further performance improvements.

Dynamic nuclear polarization (DNP) of the Li 7 NMR lithium metal signal via an Overhauser mechanism was predicted and demonstrated over 70 years ago, and applied more recently at high fields and room temperature to study electrodeposited microstructural lithium metal. Here we further explore the requirements and limitations of the Overhauser effect for alkali metals. Following a theoretical treatment of the Overhauser mechanism, we explain why only relatively modest enhancements are obtained ( ∼ 50 compared to a theoretical maximum of 1700 for Li 7 ) at a magnetic field strength of 9.4 T. Through a systematic investigation of the Li 7 enhancement as a function of experimental parameters (magnetic field strength, B 0 , temperature, microwave power, and MAS frequency), we show that the experimental enhancements are limited by the short spin-lattice relaxation times ( T 1 e and T 2 e ) of the electrons in lithium metal, and thus the degree of saturation of the electron spin resonance (ESR) transition. The enhancements of Li 6 and Na 23 are similarly limited by the degree of electron spin saturation, the effect being more pronounced for Na 23 ; enhancements are improved by using higher microwave power and for Na 23 , with its temperature-dependent T 1 e , by performing experiments at lower temperatures. Sample considerations, such as sample volume, metal particle sizes, and dilution of metal particles, are also important for obtaining high enhancements. Studies of the hyperpolarization mechanism of the spins in diamagnetic environment at the interfaces between lithium metal and its passivating film suggest that both spin diffusion and direct chemical exchange play a role, although the role of direct cross-relaxation with the conduction electrons requires further investigation. The hyperpolarization is then transferred to spins farther from the metal surface via spin diffusion, allowing only the spins within less than ∼ 10 nm from the metal surface to be hyperpolarized. Based on the insights gained through this work, lithium metal possesses a possibly unique combination of beneficial qualities, making it a highly suitable system for Overhauser DNP. The high enhancements achieved for Li 6 ( ∼ 150 ) using a 1.3 mm rotor are particularly promising for studying lithium metal samples extracted from batteries, making use of the generally higher resolution of Li 6 over Li 7 .

(Text in French) Watch the invisible: an NMR method to decipher the first steps of crystallization. Crystallization, a complex process leading to the formation of an ordered solid, plays a central role in many industrial and scientific fields. However, understanding its various stages, particularly nucleation, remains limited. This work presents an innovative strategy combining nuclear magnetic resonance (NMR) and dynamic nuclear polarization (DNP) to monitor the crystallization of glycine, a polymorphic molecule. By halting the process at specific time points through rapid freezing at 100 K, it becomes possible to identify transient phases such as the beta polymorph. Subsequently, the use of a functionalized mesoporous material slows down the process, revealing the early stages of crystallization. This methodology represents a powerful tool for studying polymorphism and could, in the long term, be applied to the identification of transient solid forms in various contexts.

In this study, nuclear magnetic resonance (NMR) is used to investigate the crystallisation behaviour of aspirin within a mesoporous SBA-15 silica material.