Investigating the role of interphases in composite electrolytes by solid- state NMR

Universidad del País Vasco / Euskal Herriko Unibertsitatea Investigating the role of interphases in composite electrolytes by solid-state NMR Pedram Ghorbanzade Thesis Directors: Prof. Senentxu Lanceros-Mendez Dr. Juan Miguel López del Amo 2024 (cc) 2025 Pedram Ghorbanzade (cc by-nc-sa 4.0) In memory of Nima Taherkhani, my kindest friend, with the most joyful laughter. Acknowledgments Throughout this research, I have been fortunate to receive professional and personal support and encouragement from many people. I wish to extend my heartfelt thanks to those whose contributions made this possible. To my supervisors, Dr. Juan Miguel Lopez del Amo and Prof. Senentxu Lanceros-Mendez. Senentxu, thank you for making things easy and offering your help whenever I needed it. Juan Miguel, there are no words to express how pleased and proud I am about working with you. I am very grateful for all the things you taught me, and the freedom you gave me to propose and follow my ideas. You are an amazing supervisor and person, and I am grateful for your continuous support in these years. It wouldn’t have been this simple without you. To my co-authors, especially Michel, Pedro, Roshan, Grazia, Arianna, and Kerman, for collaborating on my project through helpful advice or their assistance with the experiments. It was a pleasure working with you in a team. Also, Pierre Ranque, who helped me a lot in my first few months at CIC, took me to the right track and mindset and gave me the selfconfidence I really needed. It was amazing having you around and I appreciate your help. To my supervisors at the University of Cambridge, Prof. Dame Clare Grey, Dr. Chris O’Keefe, and Dr. Sundeep Vema, for hosting me and offering their support with my project during my stay. I learned a lot from you, both about science and work ethics. Thanks for making my 3month stay an unforgettable experience. Also, for giving me the chance to present my work at the group meeting, which I truly enjoyed. To Angela, Ramon, Mohammed, Richard, Jana, Astrid, Stefan, Farheen, Kieran, Megan, and other group members for the socialization in the canteen, climbing gym, or the pub. To my amazing friends, the members of the Malaka group, who guaranteed my mental health. Especially, to Liam, Remy, Ross, Ade, Arianna, Andrei, Rosa, Dimitris, Lu, Simon, Davide, Pierre, Paul, Steven, Nico, and Elisabetta. Thanks for your company and for letting me be myself. I made wonderful memories with you, from our stay at the rural house to the movie nights and pintxo-potes. You were the best part of my life in the last 3 years and I’m super happy to have you around. To my NMR mate, Nahom, for all those long scientific and non-scientific chats we had, and the ping-pong games we played together. Besides your big heart and humor, your passion for science and dedication are inspiring. I am happy to have you as a friend in the last 5 years. To my great colleagues at CIC, for all the fun time we spent together inside and outside work. The football games, Sagardotegias, Kutxi nights, ping-pong games, Barbacoas, Tortillas, team building activities, etc. Also, for helping me learn Spanish and integrate. Even the lab cleaning is fun with you! To Asier, Esti, Mercedes, and Sara, for their great support with all the paperwork regarding my relocation and immigration. I truly appreciate your assistance which started long before I joined CIC and will last until my final day here. To Elena, who manages to find great solutions for any complicated case and makes everyone’s work at CIC easy. You are amazing. To my old friends, Mahsa and Ghazal, whom I miss a lot. I feel so energized with you and I can’t wait to see you again. Every time spent with you is quality time! Also, to my Couchsurfing friend, Aya for hosting me in Vienna, but especially for our week-long hike in the Pyrenees. Even its memories make me appreciate life and feel happy. I also thank my dear family, who have always supported me and given me the confidence and freedom to make decisions. Being far from you hasn’t been easy, but I’m happy about the great memories we made when we met in Turkey, Austria, and Nepal. Each of these trips was a boost of energy and motivation for me, and I hope all of us can gather sometime soon. Finally, I would like to thank the whole DESTINY team, particularly Prof. Masquelier, Beatrice, and Louise for their great work in organizing this amazing project. I feel very proud to have been part of this amazing program. This project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant Agreement N°945357. Table of Contents Summary ............................................................................................................................... 1 Resumen ............................................................................................................................... 5 1. Introduction ................................................................................................................... 10 1.1 Energy storage scenario .......................................................................................... 10 1.2 Development of Batteries........................................................................................ 11 1.3 Li-ion batteries ....................................................................................................... 13 1.4 Solid State Batteries................................................................................................ 14 1.4.1 Solid Polymer Electrolytes ................................................................................... 16 1.4.2 Solid inorganic electrolytes .................................................................................. 17 1.4.3 Composite electrolytes ....................................................................................... 19 1.5 Solid-State NMR ..................................................................................................... 22 1.5.1 One-Pulse and Hahn Echo ................................................................................... 25 1.5.2 Saturation Recovery ............................................................................................ 26 1.5.3 Exchange Spectroscopy ....................................................................................... 27 1.5.4 Cross Polarization and Heteronuclear Correlation ................................................... 28 1.5.5 CP-spin diffusion ................................................................................................ 30 1.5.6 Isotope-exchange experiments ............................................................................ 30 1.5.7 Variable Temperature solid-state NMR .................................................................. 31 1.5.8 Pulse Field Gradient NMR and Diffusion Experiments ............................................. 31 1.6 Scope and field of research ..................................................................................... 33 2. Methodologies .............................................................................................................. 34 2.1 Composite Electrolyte Preparation ........................................................................... 34 2.2 Characterization Techniques .................................................................................... 35 2.2.1 Solid-state NMR................................................................................................ 35 2.2.2 Electrochemical Characterizations .................................................................... 36 2.2.3 Scanning Electron Microscopy .......................................................................... 37 2.2.4 Powder X-ray Diffraction .................................................................................... 37 2.2.5 Raman Spectroscopy ........................................................................................ 37 2.2.6 Mechanical Testing ........................................................................................... 37 2.2.7 Thermogravimetric Analysis .............................................................................. 38 2.2.8 Differential Scanning Calorimetry ...................................................................... 38 3. General Objective ......................................................................................................... 39 4. Result Summary and Discussion ................................................................................... 41 5. References: .................................................................................................................. 44 6. Concluding Remarks and Future Impacts ...................................................................... 49 7. Annex ........................................................................................................................... 50 7.1 Publication #1: ..................................................................................................... 50 7.2 Publication #2: ..................................................................................................... 62 7.3 Publication #3: ..................................................................................................... 72 7.4 Publication #4: ..................................................................................................... 80 7.5 Publication #5: ..................................................................................................... 97 7.6 Publication #6: ................................................................................................... 106 1 Summary The mitigation of global warming and climate change, as one of the biggest challenges ahead of humans, requires a rapid transition in energy sources from fossil fuels to renewable energies. Electrification of the transport sector is considered a priority to achieve the worldwide goal of limiting the global temperature rise below 2°C. This electrification requires the development of advanced battery technologies for various applications. Commercialized in 1990, Lithium-ion batteries (LIBs) are currently the most mature and advanced battery technology, offering a combination of high energy and power density. These batteries consist of a negative electrode (usually graphite), a positive electrode (usually lithium nickel manganese cobalt oxide (NMC) or lithium iron phosphate (LFP)), and a liquid electrolyte (LiPF6 dissolved in organic solvents). With more than 30 years of research and development, commercial LIBs deliver high gravimetric and volumetric energy densities above 200 Wh/kg and 400 Wh/L. However, the volatile and flammable organic solvents used in the electrolyte of LIBs raise non-negligible safety concerns regarding their application in Electric Vehicles (EVs). In addition, despite their merits, LIBs are not able to meet the performance requirements of a wide range of applications. For instance, the characteristics of batteries used in an EV are significantly different from the ones used for grid applications. Thus, meeting the huge energy demand requires the advancement of alternative battery technologies with different characteristics. Solid-state batteries are an emerging battery technology, aiming to improve both the safety and energy density of LIBs. Solid-state batteries (SSBs) differ from LIBs in that the liquid electrolyte and separator are replaced by a solid membrane, which has superior thermal and mechanical stability and can effectively block the growth of Li dendrites. These properties enable the use of Li metal as the negative electrode, which has a remarkably higher theoretical energy density compared to the commonly used graphite electrodes. These advantages make solid-state batteries an attractive choice for next-generation batteries. There are three main types of solid electrolytes used in SSBs. The first is polymer electrolytes, which offer excellent processability and flexibility, allowing for good interfacial contact with the electrodes. However, their room-temperature ionic conductivity is relatively low, limiting their performance in certain applications. The second type includes inorganic lithiumconducting electrolytes such as oxides, sulfides, and halides. These materials exhibit high ionic 8 particularmente cuando se manipulan polvos de LLZO para CPEs. Además, se detallan las fortalezas y limitaciones de las técnicas de caracterización más utilizadas, como el análisis termogravimétrico (TGA), la difracción de rayos X (XRD), la espectroscopía Raman, la resonancia magnética nuclear (RMN), la espectroscopía de fotoelectrones de rayos X (XPS), y la espectroscopía de impedancia electroquímica (EIS) en el análisis de LLZO. El siguiente trabajo tenía como objetivo diseñar CPEs de altas prestaciones basados en una matriz polimérica plastificada a base de PEO y partículas de Li1.3Al0.3Ti1.7(PO4)3 (LATP). Este estudio investigó las interacciones entre el LATP y la matriz polimérica, y empleó varias técnicas de RMN de estado sólido para dilucidar el mecanismo de transporte de iones tanto en cortas distancias (dinámica local) como a largas distancias. Los resultados mostraron que la participación del LATP en el transporte de iones a larga distancia se observa principalmente a altas temperaturas, lo que se atribuye principalmente al aumento del intercambio de Li+ en la interfase entre las fases orgánicas e inorgánicas. El análisis de CPEs con diferentes contenidos de LATP concluyó que, a pesar de su mínima contribución al transporte iónico, la presencia de LATP en fracciones moderadas (10 vol%) mejora significativamente las propiedades mecánicas de la membrana y su resistencia al crecimiento de dendritas de litio. El siguiente estudio de esta tesis se centra en las interfases formadas en un CPE compuesto por argiroditas Li6PS5Cl y una matriz de PEO-LiTFSI. En este trabajo, se utilizó ampliamente la RMN de estado sólido para identificar las especies presentes en la interfase y demostrar su papel en el transporte iónico. Los resultados mostraron que, al entrar en contacto con el PEO, el Li6PS5Cl se descompone en sus precursores, formando complejos con PEO como P(EO)3-LiCl y PEO-Li3PS4. Estos complejos exhiben una movilidad muy limitada, lo que ralentiza el intercambio de Li a través de la interfase entre las dos fases. Se demostró que la composición química y el grado de desorden de las argiroditas afectan sustancialmente su tasa de descomposición y la extensión de las especies presentes en la interfase. Se concluyó que, para realizar el potencial de los CPEs, es necesario optimizar la composición de ambas fases para mejorar la compatibilidad. La investigación final profundizó en la compatibilidad y las reacciones interfaciales entre los electrolitos de haluro Li2.1Ga0.1Zr0.9Cl6 y los argiroditos Li6PS5Cl. Una combinación de EIS, DRT y RMN en estado sólido reveló que estos electrolitos son químicamente incompatibles y sufren una descomposición parcial al entrar en contacto, formando fases secundarias en la interfase. Debido a su movilidad local más lenta, estas fases secundarias aumentan la 9 resistencia al transporte de iones a través de la interfase. A pesar de la incompatibilidad, las reacciones interfaciales se estabilizan con relativa rapidez, permitiendo el funcionamiento de la celda a largo plazo. No obstante, se observó que estas reacciones pueden progresar aún más con el calentamiento, lo que convierte a la temperatura en un parámetro importante en la manipulación y el funcionamiento de estos electrolitos. Es importante destacar que se observó un intercambio espontáneo de iones de Li entre los electrolitos de haluro y argirodita, participando directamente en este proceso las fases secundarias de la interfase. La presente tesis destaca parámetros críticos que gobiernan el transporte iónico en CPEs, particularmente el papel de las interfases en el intercambio de iones entre las fases orgánicas e inorgánicas. Al descubrir los mecanismos subyacentes, este trabajo proporciona información valiosa para el diseño estratégico de CPEs con mayor conductividad iónica a temperatura ambiente y un rendimiento mejorado, acercando el potencial de la tecnología de CPE a aplicaciones prácticas. 10 1. Introduction 1.1 Energy storage scenario More than a century of burning fossil fuels as well as unsustainable energy and land use have contributed to an increase in the global surface temperature, reaching 1.1°C above preindustrial levels.1 The future negative impacts of climate change are expected to be vast, costing significantly more than preventing it.2 Thus, keeping the global temperature rise below 2°C by 2050 has been set as a worldwide goal to mitigate the severe consequences of climate change.3 As reported by the International Renewable Energy Agency (IRENA)3, about twothirds of greenhouse gas emissions originate from energy production and consumption. Therefore, meeting the below 2°C goal requires a profound transformation in the energy system. Renewable energy and increasing energy efficiency not only can reduce 90% of CO2 emissions, but they also provide other benefits such as making energy access affordable and improving energy security3, an issue that was highlighted during the armed conflict in Ukraine.4 In addition, although this energy transition requires a huge investment from the governments, it would generate millions of additional jobs and make economic sense when taking the costsavings in the long term into account.3 Nevertheless, it must be noted that the energy sector alone cannot provide all the solutions to the major issue of climate change. According to IRENA, electrification is a top short and mid-term priority to set the energy system on the path needed to achieve the below 2°C objective.5 Electrification is a key solution for reducing gas emissions, particularly if paired with renewable energy sources.2 The impact of electrification, renewables, and energy efficiency is shown in Figure 1. A combination of renewable energy and deep electrification can reduce CO2 emissions by 60%. Adding energy efficiency and direct use of renewables, this share reaches a remarkable 90%.2 Figure 1 The share of different sectors in annual energy-related CO2 emissions and the impact of electrification and renewables in reducing the emissions. Reproduced with permission from IRENA.2 11 Three main sectors that require electrification are buildings, industry, and transport. Within the transport sector, electricity provides less than 1% of the total energy consumed for global transportation, including passengers and cargo.5 The electrification of this sector is mainly associated with the use of Battery Electric Vehicles (BEVs), Hydrogen and Fuel Cell Electric Vehicles (FECVs). BEVs have developed significantly and experienced huge market growth in the last decade. For instance, between 2017 and 2023, EV sales have increased from 1 to around 14 million, reaching an 18% share in total new cars sold in 2023.6 In Europe, the second largest EV market, more than one in every 5 cars sold was electric.6 The exponential growth of electric car stock in the last decade is demonstrated in Figure 2. Besides national policies and incentives, one of the main reasons behind this rapid growth is that EVs are becoming more competitive relative to internal combustion engine vehicles, both in performance and price.7 This improved competitivity also stems from the advancements in Li-ion battery (LIB) technology over the last decades, since their commercialization in 1991 by SONY. For instance, the gravimetric and volumetric energy density of LIBs have increased from 80 Wh/kg and 200 Wh/L in 1991 to around 260 Wh/Kg and 700 Wh/L in 20158, while the price of LIBs on the pack level has dropped by a factor of 7.9 Figure 2 Global electric car stock trends between 2010 – 2023. Reprinted with permission from Ref.6 Despite all these improvements and the huge growth of the EV market, the transport sector, with a 25% share of global energy-related CO2 emissions,10 still lags in the energy transition and requires acceleration.2,3 In addition, electrification requires a wide variety of batteries for a broad range of applications. For instance, a battery designed for aviation requires a much higher power than portable devices, with cost being of lower importance.9 Thus, the development of batteries in different technologies or chemistries is vital. 1.2 Development of Batteries A battery is a device that stores energy and typically consists of multiple cells put together. Each cell contains two electrodes with different electrical potentials, separated by an electrolyte. Batteries operate based on reduction-oxidation (redox) reactions that occur at the electrodes. In these reactions, electrons and ions are transferred: electrons flow through an 12 external circuit connecting the electrodes, while ions migrate through the electrolyte. The electrolyte facilitates ion diffusion but blocks electron flow, preventing short circuits. The specific redox reactions at each electrode, along with the directions of ion and electron flow during charging and discharging, are illustrated schematically in Figure 3. Figure 3 Schematic representation of a battery during charge and discharge. Two critical metrics that determine a battery's performance are energy density and power density. Energy density refers to the amount of energy a battery can store per unit volume or weight. A higher energy density allows for longer operation times between charges. This is especially important in applications like electric vehicles and portable electronics, where maximizing the energy stored in a limited space is crucial. Power density, on the other hand, measures how quickly energy can be delivered by the battery for a given volume or weight. A battery with high power density can provide a large amount of energy in a short period, critical for devices that require quick acceleration or immediate response, such as power tools or electric vehicles. The batteries are usually classified into primary and secondary. The primary batteries reach their end-of-life at full discharge state, while secondary batteries can be recharged and reused. Although primary batteries are important in many applications particularly in medical and military fields, addressing the previously discussed global warming issues requires advanced secondary batteries. The rather simple concept of batteries has led to the development of a wide variety of batteries, from nickel-cadmium (Ni-Cd) and lead-acid to lithium-ion batteries, and more recently, sodium-ion batteries. Thanks to the high electrochemical potential of lithium and its low weight, lithium-ion batteries are the most popular, because they offer a combination of high energy density, high power, and long cyclability, making them ideal for a broad range of applications.11 The superiority of Li-based batteries compared to other mature technologies is demonstrated in Figure 4. Negative Electrode Positive Electrode Oxidation Reduction Negative Electrode Positive Electrode Reduction Oxidation 13 Figure 4 The gravimetric and volumetric energy density of different secondary battery technologies. Reprinted with permission from Kubota et al.12 1.3 Li-ion batteries With more than 3 decades of research and development, LIBs are currently the most advanced type of batteries, offering gravimetric and volumetric energy densities above 200 Wh/Kg and 400 Wh/L.13,14 In commercial LIBs, schematically illustrated in Figure 5, graphite is the material of choice for the negative electrode. The charge storage mechanism of graphite anodes is based on the intercalation of Li atoms between the graphite planes, which can continue up to one Li per 6 carbon atoms.15 Upon discharge, deintercalation takes place and the Li ions leave the graphite planes and move toward the positive electrode. Figure 5 illustration of a typical Li-ion battery and its different components. Graphite’s gravimetric capacity of 372 mAh/g is lower than other anode materials such as LTO, Si, Ge, and Li metal. Nevertheless, graphite, is still the preferred option in commercial batteries, as it offers a good balance between cost, lithiation potential, Li diffusivity, electrical conductivity, and low volume change.15,16 On the positive electrode, LiNi1-x-yCoxAlyO2 (NCA) LiNixCoyMnzO2 (NMC), and LiFePO4 (LFP) are the most common active materials. While NCA and NMC offer a higher energy density, LFP is cheaper and shows a better rate-performance and cycle life.17,18 Thus, the choice of the cathode material depends on the application requirements. 14 A key component in LIBs is the separator, a porous membrane typically made of glass fiber or polypropylene. It physically separates the electrodes while enabling ion transport by being soaked in electrolyte. Some of the most important characteristics of a separator include its chemical stability, thickness, porosity, mechanical and thermal properties, tortuosity, electrolyte uptake, and cost.19 While some of these parameters can directly affect the rate performance, cyclability, and safety of the cell,19 the efficacy of the separator is closely tied to the choice and properties of the electrolyte. Although the electrolytes should in theory be selected based on the properties and requirements of the anode and cathode materials, the electrolyte used in most commercial LIBs is Lithium hexafluorophosphate salt (LiPF6), dissolved in organic solvents. This is mainly due to its balanced properties such as its high electrochemical stability limit of 4.5 V and a rather low cost. 20 Instead of changing the major electrolyte components, most research activities targeted electrolyte additives to control the interface and electrolyte properties.20 The additives are generally designed to decompose at a certain voltage potential and form specific decomposition products in the form of a protective layer on the surface of the electrodes.20–22 Depending on the structure and concentration of the additive, the protective layer can affect the chemical composition and the morphology of the solid electrolyte interphase (SEI) and impact the performance and cycle life of the cell.21,22 LIBs currently offer the highest performance and are expected to continue dominating the market in the coming years. Nonetheless, there are still some concerns and issues with LIBs. Firstly, LIBs are approaching their physicochemical limit.13 Although their gravimetric and volumetric energy densities have been increasing at a decent pace, without any unforeseen technological development, they are unlikely to surpass their physicochemical limit estimated at around 400 Wh/kg and 800 Wh/L.13 Secondly, LIBs are insufficient for all types of applications as each has different specific requirements. While LIB is the top contender for EVs, other technologies might be advantageous for maritime, aviation, or grid energy storage.9,23 Finally, LIBs contain combustible liquid organic electrolytes and thus are still associated with safety concerns, particularly in elevated-temperature and high-rate applications.13,24 Multiple electrochemical reactions occurring inside the battery generate heat that may not be efficiently dissipated in the case of mechanical, electrical, or thermal abuse. In this case, the separator is damaged causing a short circuit, or oxygen is released from the cathode side, enabling further (electro)chemical reactions. This leads to thermal runaway which can generate smoke, fire, or explosion.25 Considering these limitations, research and development in alternative battery technologies are essential. These alternative technologies do not necessarily compete with LIBs but rather collaborate to meet the huge energy requirements previously mentioned. These batteries could potentially offer different characteristics such as lower cost (Na-ion or aqueous), long life (Li-O), fast charge (organic, novel architecture), or higher energy density and safety (solid-state). 1.4 Solid State Batteries Solid state batteries (SSBs) have attracted huge attention in recent years, and together with Na-ion batteries are among the most promising technologies as alternatives to Li-ion batteries. 15 SSBs differ from LIBs in that the electrolyte is an ion-conducting solid that also plays the separator role (see Figure 6). Figure 6 illustration of Li-ion batteries and solid-state batteries. Reproduced with permission from Grady et al. 26 Moving from a flammable liquid electrolyte to a solid membrane with higher mechanical and thermal stability in theory improves the safety of the battery cells13, although this is yet to be proved in larger-scale battery packs. In addition, higher mechanical properties of solids in principle hinder the growth of Li dendrites, facilitating the use of Li metal as the negative electrode which offers significantly higher specific capacity and energy density.13 Finally, in inorganic solid electrolytes, huge concentration gradients are avoided because the Li ions are the only mobile species. As a result, higher current densities and faster cell charge/discharge are possible.13 Solid electrolytes are assessed based on a comprehensive set of criteria that determine their suitability for application in a battery cell. Ionic conductivity is one of the most critical factors, as it represents how quickly and efficiently ions can travel through the electrolyte. High ionic conductivity is crucial for achieving high charge and discharge rates, as well as maintaining voltage stability. While ionic conductivity takes the mobility of both cations and anions into account, transport number (t+ or t-) indicates the fraction of charge that is carried by the cations or anions.27 A low cation transport number can lead to the formation of a salt concentration gradient in the electrolyte, which ultimately limits the rate capability of the cell and causes dendrite growth.27,28 The next key parameter is interfacial resistance, which refers to the resistance encountered at the interface between the electrolyte and the electrodes. This resistance, which is perhaps one of the most challenging aspects of solid-state batteries, has a significant impact on the rate-capability of the electrolyte and the overall cell performance. In addition to these electrochemical properties, mechanical properties are also vital, as the electrolyte must withstand stresses and strains during operation without cracking or deforming. Additionally, thermal, chemical, and electrochemical stability are crucial, as degradation at high potentials (against high-voltage cathodes) or very low potentials (against lithium) is undesirable and leads to battery failure. Lastly, electrolytes must be cost-effective and easy to process. While these factors may be less critical in lab-scale research, they are essential for commercialization and large-scale production. 16 Solid electrolytes are commonly divided into three main groups29: i) solid polymer electrolytes ii) solid inorganic electrolytes and iii) composite electrolytes. Each of them is discussed below. 1.4.1 Solid Polymer Electrolytes Solid Polymer Electrolytes (SPEs) consist of a Li salt incorporated in a polymer matrix. The main advantage of SPEs is their excellent processability and flexibility, enabling thin membranes with low electrode/solid electrolyte interface resistance. On the other hand, SPEs have relatively low room temperature ionic conductivities (<10-4 S.cm-1) and low transport numbers (<0.3)30. While the ionic conductivity increases with the temperature, operation above 80°C sacrifices the mechanical properties and electrochemical stability of the SPE31–33, increasing the chance of Li dendrite growth and short circuit. Thus, it is essential to enhance the roomtemperature ionic conductivity of SPEs, without compromising their thermal, electrochemical, and mechanical properties. The most common approaches are adding plasticizers, incorporating ionic liquids, developing new salts, or modifying the polymer (grafted copolymers). The most extensively researched polymer electrolyte is Poly(ethylene oxide) (PEO), which is a semicrystalline polyether. With a high donor number, PEO forms complexes with Li+, and its chain flexibility facilitates intraand inter-chain transport of Li+34, as shown in Figure 7. Figure 7 Mechanism of ion transport in PEO. Reproduced with permission from Xue et al.34 Among many Li salts explored for PEO systems, Lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) is the most popular, thanks to its bulky nature that ensures facile salt dissociation, required for high ionic conductivity. The structure of this Li salt is shown in Figure 8. Increasing the salt content in SPE decreases the crystallinity of PEO but compromises the mechanical properties of PEO and the mobility of EO groups. Considering this trade-off, the EO: Li ratios of 20:1 or 16:1 are the most common. 17 Figure 8 Molecular structure of LiTFSI 1.4.2 Solid inorganic electrolytes Solid inorganic electrolytes have very different properties from polymers. They present excellent thermal and mechanical properties and significantly higher ionic conductivity up to 10-3 S.cm-1. A significant characteristic of inorganic electrolytes is their Li+ transport number of ~1 because Li ions are the only mobile species. This prevents concentration gradients, enables high-rate applications, and improves battery lifetime and safety.35 In a crystalline solid electrolyte, the mobile ions pass between two crystallographic sites for lithium, along the minimum energy pathway. This energy barrier, called migration energy, greatly impacts the ionic conductivity of the electrolyte. Ion transport and conductivity are also influenced by defects, which could be in the form of interstitials, vacancies, partial occupancy, etc.35,36 Significant efforts in designing crystal structures with optimal migration paths and optimal defects have resulted in several groups of highly conductive electrolytes, with oxides, sulfides, and halides being the most promising ones.37,38 1.4.2.1 Garnets The high ionic conductivity (0.1 – 1 mS/cm) and excellent stability against Li metal, make Li7La3Zr2O12 (LLZO) a great candidate for solid-state Li-metal batteries.39 In the LLZO crystal structure, there are 3 interstitial sites for Li-ions namely i) tetrahedral (24d), ii) octahedral (48g), and iii) distorted octahedral (96h). The 24d tetrahedra and 48g/96h octahedra share a face, which significantly facilitates Li-ion migration, achieving high ionic conductivities.40 LLZO has two main polymorphs namely cubic and tetragonal. In the tetragonal phase, the tetrahedral Li sites are fully filled. As a result, it is two orders of magnitude less conductive than the cubic phase.40 Thus, LLZO is commonly doped with elements such as Ta, Ga, Nb, and Al to stabilize the highly conductive cubic phase at room temperature and enhance its ionic conductivity by generating vacancies and disorder.41,42 The wide electrochemical stability window of LLZO, which is the voltage range it can withstand without undergoing redox reactions,36 makes it compatible with Li metal and high-voltage cathodes such as NMC (LiNi0.8Co0.1Mn0.1O2) when the kinetic limitations are considered.40,43 This makes LLZO an ideal material for a fundamental understanding of the chemical and electrochemical processes in solid-state batteries. From a practical standpoint, it is necessary to sinter LLZO at high temperatures to reduce the presence of grain boundaries. This is crucial because grain boundaries can obstruct the 24 • Chemical shift anisotropy (CSA): The magnetic field experienced by each nucleus depending on their orientation versus the B0. These interactions called CSA contribute to the relaxation process and broaden the observed NMR signal.70 • Dipolar coupling: The magnetic moment associated with each nucleus can generate a weak magnetic field that can be sensed by the nearby nuclei. These interactions are reversible, meaning that the first spin can also experience the magnetic field created by the second spin. These interactions which are very sensitive to internuclear distances, are known as dipolar coupling and induce relaxation.70 • Quadrupolar coupling: In nuclei with a spin number of I > ½, the electron charge distribution around the nucleus is asymmetric and non-spherical. This creates an electric field gradient (EFG) which interacts with the nucleus, causing relaxation and line broadening. • Paramagnetic relaxation: This mechanism occurs only in species having unpaired electrons. Unpaired electrons with large magnetic moments generate a strong local magnetic field which can interact with the nearby nuclei, causing rapid relaxation.70 Solid materials contain numerous crystallites, each oriented at different angles relative to the magnetic field. Since the anisotropic NMR interactions, such as chemical shift anisotropy (CSA), dipolar coupling, and quadrupolar coupling, depend on these orientations, each crystallite experiences a slightly different magnetic field. As a result, the NMR spectrum of a powdered sample tends to be broad and asymmetric, forming a distinctive pattern known as powder pattern, which represents the distribution of crystallite orientations relative to the magnetic field. Unlike in liquids, where rapid molecular tumbling averages out these anisotropies and results in sharp signals, the absence of such motion in solids leads to broad and complex spectra. This issue can be addressed by introducing artificial motions through a technique called Magic Angle Spinning (MAS). By spinning the sample at a very fast rate at an angle of 54.7 degrees relative to the B0, MAS effectively averages out many of these anisotropic interactions, leading to narrower signals and higher resolution. For MAS to be effective, the spinning rate must exceed the linewidths of the anisotropic interactions, which can be in the range of several kHz. Although anisotropic interactions were once viewed as a significant challenge in NMR, they contain valuable structural and dynamic information. Thus, in some cases, it may be necessary to restore these interactions using techniques like Rotational Echo Double Resonance (REDOR) which reintroduces the dipolar interactions to study parameters like internuclear distances. While most elements have an NMR-active isotope and can theoretically be measured, recording an NMR spectrum with reasonable quality using a standard magnet generally requires the nuclei to have a combination of high natural abundance, high gyromagnetic ratio, and low quadrupole moment. This restricts the range of nuclei that can be easily studied, excluding nuclei like 17O, and 35Cl, due to the complexity of sample preparation, measurement, or data interpretation. However, several elements involved in battery systems are well-suited for NMR analysis and could be measured with relative ease. These include: • Proton (1H): Benefiting from the highest gyromagnetic ratio among all nuclei and an isotope natural abundance close to 100%, proton (1H) is a highly sensitive nucleus with 25 strong NMR signals, ideal for studying hydrogen-containing samples. For instance, the surface reactivity and protonation during storage or processing steps can easily be analyzed using 1H NMR. However, due to the dipolar couplings, the 1H spectra suffer from signal broadening in solid-state NMR, making MAS often crucial. • Fluorine (19F): Thanks to its high gyromagnetic ratio and 100% abundance, 19F NMR measurements are fairly simple. The wide chemical shift range of 19F allows for the resolution of distinct fluorine environments, making it especially useful in the analysis of complex electrolytes and the identification of different fluoride species. Given its presence in nearly all organic Li-salts, 19F is an important nucleus for studying electrolyte materials. In the battery field, 19F NMR is used to investigate salt agglomeration or decomposition processes. In addition, 19F diffusion experiments are frequently employed in liquid, gel, and solid polymer electrolytes to probe the longrange dynamics of the anions. • Phosphorus (31P): With 100% abundance and a decent gyromagnetic ratio, 31P is an important nucleus in NMR, especially since it is present in a wide range of organic and inorganic materials. The chemical shift and shielding effects in 31P NMR provide detailed information about the local environment, making it a powerful tool for studying the coordination chemistry of phosphorus in electrolyte formulations. In solid-state battery research, 31P NMR is commonly employed to investigate sulfidebased electrolytes (LPSCl), NASICON-type electrolytes (LATP), and Oxynitrides (LiPON). • Lithium (6Li and 7Li): As the main element in batteries, Lithium has two NMR-active nuclei, both widely used in NMR studies. 7Li, the more abundant isotope (92.5%), has a higher gyromagnetic ratio, resulting in greater sensitivity and intense signals. On the other hand, 6Li, while less abundant, has a lower quadrupolar moment, leading to sharper signals and higher-resolution spectra. Therefore, 6Li is often preferred for indepth investigation of local chemical environments, although at the cost of requiring higher sample concentrations or longer experiment times. Interestingly, the lower abundance of 6Li could be advantageous in particular cases such as EXSY experiments, where the undesired spin diffusion is minimized. Overall, both 6Li and 7Li NMR are invaluable for not only structural determination but also for studying the ion-transport mechanisms and Li dynamics in electrolytes. As an important feature, metallic Li exhibits a knight shift, a phenomenon that significantly shifts its NMR signal from that of diamagnetic lithium. This shift, which occurs due to interactions between the nucleus and the unpaired electrons in metals,75 enables the study of Li plating and Li dendrite growth in the batteries. Herein, some of the most useful and common techniques used in studying solid composite electrolytes are explained. 1.5.1 One-Pulse and Hahn Echo One-pulse is a simple but widely used pulse sequence in NMR spectroscopy. In this pulse sequence (shown in Figure 12.a) a 90° pulse flips the magnetization from the z-axis to the xy plane, and then the FID is recorded. In some cases, rapid magnetization dephasing due to T2 relaxation causes information loss and signal decay. Additionally, one-pulse suffers from dead 26 time issue, which refers to a short period (typically a few microseconds) immediately after the applied pulse during which the receiver cannot detect the signal due to electronic limitations. Thus, the initial portions of the FID and its corresponding information are lost, and the spectrum is slightly distorted.76 To address these challenges, the Hahn Echo pulse sequence is proposed, which is mainly applied to nuclei with spin number I = ½.77 In this pulse sequence (shown in Figure 12.b) after the 90° pulse and a delay time (τ), a 180° pulse is applied to reverse the magnetization dephasing in the xy plane. After another delay time (τ), the spins are rephased, and the FID is recorded.78 The word “echo” in this pulse sequence refers to the initial dephasing and subsequent rephasing of the signal. This pulse sequence allows for recovering the information lost both during the dephasing and the dead time. Hahn Echo is typically limited to nuclei with spin number I = ½, as in quadrupolar nuclei with I > ½ the refocusing of the spins into an echo signal is challenging. Conducting echo experiments for these nuclei requires other pulse sequences such as quadrupole echo and Multiple Quantum Magic Angle Spinning (MQMAS) which are out of the scope of this work.76,77 Figure 12 Pulse sequence of a) One-Pulse and b) Hahn-Echo NMR experiments. 1.5.2 Saturation Recovery In this experiment which aims at measuring the T1 relaxation time, magnetization is initially forced to zero by a train of pulses. The nullified magnetization is allowed to recover during a time gap t1, and then a 90° pulse is applied to record the signal. As shown in Figures 13.a and 13.b, Repeating this experiment at different time intervals results in a build-up curve from which the time constant can be calculated.79 Since T1 relaxation times can vary between different species due to differences in structure and local dynamics, analyzing these relaxation times provides valuable insights into the composition, molecular structure, and dynamic behavior of the sample. b 90 180 n o 90 a n - l 27 Figure 13 a) Pulse sequence and b) build-up curve of Saturation Recovery experiment. Reproduced from Forse.80 1.5.3 Exchange Spectroscopy In many systems, ions or molecules undergo spontaneous and continuous exchange between two distinct environments or sites. Understanding the dynamics and characteristics of such exchanges could be crucial, for instance for determining the ion transport mechanisms in solid composite electrolytes. This type of exchange process can be effectively investigated using exchange spectroscopy (EXSY) experiments. Like other two-dimensional NMR techniques, this experiment comprises several sequential steps: preparation, evolution, mixing, and detection,70 as Figure 14 illustrates. Figure 14 a) Schematic presentation, b) pulse sequence, and c) data representation of an EXSY NMR experiment. Reproduced from Keeler.70 The pulse sequence in an EXSY experiment involves three 90° pulses. While these pulses affect all spins uniformly, it is conceptually simpler to focus on an individual spin. The first pulse flips the magnetization to the XY plane, where it evolves during the t1. The second pulse then returns this magnetization to the z-axis, effectively frequency labeling the spin by encoding its environment. During the subsequent mixing time, a chemical exchange between the spin at two different sites may occur. In this case, the magnetization initially generated on spin-1 is transferred to spin-2. The final pulse then rotates this magnetization back to the XY plane, allowing for the acquisition of the FID, which achieves the spectrum in the F2 (or direct) dimension after the Fourier transformation. By varying the t1 across multiple experiments and applying a second Fourier transformation, spectra are obtained in a new dimension, called F1 or indirect dimension. As shown in Figure 14.c, the resulting EXSY NMR data are typically 90 t1 90 n a b T1 ms Intensity Preparation Mixing Evolution t1 Detection t2 90 90 90 a b t1tmix F2 F1 Exchanged Not Exchanged c 28 displayed as contour plots, where the X and Y axes correspond to the chemical shifts in the F2 and F1 dimensions, respectively. Signals that appear along the diagonal line represent nuclei that did not undergo exchange during the mixing time and thus remained in their original environment. Conversely, nuclei that participated in exchange processes produce off-diagonal peaks, also known as cross-peaks. Conducting EXSY experiments at varying mixing times allows for the study of the rate at which exchange processes occur. However, this analysis requires high-quality data and careful analysis, as too long mixing times can lead to the decay of signal intensity due to the relaxation of magnetization. In fact, one of the limitations of EXSY experiments is the requirement that the mixing time must be shorter than the relaxation time of the spins.81 Therefore, selecting an appropriate mixing time is critical and typically restricted to a few hundred milliseconds in most solid composite electrolytes. It is important to note that the magnetization transfer during the mixing time which gives rise to the cross-peaks may not only result from chemical exchange but also from spin diffusion, a phenomenon caused by homonuclear dipolar interactions between two abundant spins. In this process, as opposed to exchange, magnetization is transferred between two spatially fixed spins.82 This process, which depends both on the gyromagnetic ratio of the nucleus and the internuclear distances, can be studied using a nuclear Overhauser effect spectroscopy (NOESY) experiment, which is identical to the EXSY but follows a different purpose.70 Distinguishing between these two mechanisms as the sources of cross-peaks in EXSY experiments for Li-based batteries can be realized by conducting EXSY experiments on 6Li instead of 7Li. Thanks to its lower gyromagnetic ratio and lower natural abundance (thus higher internuclear distance), 6Li reduces the likelihood of spin diffusion, allowing cross-peaks to be attributed more confidently to chemical exchange. 1.5.4 Cross Polarization and Heteronuclear Correlation Cross Polarization (CP) is a widely used NMR technique that can significantly improve the sensitivity of the NMR spectra of nuclei with low gyromagnetic ratio or low natural abundance. In the CP experiment, magnetization is transferred from an abundant nucleus, such as 1H or 19F, to a less abundant nucleus, like 6Li or 13C. This transfer is based on through-space dipolar interactions, which are inversely proportional to the cube of the internuclear distance, according to equation 2.80 equation (2): 𝜔12 ∝ 𝛾1𝛾2 𝑟3 As shown in Figure 15.a, the CP pulse sequence typically begins with the excitation of the abundant nucleus, followed by a contact period during which the magnetization is transferred to the less abundant nucleus. The Fourier Transform of the recorded FID results in a single spectrum along the chemical shift axis of the less-abundant nucleus. In this spectrum, each signal corresponds to a site of the non-abundant nucleus that is close to and thus effectively coupled with the abundant nucleus. If a particular site in the low-abundance nucleus does not 29 participate in the magnetization transfer, due to distance for example, it will not appear in the spectrum. CP is highly beneficial for improving the sensitivity of the NMR spectra of nuclei with low gyromagnetic ratio or low abundance. The maximum sensitivity improvement for a single scan is 𝛾𝐼𝛾𝑠 ⁄. However, since the abundant nucleus typically has a shorter relaxation time, CP allows for a higher number of scans to be recorded within the same time frame compared to direct excitation, resulting in further enhancement of the signal-to-noise ratio.80 This is crucial for analyzing nuclei with low abundance like 13C in systems where direct observation is challenging due to low signal intensity. Building on the principles of CP, 2D Heteronuclear Correlation (HETCOR) is a slightly more advanced version of CP designed to investigate the interactions between different types of nuclei within a sample. As shown in Figure 15.b, the HETCOR pulse sequence differs from the one of CP in that it includes an additional evolution step before the cross-polarization. Once the magnetization of the abundant nucleus is excited, it undergoes a period of evolution to encode information about its local environment. This evolution time referred to as t1, is systematically varied throughout the experiment to introduce an extra dimension to the data and construct a two-dimensional spectrum in which correlation signals indicate through-space magnetization transfer, hence spatial proximity.83 Therefore, HETCOR is invaluable for characterizing complex materials like solid electrolytes, where understanding the local structure and interactions between components is crucial. Figure 15 Graphical presentation of the pulse program of a) CP-MAS and b) HETCOR with their corresponding data representation. It must be noted that although CP and HTECOR are beneficial for enhancing signal sensitivity and providing structural information, these methods are not inherently quantitative. This is because the efficiency of magnetization transfer depends on the size of the dipolar a b F2 CP CP Decoupling optional t2 90 I S F2 F1 Correlation signal No correlation CP CP Decoupling optional t1 t2 90 I S 30 interactions and, therefore on several factors such as the proximity of nuclei, and the relaxation dynamics of the system. These factors can vary significantly across different sites within a material, causing complications in the quantitative analysis. As a result, CP and HETCOR cannot be reliably used to determine the absolute quantities or relative concentrations of different sites within the sample. For applications requiring precise quantitative analysis, alternative NMR techniques or careful calibrations may be necessary. Additionally, although the correlations in HETCOR and CP depend on the internuclear distances, for precise measurements of the distance between certain nuclei, other NMR pulse sequences such as Rotational-Echo Double-Resonance (REDOR) or Transfer-Echo DoubleResonance NMR (TEDOR) are employed.84 1.5.5 CP-spin diffusion In some EXSY experiments, if the intensity of the signal of interest is too low, or if selective EXSY of certain signals is desired, a combination of CP and EXSY experiments may be used, which herein is referred to as “CP-spin diffusion”. In this experiment, a CP step is introduced at the beginning of the EXSY sequence to selectively enhance the signal intensity of the nuclei that are spatially close to the abundant nucleus. While this selective enhancement limits the scope of the EXSY experiment to regions that are close to the abundant nuclei, -which could sometimes be beneficial-, it significantly boosts the intensity of the signals corresponding to the correlated sites. As a result, CP-spin diffusion allows for sensitive exploration of exchange processes in systems where the signals might otherwise be too weak to detect. This approach is very useful in studies of solid electrolytes, in which enhancing the intensity of certain signals at the interphase is critical for obtaining meaningful insights into the dynamics and interactions. By omitting the evolution step, CP-spin diffusion experiments can be performed as onedimensional (1D) experiments. This modification results in a substantial reduction in the overall experiment time, making it more efficient and less resource-intensive. However, this simplification comes at the cost of losing the additional information provided by the second dimension in a typical 2D NMR experiment. Consequently, 1D versions of these experiments are best suited for systems where the molecular environment and dynamics are already well resolved. In addition, these 1D measurements are also very beneficial when one of the signals disappears after the CP step. In such cases, comparing the CP-spin diffusion spectrum with CP and direct excitation spectra can yield valuable information on the exchange processes within the sample. 1.5.6 Isotope-exchange experiments These experiments are specific to systems containing elements with multiple NMR-active isotopes, such as lithium (7Li and 6Li) and hydrogen (1H and 2H). The main idea is to substitute one isotope for another and observe the resulting changes in the NMR spectra. This opens opportunities for innovative experiments tailored to the system in question. Among several possible combinations, A common approach involves introducing a small quantity of 6Li into a system predominantly composed of 7Li.85–87 Then, by tracking the movement and interaction of the introduced isotope over time or during cell cycling, critical insights into transport mechanisms, diffusion rates, and the interactions between different environments within the 31 sample are obtained. This technique is particularly valuable for investigating ion transport in complex, multi-component systems like solid-state batteries.87 However, these experiments are generally more costly and require special care in sample preparation, measurement, and data analysis. 1.5.7 Variable Temperature solid-state NMR Variable temperature NMR experiments used in this thesis rely on the principle that the linewidth of an NMR signal is inversely proportional to the T2 relaxation time. Thus, measuring the signal linewidth over a broad temperature range can indirectly reveal the impact of thermally activated motions on T2 relaxation.71 At very low temperatures, also known as rigid lattice regime, the ion motions are too slow to average out the dipolar interactions, resulting in broadened NMR signals.88 Upon heating, the ion or molecular motions are increased, continuously averaging the dipolar interactions and decreasing the signal linewidth, as demonstrated in Figure 16. This process is called motional narrowing, from which the Li+ jump rates can be roughly estimated.88 This can continue until an extreme narrowing regime is achieved, in which further heating has no impact on the linewidth.71,88 When plotting the linewidth vs temperature, both plateaus at high and low temperatures should ideally be observed. However, the onset temperatures highly depend on the Li dynamics of the sample. For instance, in a highly Li-conducting sample like Li6.6P0.4Ge0.6S5I, probing the ion-hopping processes requires NMR measurements at temperatures as low as 9 K, thus needing a special setup.73 The variable temperature measurements are normally conducted in static mode, i.e. without MAS. This not only simplifies the data interpretation but also allows for a significantly broader temperature range. Figure 16 Schematic representation of a variable temperature NMR experiment and data representation for studying molecular or ion dynamics. 1.5.8 Pulse Field Gradient NMR and Diffusion Experiments The diffusion coefficient of various nuclei can be determined through Pulse Field Gradient (PFG) NMR experiments, in which as opposed to usual NMR experiments, the magnetic field is varied in strength along a specific axis within the sample. In these experiments, the presence of a field gradient causes the magnetization precession of nuclei to differ based on their position within the sample, leading to dephasing.71 As demonstrated in Figure 17, the method F M z 1 T -1 Motional Narrowing Temperature 7 i chemical shift ppm 32 involves applying two field gradient pulses to dephase and then rephase the magnetization. In the case of diffusion during the delay time between these pulses (Δ), the magnetization would not completely refocus, resulting in signal decay. By conducting these experiments with different gradient strengths and analyzing the signal intensity, the diffusion coefficient is determined. Figure 17 Pulse sequence of Pulse Field Gradient Stimulated Echo experiment. and the signal decay due to diffusion. 1.5.9 In situ and operando NMR experiments Recording NMR spectra of battery materials during cell operation, known as operando measurements, enables real-time investigation of ion dynamics, interactions, and phase evolution throughout the cycling process. This technique, which is schematically presented in Figure 18, offers valuable insights into the chemical and structural changes of the cell components. In the case of solid-state batteries, it is especially useful in studying the Li plating/stripping and dendrite growth. In situ NMR is a great method to observe and quantify the dead Li, a concept that is particularly important in anode-free batteries.89,90 Nevertheless, despite all the advantages, this technique comes with significant limitations. First, due to practical challenges, operando measurements cannot be combined with magic angle spinning, restricting the analysis to a static mode with lower resolution. Additionally, current technology does not allow for straightforward measurements in standard cell configurations such as coin cells or Swageloks. Instead, specialized cell setups are required which differ considerably from conventional ones. For example, it is often difficult to apply and maintain consistent stack pressure in these cells, causing significant challenges in studying solid-state batteries. GF 90 rf gradient l ld d nt t m l t d o 180 GF GGradient eld strength Intensity 33 Figure 18 Schematic illustration of in situ NMR setup and its application in studying Li plating, dendrite formation, and dead lithium quantification. Reproduced with permission from Xiang et al.90 1.6 Scope and field of research A review of the literature highlights that despite the promising potential of the CPEs in solidstate batteries, additional efforts are needed to address the existing challenges. The main areas that require improvement include their room-temperature ionic conductivity, interface resistance with the Li metal, and compatibility of the phases. Although CPEs have been explored in several studies and the concept is not entirely new, there remain knowledge gaps about several fundamental aspects of their properties. Most importantly, the mechanism of ion transport through the electrolyte, particularly across the interphases, has not been thoroughly investigated. In addition, the impact of processing conditions such as humidity or temperature on the performance of the CPEs remains unclear or poorly understood. Finally, little effort has been devoted to investigating the interactions between the inorganic fillers and the polymer matrix. This is while these interactions are expected to play a critical role in mitigating the Li dendrite growth and thus significantly enhance the rate performance and cyclability of the cells. Addressing these gaps in knowledge is essential for driving innovation, overcoming existing challenges, and realizing the full potential of CPE technology in advancing solid-state batteries. ot nt o t t NMR 40 in publication #2, titled: Influence of the LLZO-PEO interface on the microand macro-scale properties of composite polymer electrolytes for solid-state batteries. These studies revealed that LLZO's instability in ambient conditions introduces significant complexities, impacting both its dynamics and surface properties. It was realized that these complexities are generally not well understood and that the methods commonly employed to study the surface layers are often inefficient. Thus, to address this knowledge gap, these issues were further explored in publication #3, titled: A vision for LLZO Carbonate Formation: Perspectives on Surface Treatment Approaches and Characterization Techniques. Following the study of LLZO, the ion transport properties and phase interactions were examined in a new CPE composed of LATP particles and a PEO-based polymer matrix. These CPEs demonstrate a better performance, attributed to the lower fraction of inorganic particles and the inclusion of plasticizers. Exploring how LATP particles interact with various components of the polymer matrix, and how these interactions influence the ion transport formed the basis of publication #4, titled: Transport Properties and Local Ions Dynamics in LATP-Based Hybrid Solid Electrolytes. In the next step, a CPE composed of Li₆PS₅Cl and PEO-LiTFSI was developed to examine their compatibility and potential interfacial reactions. Solid-state NMR was employed to analyze the reaction products and their influence on ion transport across the interphase. Furthermore, the study explored how dopant-induced disorder in the argyrodite crystal structure affects the kinetics of these interfacial reactions. This study was discussed in publication #5, titled: Unveiling the Reactivity and the Li-Ion Exchange at the PEO-Li6PS5Cl Interphase: Insights from Solid-State NMR. Lastly, an all-inorganic composite electrolyte based on an argyrodite (Li₆PS₅Cl) and a halide (Li2.1Ga0.1Zr0.9Cl6) electrolyte was prepared to gain insight into the interfacial reactions of these solid electrolytes. In this study, solid-state NMR, EIS, and DRT analysis were used to investigate their chemical compatibility and to determine the role of the interphase species in the local and long-range ion transport. The findings of this study are presented in publication #6, titled: Insights into the Compatibility and Interphases of LPSCl Argyrodites and Zr-based Halide Electrolytes for Solid-State Batteries. 41 4. Result Summary and Discussion Publication #1: The first publication of this thesis focuses on the reaction of LLZO in air and the formation of secondary phases on its surface. The study reveals that pristine LLZO, even when stored in a glovebox with an inert atmosphere, contains noticeable amounts of secondary phases, particularly Li2CO3 and LiOH layers on its surface, and hydrogens inside its crystal structure. Considering the well-established negative impact of these phases on the performance of LLZO, a heat treatment process was proposed for their elimination. Based on the TGA profile, the LLZO powders were heated at 350°C, 550°C, and 750°C to monitor the LLZO chemical and phase evolution during this heat treatment, using XRD, solid-state NMR, and Raman spectroscopy. The results demonstrate that heating at 750°C is highly effective in removing both the LiOH and Li2CO3 surface layers and the protons inside the structure. Additionally, variable temperature NMR experiments revealed that this heat treatment significantly enhances the local Li dynamics of LLZO. Lastly, it was shown that although the heat treatment is not particularly effective as a preparatory step before sintering the ceramic pellets, it notably improves the morphology and integrity of the CPEs. Publication #2: In the next step, the impact of these interphase species was explored by comparing two ceramic-rich CPEs composed of pristine and heat-treated LLZO. The findings indicated that removing Li2CO3 and LiOH enhances the interactions between PEO and LLZO, resulting in a more uniform phase distribution and improved mechanical properties. Solid-state NMR experiments showed that removing these surface layers reduces the interfacial Li exchange between LLZO and PEO. Further investigation identified LiOH as a key facilitator in this exchange, aiding Li+ transport across the interphase. However, given that the interfacial Li exchange occurs relatively slowly (in the order of hundreds of milliseconds), LLZO's contribution to ion conduction remains insignificant. This conclusion was further supported by the observation that the CPE containing heat-treated LLZO exhibited an ionic conductivity 3 times higher than its pristine counterpart. Additionally, the heat treatment enhanced Li plating kinetics and reduced the solid electrolyte/electrode interface resistance by almost one order of magnitude. These improvements were also reflected in the reduced overpotential and prolonged cyclability of the CPEs in symmetric cells, ultimately demonstrating the superior performance of these CPEs in blocking Li dendrite growth through the electrolyte. The prolonged cyclability of the CPE with heat-treated LLZO, likely due to the slower rate of lithium dendrite growth, encouraged a new project to investigate the Li dendrite growth through solid electrolytes using operando solid-state NMR. While operando NMR has been successfully employed in Li-ion cells with liquid electrolytes, its application to solid electrolytes remains limited. The main challenges of using this technique in solid-state batteries are: 1. Stack pressure: Efficient Li plating/stripping tests in solid-state batteries require consistent pressure to fill the generated voids and maintain a stable overpotential. 42 2. Ionic conductivity: The room temperature ionic conductivity of CPEs and polymer electrolytes is relatively low, making plating/stripping at room temperature feasible only at very low currents. The NMR operando cells, purchased from eProbe, do not contain a built-in spring for maintaining stack pressure. On the other hand, adding an iron-based spring is also not possible, as iron is ferromagnetic and unsuitable for being used inside the NMR magnet. Thus, several layers of Viton rubber were included to mitigate the pressure issue. The second challenge regarding ionic conductivity was addressed by performing the electrochemical and NMR experiments at high temperatures. However, it was realized that the cells are not perfectly airtight at elevated temperatures, causing the passivation of the Li metal. To overcome this issue, pure argon as an inert gas was used to heat the sample inside the NMR probe. This approach allows for cell operation for about 20 hours for a 50-liter argon cylinder. However, this period is still insufficient to observe the growth of Li dendrites. Additionally, using argon flow for several days or weeks is not economically feasible. Thus, focusing more on practicality, ongoing efforts aim at improving the room temperature ionic conductivity and increasing the stack pressure by non-magnetic compression springs to enable the cell operation at room temperature. This work is in progress and the solutions are anticipated for the coming year. Publication #3: After identifying the negative impact of surface layers in LLZO in CPEs, a brief concept publication was developed to discuss and compare several methods of removing these layers. In addition, this work presents the advantages and limitations of common characterization techniques such as XRD, NMR, Raman, etc. in analyzing LLZO samples. Focusing on practical aspects of LLZO handling, this guide aims to help researchers better understand the complications of these systems and conduct their experiments more accurately. Publication #4: The next work delved into ion transport in a CPE composed of PEO-based polymer matrix and LATP particles. The short-range (local) and long-range conductivity were separately analyzed in hybrid electrolytes containing up to 20 vol% LATP to assess the participation of LATP in ion transport. Saturation recovery and variable temperature 19F NMR experiments demonstrated that the addition of LATP has minimal impact on the local ions’ dynamics. The findings from 6Li EXSY NMR and 6Li/7Li isotope exchange experiments indicated that an interfacial Li+ exchange does occur between the phases at a local scale. However, conductivity measurements revealed that this exchange does not translate into improving long-range conductivities, due to the high interface resistance for Li+ transport between the LATP and the polymer phases. Nevertheless, by increasing the temperature from 20°C to 60°C, the time constant of this Li+ exchange decreases by a factor of 6, facilitating the participation of LATP in long-range transport, observed by electrochemical conductivity measurements. Combining PFG-NMR and electrochemical tests, it was revealed that at low or moderate fractions (10 vol%), LATP particles have little to no impact on the ion transport properties of the CPEs. 43 However, at this concentration, they notably improve the mechanical properties of the electrolyte membranes, which resulted in both improved processability and dendrite growth resistance. Publication #5: In the subsequent publication, the interphases between Li6PS5Cl (LPSCl) and PEO-LiTFSI were investigated, as understanding the nature of these interphases and their role in ion transport is crucial for both multilayer systems and CPEs. Detailed solid-state NMR analysis of this system revealed that when highly disordered LPSCl is brought into contact with highly polar PEO, its structure degrades to form P(EO)3-LiCl and PEO-Li3PS4 complexes, as well as polysulfides. 6Li EXSY and 1H-6Li Spin-diffusion NMR experiments demonstrated a direct Li exchange between the PEO and LPSCl, which is not mediated by the species at the interphase, particularly PEO-LiCl and PEO-Li3PS4. Further insights into this system were obtained by replacing the LPSCl with a more disordered Li5.4PS4.6ClBr0.4 (LPSClBr), which reacted more intensely with PEO and formed a higher content of decomposition products at the interphase. The NMR analysis indicated that the mentioned interphase species exhibit very slow mobility, hindering the Li exchange between the phases. This study concluded that the defect chemistry of argyrodites greatly impacts their stability against highly polar polymers like PEO, influencing the ion transport across the interphase. The observed Li exchange and its dependence on the decomposition products suggest the significant potential of these CPEs if the chemistry of the polymer and the argyrodite are slightly modified to improve their compatibility. Publication #6: The last publication investigates the chemical compatibility between Li6PS5Cl argyrodite and a Li2.1Ga0.1Zr0.9Cl6 (LGZC) halide electrolyte, as Li6PS5Cl is widely used as an interlayer between the Li metal and halide electrolytes to mitigate their instability. Solid-state NMR experiments revealed that the chemical incompatibility between these electrolytes drives interfacial reactions, involving partial decomposition of both halide and sulfide. The decomposition products at the interphase exhibit slower dynamics than the main phases, thus increasing the resistance of ion transport through the interphase. EIS, DRT, and solid-state NMR analysis indicated that interphase reaches stability after a short period, although the decomposition advances at elevated temperatures. Notably, 6Li and 7Li EXSY NMR experiments revealed the spontaneous Li-ion exchange not only between the primary halide and argyrodite phases but also between the decomposition products at the interphase. These observations explain why incorporating a thin layer of Li6PS5Cl enhances cell performance in multilayer systems. Although this work focused on LGZC/LPSCl interactions, its simple methodology used for examining the chemical compatibility and interfacial reactions can be readily applied to other halide/sulfide composites with different chemical compositions. 44 5. References: 1. Intergovernmental Panel on Climate Change. IPCC, 2023: Climate Change 2023: Synthesis Report. 2023 . 2. International Renewable Energy Agency. Global Energy Transformation: A Roadmap to 2050 (2019 Edition). Abu Dhabi . 3. International Renewable Energy Agency. Global Energy Transformation: A Roadmap to 2050. Abu Dhabi . 4. In focus: EU energy security and gas supplies - European Commission. https: energy.ec.europa.eu news focus-eu-energy-security-and-gas-supplies-2024-02-15_en. 5. International Renewable Energy Agency. Smart Electrification with Renewables: Driving the Transformation of Energy Services. Abu Dhabi . 6. International Energy Agency. Global EV Outlook 2024. 2024 . 7. International Energy Agency. Global EV Outlook 2023: Catching up with Climate Ambitions. OECD, 2023 . 8. Placke, T., loepsch, R., Dühnen, S. & inter, M. ithium ion, lithium metal, and alternative rechargeable battery technologies: the odyssey for high energy density. J Solid State Electrochem , 1939–1964 2017 . 9. Trahey, . et al. Energy storage emerging: A perspective from the Joint Center for Energy Storage Research. Proceedings of the National Academy of Sciences 7, 12550–12557 2020 . 10. International Renewable Energy Agency. World Energy Transitions Outlook 2023: 1.5°C Pathway. 2023 . 11. Gao, Y., Pan, Z., Sun, J., iu, Z. & ang, J. igh-Energy Batteries: Beyond ithium-Ion and Their ong Road to Commercialisation. Nano-Micro Lett. 4, 94 2022 . 12. ubota, ., Dahbi, M., osaka, T., umakura, S. & omaba, S. Towards -Ion and Na-Ion Batteries as “Beyond i-Ion”. The Chemical Record 8, 459–479 2018 . 13. Janek, J. & Zeier, . G. A solid future for battery development. Nat Energy , 1–4 2016 . 14. Cao, ., Zhang, J. & i, . Batteries with high theoretical energy densities. Energy Storage Materials 6, 46–55 2020 . 15. Nitta, N., u, F., ee, J. T. & Yushin, G. i-ion battery materials: present and future. Materials Today 8, 252–264 2015 . 16. Asenbauer, J. et al. The success story of graphite as a lithium-ion anode material – fundamentals, remaining challenges, and recent developments including silicon oxide composites. Sustainable Energy & Fuels 4, 5387–5416 2020 . 17. Manthiram, A. A reflection on lithium-ion battery cathode chemistry. Nat Commun , 1550 2020 . 18. Murdock, B. E., Toghill, . E. & Tapia-Ruiz, N. A Perspective on the Sustainability of Cathode Materials used in ithium-Ion Batteries. Advanced Energy Materials , 2102028 2021 . 19. ingappan, N., ee, ., Passerini, S. & Pecht, M. A comprehensive review of separator membranes in lithium-ion batteries. Renewable and Sustainable Energy Reviews 87, 113726 2023 . 20. Xu, . Electrolytes and Interphases in i-Ion Batteries and Beyond. Chem. Rev. 4, 11503–11618 2014 . 21. Michan, A. . et al. Fluoroethylene Carbonate and Vinylene Carbonate Reduction: Understanding ithium-Ion Battery Electrolyte Additives and Solid Electrolyte Interphase Formation. Chem. Mater. 8, 8149–8159 2016 . 45 22. aregewoin, A. M., otango, A. S. & wang, B.-J. Electrolyte additives for lithium ion battery electrodes: progress and perspectives. Energy Environ. Sci. 9, 1955–1988 2016 . 23. Albertus, P., Manser, J. S. & itzelman, S. ong-Duration Electricity Storage Applications, Economics, and Technologies. Joule 4, 21–32 2020 . 24. Armand, M. & Tarascon, J.-M. Building better batteries. Nature 45 , 652–657 2008 . 25. Chen, Y. et al. A review of lithium-ion battery safety concerns: The issues, strategies, and testing standards. Journal of Energy Chemistry 59, 83–99 2021 . 26. Grady, Z. A., ilkinson, C. J., Randall, C. A. & Mauro, J. C. Emerging Role of Non-crystalline Electrolytes in Solid-State Battery Research. Front. Energy Res. 8, 2020 . 27. Ghorbanzade, P., oaiza, . C. & Johansson, P. Plasticized and salt-doped single-ion conducting polymer electrolytes for lithium batteries. RSC Adv. , 18164–18167 2022 . 28. Xiao, J. ow lithium dendrites form in liquid batteries. Science 366, 426–427 2019 . 29. i, S. et al. Progress and Perspective of Ceramic Polymer Composite Solid Electrolytes for ithium Batteries. Advanced Science 7, 1903088 2020 . 30. Villaluenga, I. et al. Negative Stefan-Maxwell Diffusion Coefficients and Complete Electrochemical Transport Characterization of omopolymer and Block Copolymer Electrolytes. J. Electrochem. Soc. 65, A2766 2018 . 31. Yue, . et al. All solid-state polymer electrolytes for high-performance lithium ion batteries. Energy Storage Materials 5, 139–164 2016 . 32. Bouchet, R. et al. Single-ion BAB triblock copolymers as highly efficient electrolytes for lithium-metal batteries. Nature Mater , 452–457 2013 . 33. Dias, F. B., Plomp, . & Veldhuis, J. B. J. Trends in polymer electrolytes for secondary lithium batteries. Journal of Power Sources 88, 169–191 2000 . 34. Xue, Z., e, D. & Xie, X. Poly ethylene oxide -based electrolytes for lithium-ion batteries. J. Mater. Chem. A 3, 19218–19253 2015 . 35. Bachman, J. C. et al. Inorganic Solid-State Electrolytes for ithium Batteries: Mechanisms and Properties Governing Ion Conduction. Chem. Rev. 6, 140–162 2016 . 36. Famprikis, T., Canepa, P., Dawson, J. A., Islam, M. S. & Masquelier, C. Fundamentals of inorganic solid-state electrolytes for batteries. Nat. Mater. 8, 1278–1291 2019 . 37. Grill, J., Steensen, S. ., Castro, D. . Q., Castelli, I. E. & Popovic-Neuber, J. Solid-state inorganic electrolytes for next generation potassium batteries. Commun Mater 5, 1–10 2024 . 38. im, . J., Balaish, M., adaguchi, M., ong, . & Rupp, J. . M. Solid-State i–Metal Batteries: Challenges and orizons of Oxide and Sul de Solid Electrolytes and Their Interfaces. Advanced Energy Materials , 2002689 2021 . 39. Murugan, R., Thangadurai, V. & eppner, . Fast ithium Ion Conduction in Garnet-Type i7 a3Zr2O12. Angewandte Chemie International Edition 46, 7778–7781 2007 . 40. ang, C. et al. Garnet-Type Solid-State Electrolytes: Materials, Interfaces, and Batteries. Chem. Rev. 0, 4257–4300 2020 . 41. Charrad, G. et al. Investigation of Chemical and Thermal Stability of i7−x a3Zr2−xTaxO12 Garnet Type Solid-State Electrolyte to Assemble Self-Standing i-based All Solid-State Battery. Energy Technology , 2300234 2023 . 42. Thangadurai, V., aack, . & eppner, . J. F. Novel Fast ithium Ion Conduction in Garnet-Type i5 a3M2O12 M = Nb, Ta . Journal of the American Ceramic Society 86, 437–440 2003 . 43. Zhang, N. et al. Mechanism Study on the Interfacial Stability of a ithium Garnet-Type Oxide Electrolyte against Cathode Materials. ACS Appl. Energy Mater. , 5968–5976 2018 . 44. Zheng, C. et al. Grain boundary modi cation in garnet electrolyte to suppress lithium dendrite growth. Chemical Engineering Journal 4 , 128508 2021 . 46 45. Accardo, G. et al. Fast and low-temperature densi cation of highly conductive i7 a3Zr2O12 ceramic electrolytes for solid-state batteries. Journal of Power Sources 585, 233632 2023 . 46. Cheng, E. J., Shara , A. & Sakamoto, J. Intergranular i metal propagation through polycrystalline i6.25Al0.25 a3Zr2O12 ceramic electrolyte. Electrochimica Acta 3, 85–91 2017 . 47. Ren, Y., Shen, Y., in, Y. & Nan, C.- . Direct observation of lithium dendrites inside garnettype lithium-ion solid electrolyte. Electrochemistry Communications 57, 27–30 2015 . 48. Cussen, E. J. The structure of lithium garnets: cation disorder and clustering in a new family of fast i conductors. Chem. Commun. 412–413 2006 doi:10.1039 B514640B. 49. Paolella, A. et al. NASICON lithium ions conductors: Materials, composites, and batteries. Current Opinion in Electrochemistry 36, 101108 2022 . 50. Ma, F. et al. Preparation and evaluation of high lithium ion conductivity i1.3Al0.3Ti1.7 PO4 3 solid electrolyte obtained using a new solution method. Solid State Ionics 95, 7–12 2016 . 51. Yin, J.- . et al. Recent Advances of ATP and Their NASICON Structure as a Solid-State Electrolyte for ithium-Ion Batteries. Advanced Engineering Materials 5, 2300566 2023 . 52. e, X., Zhu, Y. & Mo, Y. Origin of fast ion diffusion in super-ionic conductors. Nat Commun 8, 15893 2017 . 53. uo, C., Yi, M., Cao, Z., ui, . & ang, Y. Review of Ionic Conductivity Properties of NASICON Type Inorganic Solid Electrolyte ATP. ACS Appl. Electron. Mater. 6, 641–657 2024 . 54. Tan, D. . S. et al. Elucidating Reversible Electrochemical Redox of i6PS5Cl Solid Electrolyte. ACS Energy Lett. 4, 2418–2427 2019 . 55. iu, Y. et al. Revealing the Impact of Cl Substitution on the Crystallization Behavior and Interfacial Stability of Superionic ithium Argyrodites. Advanced Functional Materials 3 , 2207978 2022 . 56. anghofer, I. et al. Substitutional disorder: structure and ion dynamics of the argyrodites i6PS5Cl, i6PS5Br and i6PS5I. Phys. Chem. Chem. Phys. , 8489–8507 2019 . 57. anghofer, I., Gadermaier, B. & ilkening, . M. R. Fast Rotational Dynamics in Argyrodite-Type i6PS5X X: Cl, Br, I as Seen by 31P Nuclear Magnetic Relaxation—On Cation– Anion Coupled Transport in Thiophosphates. Chem. Mater. 3 , 4591–4597 2019 . 58. Golov, A. & Carrasco, J. Molecularevel Insight into the Interfacial Reactivity and Ionic Conductivity of a i-Argyrodite i6PS5Cl Solid Electrolyte at Bare and Coated i-Metal Anodes. ACS Appl Mater Interfaces 3, 43734–43745 2021 . 59. Simon, F. J., anauer, M., enss, A., Richter, F. . & Janek, J. Properties of the Interphase Formed between Argyrodite-Type i6PS5Cl and Polymer-Based PEO10: iTFSI. ACS Appl. Mater. Interfaces , 42186–42196 2019 . 60. Muruganantham, R., u, .- ., o, Y. & iu, .-R. Improving electrochemical stability by modifying the i6PS5Cl entailed with the mixed phase of i6.4 a3Zr1.4Ta0.6O12 composite solid electrolytes for all-solid-state lithium battery applications. Surface and Coatings Technology 479, 130480 2024 . 61. i, Y. et al. Interface stability of iCl-rich argyrodite i6PS5Cl with propylene carbonate boosts high-performance lithium batteries. Electrochimica Acta 363, 137128 2020 . 62. Tuo, ., Sun, C. & iu, S. Recent Progress in and Perspectives on Emerging alide Superionic Conductors for All-Solid-State Batteries. Electrochem. Energy Rev. 6, 17 2023 . 63. Tao, B., Zhong, D., i, ., ang, G. & Chang, . alide solid-state electrolytes for all-solidstate batteries: structural design, synthesis, environmental stability, interface optimization and challenges. Chemical Science 4, 8693–8722 2023 . 47 64. Riegger, . M., Schlem, R., Sann, J., Zeier, . G. & Janek, J. ithium-Metal Anode Instability of the Superionic alide Solid Electrolytes and the Implications for Solid-State Batteries. Angewandte Chemie International Edition 60, 6718–6723 2021 . 65. iu, X., i, X., i, . & u, . B. Recent Progress of ybrid Solid-State Electrolytes for ithium Batteries. Chemistry – A European Journal 4, 18293–18306 2018 . 66. Arya, A. & Sharma, A. . Polymer electrolytes for lithium ion batteries: a critical study. Ionics 3, 497–540 2017 . 67. uhnert, E. et al. owering the Interfacial Resistance in i6.4 a3Zr1.4Ta0.6O12|Poly Ethylene Oxide Composite Electrolytes. Cell Reports Physical Science , 100214 2020 . 68. Zagórski, J. et al. Garnet–Polymer Composite Electrolytes: New Insights on ocal i-Ion Dynamics and Electrodeposition Stability with i Metal Anodes. ACS Appl. Energy Mater. , 1734– 1746 2019 . 69. Ghorbanzade, P., ópez-Aranguren, P. & ópez del Amo, J. M. A Vision for ZO Carbonate Formation: Perspectives on Surface Treatment Approaches and Characterization Techniques. ChemElectroChem , e202400136 2024 . 70. eeler, J. Understanding NMR Spectroscopy. John iley and Sons, Chichester, U. , 2010 . 71. Vinod Chandran, C. & eitjans, P. Solid-State NMR Studies of ithium Ion Dynamics Across Materials Classes. in Annual Reports on NMR Spectroscopy vol. 89 1–102 Elsevier, 2016 . 72. Nardelli, F. et al. Anisotropy and NMR spectroscopy. Rend. Fis. Acc. Lincei 3 , 999–1010 2020 . 73. ogrefe, ., Stainer, F., Minafra, N., Zeier, . G. & ilkening, . M. R. NMR Down to Cryogenic Temperatures: Accessing the Rateimiting Step of i Transport in Argyrodite Electrolytes. Chem. Mater. 36, 6527–6534 2024 . 74. Yan, Z. & Zhang, R. Measurement of spin-lattice relaxation times in multiphase polymer systems. Journal of Magnetic Resonance 357, 107597 2023 . 75. Pecher, O., Carretero-González, J., Griffith, . J. & Grey, C. P. Materials’ Methods: NMR in Battery Research. Chem. Mater. 9, 213–242 2017 . 76. ashi, . et al. Experimental Comparison of Solid-state NMR Spectra for Quadrupolar Nuclei Using Various Spin-echo Sequences. Chemistry Letters 49, 68–70 2020 . 77. Ulrich, A. S. & Grage, S. . 2 NMR. in Studies in Physical and Theoretical Chemistry eds. Ando, I. & Asakura, T. vol. 84 190–211 Elsevier, 1998 . 78. Carr, . Y. & Purcell, E. M. Effects of Diffusion on Free Precession in Nuclear Magnetic Resonance Experiments. Phys. Rev. 94, 630–638 1954 . 79. Valori, A., McDonald, P. J. & Scrivener, . . The morphology of C–S– : essons from 1 nuclear magnetic resonance relaxometry. Cement and Concrete Research 49, 65–81 2013 . 80. Forse, A. Nuclear Magnetic Resonance Studies of Ion Adsorption in Supercapacitor Electrodes. 2015 doi:10.17863 CAM.26359. 81. leckner, I. R. & Foster, M. P. An introduction to NMR-based approaches for measuring protein dynamics. Biochimica et Biophysica Acta (BBA) - Proteins and Proteomics 8 4, 942–968 2011 . 82. Saalwächter, . & Reichert, D. Polymer Applications of NMR. in Encyclopedia of Spectroscopy and Spectrometry (Second Edition) ed. indon, J. C. 2221–2236 Academic Press, Oxford, 2010 . doi:10.1016 B978-0-12-374413-5.00032-4. 83. nicker, ., Velasco-Molina, M. & nicker, M. 2D Solid-State ETCOR 1 -13C NMR Experiments with Variable Cross Polarization Times as a Tool for a Better Understanding of the Chemistry of Cellulose-Based Pyrochars—A Tutorial. Applied Sciences , 8569 2021 . 48 84. Fyfe, C. A., ewis, A. R. & Chézeau, J.-M. A comparison of NMR distance determinations in the solid state by cross polarization, REDOR, and TEDOR techniques. Canadian Journal of Chemistry 2011 doi:10.1139 v99-199. 85. Gunnarsdóttir, A. B., Vema, S., Menkin, S., Marbella, . E. & Grey, C. P. Investigating the effect of a fluoroethylene carbonate additive on lithium deposition and the solid electrolyte interphase in lithium metal batteries using in situ NMR spectroscopy. J. Mater. Chem. A 8, 14975– 14992 2020 . 86. Berthault, M. et al. ithium isotope tracing in silicon-based electrodes using solid-state MAS NMR: a powerful comprehensive tool for the characterization of lithium batteries. Phys. Chem. Chem. Phys. 5, 22145–22154 2023 . 87. Zheng, J., Tang, M. & u, Y.-Y. ithium Ion Pathway within i7 a3Zr2O12-Polyethylene Oxide Composite Electrolytes. Angewandte Chemie International Edition 55, 12538–12542 2016 . 88. Epp, V., Gün, Ö., Deiseroth, .-J. & ilkening, M. ong-range i dynamics in the lithium argyrodite i 7 PSe 6 as probed by rotating-frame spin–lattice relaxation NMR. Physical Chemistry Chemical Physics 5, 7123–7132 2013 . 89. Gunnarsdóttir, A. B., Amanchukwu, C. V., Menkin, S. & Grey, C. P. Noninvasive In Situ NMR Study of “Dead ithium” Formation and ithium Corrosion in Full-Cell ithium Metal Batteries. J. Am. Chem. Soc. 4 , 20814–20827 2020 . 90. Xiang, Y. et al. Quantitatively analyzing the failure processes of rechargeable i metal batteries. Science Advances 7, eabj3423 2021 . 91. ang, G., iu, ., iang, Y., ang, C. & Fan, .-Z. Composite polymer electrolyte with three-dimensional ion transport channels constructed by NaCl template for solid-state lithium metal batteries. Energy Storage Materials 45, 1212–1219 2022 . 92. iu, . et al. Enhancing ionic conductivity in composite polymer electrolytes with wellaligned ceramic nanowires. Nat Energy , 1–7 2017 . 93. Massiot, D. et al. Modelling oneand two-dimensional solid-state NMR spectra. Magnetic Resonance in Chemistry 40, 70–76 2002 . 94. uhn, A., ilkening, M. & eitjans, P. Mechanically induced decrease of the i conductivity in an alumosilicate glass. Solid State Ionics 80, 302–307 2009 . 49 6. Concluding Remarks and Future Impacts In this thesis, solid-state NMR has been successfully employed as a powerful tool to address key challenges in designing composite polymer electrolytes (CPEs) and investigating interfacial reactions in solid-state battery materials. This work contributes to a deeper understanding of the ion transport processes at the interphases, which are critical to the development of advanced solid electrolytes. Throughout the work, solid-state NMR played a crucial role in providing deep insights into the interphases within these systems, particularly the ion transport mechanisms and phase interactions. In the first part of the research, NMR was used as a tool for the rational design of CPEs, investigating both the structure and Li dynamics of different phases, as well as the Li-ion exchange processes between the organic and inorganic components. The interphase was shown to be a critical zone of ion transport, impacting not only the local ion exchange, but also the microstructure, long-rage ion transport, and mechanical properties of the CPEs. Using solid-state NMR, valuable insights into the dynamics at the interfaces were obtained, demonstrating the role of interphase species such as passivation layers or decomposition products in facilitating or impeding the Li-ion conduction in CPEs. In the second part, solid-state NMR was employed to investigate the compatibility and interfacial reactions between various solid electrolytes, using CPEs as model systems. This part focused on understanding how different solid electrolytes, such as halides, polymer electrolytes, and argyrodites interact when brought into contact. The work showed that interfacial reactions, while sometimes inevitable, can be managed and mitigated through careful material selection. Importantly, it was shown that the chemical compatibility between different electrolytes is highly influenced by the structural disorder and the chemical composition of each phase, making it challenging to propose general conclusions on the compatibility between different families of materials. However, the methods developed in this study can be applied to various electrolyte systems to assess their compatibility and identify potential issues in their design. Although this thesis primarily focused on the fundamental understanding of composite electrolytes, it was demonstrated that high-performance solid electrolytes with improved conductivity and cycling behavior can be developed by modifying the interphases between the organic and inorganic phases. Thus, the methods used, particularly solid-state NMR, are not only effective for describing the interfacial reactions but also valuable for addressing the performance limitations of these systems. The findings and methodologies applied in this research establish a foundation for further exploration of interphases in solid-state batteries. By demonstrating how solid-state NMR can effectively determine ion exchange, monitor chemical evolution, and analyze interphase reactions, this study paves the way for refining electrolyte designs. These insights will guide future efforts in optimizing material compatibility and improving the performance of solid electrolytes, driving advancements in reliable and efficient solid-state battery technologies. phases. Fig. 3 shows the identied phases for each sample and the complete removal of Li 2 CO 3 aer heat treating at 750 °C. While T-Pristine shows only a low temperature cubic phase, in all the heat treated samples, the tetragonal phase is present, coexisting with the low temperature cubic phase in T-350, and with the highly conductive cubic phase in T-550 and T-750. Solid state NMR experiments were conducted on the pristine and heat treated samples to further determine the impact of temperature on the composition and to characterize local Li + dynamics. The 1 H NMR spectrum of T-Pristine is characterized by three main signals (Fig. 4a) appearing at −1.6 ppm, 0.9 ppm, and 3.9 ppm. The signal at −1.6 ppm can be easily ascribed to LiOH in agreement with values previously observed for this compound formed as a secondary product of garnet protonation. 33–37 The signal at 3.9 ppm was also previously observed upon LLZO protonation and is ascribed to protons inside the garnet structure. 33–35 The 1 H– 7 Li HETCOR correlation experiment performed for this sample (Fig. 4c) clearly shows the strong correlation of 1 H signals at −1.6 and 3.9 with 7 Li, in agreement with protons in LiOH and protons coexisting with lithium in Li 7−x H x La 3 Zr 2 O 12 . The signal at 0.9 ppm, which was previously observed but not specically assigned, 33,35 is relatively sharp in the 1 H spectrum, but does not show a clear correlation with Li in the CP-based spectrum of Fig. 4b. These proton signals could be attributed to the protons of the water molecules adsorbed at the surface of LLZO or included at its surface defects, forming LLZO$H 2 O through reaction (4). 12 Li 7−x H x La 3 Zr 2 O 12 +xH 2 O/Li 7−x H x La 3 Zr 2 O 12 $nH 2 O(4) To further explore the nature and disposition of the protonated phases ascribed to the observed proton signals, 1 H– 1 H EXSY homonuclear correlations were determined. This experiment was performed for T-Pristine with a mixing time of 128 ms. The resulting spectrum shown in Fig. 4c is characterized by a diagonal with autocorrelation signals observed in the 1D spectrum and off-diagonal correlations between signals at 3.9 and 0.9 ppm as well as between signals at −1.6 and 0.9 ppm. Since signals at 3.9 and −1.6 ppm are assigned to protons inside the garnet structure and LiOH at the surface of the particles respectively, the protons observed at 0.9 ppm should be placed in close vicinity of both phases. This is in agreement with water molecules adsorbed at the surface of LLZO. The narrower character of this signal at 0.9 ppm is also in agreement with the faster local mobility expected for a water molecule. Additional 1 H– 1 H EXSY NMR experiments performed with very short (0.1 ms) and long (256 ms) mixing times presented in Fig. S1,† indicate that off-diagonal correlations are caused by chemical exchange. The 1 H and 7 Li NMR spectra as well as 7 Li T 1 times of the pristine garnet, T-350, T-550, and T-750 (Fig. 5a–c respectively) were recorded and compared in order to observe the effect of the heat treatment on the Li local mobilities and states of protonation. From the inspection of the 1 H NMR signal Fig. 4 (a) 1 H NMR spectrum, (b) 7 Li– 1 H 2D heteronuclear correlation, (c) 1 H– 1 H 2D correlation with a mixing time of 128 ms and (d) 7 Li NMR spectrum of T-Pristine. Fig. 5 (a) 1 H NMR spectra and (b) 7 Li NMR spectra and (c) 7 Li saturation recovery experiment of T-Pristine, T-350, T-550 and T-750. 11678 |J. Mater. Chem. A,2023,11, 11675–11683 This journal is © The Royal Society of Chemistry 2023 Journal of Materials Chemistry A Paper Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online evolution upon heat treatment (Fig. 5a), a signicant reduction of intensity of the proton signal is observed already for the sample T-350. The reduction of proton intensity observed involves the three signals present in the spectrum of T-Pristine, suggesting the removal of water molecules from LLZO$H 2 O, protons from inside the structure, and LiOH from the garnet surface through reactions (5)–(7) respectively. Li 7−x H x La 3 Zr 2 O 12 $nH 2 O/Li 7−x H x La 3 Zr 2 O 12 +xH 2 O(5) Li7xHxLa3Zr2O12/Li7xLa3Zr2O 12x 2 þx 2H2O(6) Li 7−x H x La 3 Zr 2 O 12 +xLiOH /Li 7 La 3 Zr 2 O 12 +xH 2 O(7) Comparing the 1 H– 1 H and 7 Li– 1 H correlations in T-Pristine and T-350 (Fig. 6a and b), we can observe that LiOH is completely removed, and only a small proton population remains. As a result, the signal intensities are low, and no magnetization transfer can be observed between protons inside the garnet. As can be observed in Fig. 5b, the 7 Li signal in both T-Pristine and T-350 is broad, although it is clear that at least for T-350, the signal has more than one component, probably due to the presence of more than one Li compound. This is in agreement with the clear carbonate band in the Raman spectra (Fig. 3), as well as 7 Li T 1 measurements (Fig. 5c and Table S1†), in which the build-up curves of both T-Pristine and T-350 can be tted only by a double exponential term, indicating the presence of two Li reservoirs with considerably different relaxation times. In general, the presence of fast Li + dynamics results in shorter relaxation times in highly conductive ceramics as compared to the long relaxation times typically observed for non-conductive salts like Li 2 CO 3 . By elimination of the protons inside the garnet structure, the Li T 1 relaxation time decreases from 17 s to 3.9 s (Table S1†). When the heat treatment temperature was increased to 550 ° C and 750 °C, the 1 H signals are further reduced, indicating additional proton elimination through reaction (6). No signals were observed for the heteronuclear 1 H– 7 Li and homonuclear 1 H– 1 H NMR correlations conducted on both samples T-550 and T-750. This result suggests complete elimination of protons in the sample and that the remaining proton signals observed in Fig. 5a are due to background proton signals from the probe or the rotor and therefore external to the sample. In general, the linewidth (FWHM) of 7 Li NMR signals is directly related to ion dynamics, as faster ion motions can effectively average out the anisotropic interactions present in the solid state, 38,39 reducing the observed linewidths. Fig. 5b shows a signicant reduction of the linewidth of 7 Li NMR resonances from 691 Hz to 79 Hz aer heat treatment at 550 °C, implying faster local dynamics for Li + in the T-550 sample. This change can be explained by a phase transition from tetragonal to highly conductive cubic, as well as by partial removal of the rigid lithium carbonates (reaction (8)) as shown previously by Raman spectroscopy in Fig. 3. Li7xLa3Zr2O 12x 2 þx 2Li2CO3/Li7La3Zr2O12 þx 2CO2(8) By further increasing the heat treatment temperature to 750 °C, a further narrowing of the 7 Li NMR signal is observed from 79 Hz to 64 Hz. In addition, the build-up curve of the saturation recovery experiment (Fig. 5c and Table S1†) can be tted by a single exponential term only aer heat treating at 750 °C, which indicates complete removal of Li 2 CO 3 and is consistent with our results from Raman measurements (Fig. 3). It is noteworthy that according to reaction (8) and based on our ICP results (Table S2†), the Li from the carbonate is not lost and rather returns to the garnet structure aer the Li 2 CO 3 decomposition. Furthermore, reaction (8) implies the recovery of the oxygen vacancies that could be generated at the LLZO surface by the proton elimination through water evaporation described in reaction (6). To perform a detailed characterization of the Li dynamics in the pristine and T-750 heat treated samples, variable temperature 7 Li Linewidth (LW) evolutions were measured and compared with those of sintered pellets. The transition from broad gaussian NMR signals observed at the rigid lattice (low temperatures) to narrow Lorentzian peaks at thermally activated states is clearly observed in all samples except in TPristine where the high temperatures required for fast Li + dynamics could not be experimentally reached (Fig. 7a). The motional narrowing (MN) curves (Fig. 7a) show that unlike TPristine, T-750 has a behaviour very similar to the sintered pellets. The resulting temperature-dependance of the 7 Li LW were tted considering a Boltzmann function (eqn (9)) to obtain the inection point and linewidth of the rigid lattice for each sample. y¼A1A2 1þeðxx0Þ=dxþA2(9) Jump rates (at T ip ) and the activation energy of motional narrowing (E MN a ) of each sample were calculated through eqn (10) (ref. 40 ) and (11) (ref. 41 and 42 ) respectively and are presented in Table 1. Fig. 6 (a) 1 H– 7 Li heteronuclear experiment and (b) 1 H– 1 H homonuclear experiment for T-Pristine (black) and T-350 (red). This journal is © The Royal Society of Chemistry 2023 J. Mater. Chem. A,2023,11, 11675–11683 | 11679 Paper Journal of Materials Chemistry A Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online s −1 MN =2p$u RL (10) uðTÞ¼uRL1þuRL B1expEMN a KBT1 þD(11) where u RL is the 7 Li linewidth in the rigid lattice, K B is the Boltzmann constant and Dis a line-broadening constant to take the temperature-independent effects such as magnetic eld inhomogeneity into account. Eqn (11) which assumes a similar temperature dependency in the fraction of vacancies and the fraction of thermally activated ions is uninuenced by the possible distribution of relaxation times in the system and therefore can be applied in a wide range of temperature. 41 Since the high temperature part of the curve could not be achieved for T-Pristine, its activation energy is roughly estimated by using eqn (12) known as the Waugh–Fedin 43 equation. E WF a =1.617 ×10 −3 $T onset (12) where T onset is the temperature in which the motional narrowing begins, and it is calculated by tting the tangential lines in the MN curve. The E MN a from solid state NMR takes both nonlocalized (longrange ion motions) and localized (dipolar relaxations) conduction into account and therefore is lower than the E a obtained by EIS. In fact, a thorough calculation of E a by solid state NMR would require additional T 1 ,T 1r and T 2 relaxation time measurements, as well as impedance spectra recorded in a wide temperature and frequency range, which is out of the scope of this study. 44–48 However, the E MN a values of T-750, Sint-Pristine and Sint-750 are noticeably lower than that of T-Pristine and are similar to what Buschmann et al. 49 reported for highly conductive cubic LLZO with Al doping. At low temperatures, the 7 Li NMR spectra of T-750 seem to be composed of 2 overlapping signals, showing different dynamics. Therefore, the spectra were deconvoluted considering Gaussian and Lorentzian functions. The area fraction (A f ) of the narrow component calculated in the range of −50 to −10 °C (Fig. 7b) indicates the fraction of Li ions with fast jump rates (10 4 s −1 ). Fig. 7c shows that the A f linearly increases upon heating, and already at −10 °C, about 40% of the Li ions are subject to fast dynamics on the NMR timescale. This further conrms our previous conclusion from 7 Li T 1 relaxation time measurements that the heat treatment is not only surfacerelated, but rather affects the dynamics of the Li atoms in the bulk of the garnet. The effect of LiOH and Li 2 CO 3 removal on the pellet's relative density and its ionic conductivity was investigated by preparing sintered pellets using T-Pristine and T-750 as starting materials. These pellets, referred to as Sint-Pristine and Sint-750 respectively, showed no differences in relative densities (92% in both cases), and very similar activation energies, presented in Fig. 7a and Table 1. The bulk ionic conductivities were calculated using eqn (13), where Ais the surface area and ris the sample thickness. s=r/RA (13) Furthermore, the activation energy from ionic conductivities was obtained through the following Arrhenius equation. Fig. 7 (a) Temperature dependency of the 7 Li linewidth for T-Pristine, T-750, Sint-Pristine, and Sint-750. Dashed lines indicate the fit with the Boltzmann equation. (b) 7 Li NMR spectra of T-750 at −50 °C to −10 °C, and deconvolution of the spectrum at −30 °C. (c) Temperature dependency of the area fractions of the narrow components obtained from the deconvolution of spectra with Gaussian and Lorentzian functions. Table 1 Temperature of the inflection point, activation energy of motional narrowing and the jump rates calculated from the variable temperature 7 Li LW analysis shown in Fig. 7a Sample T ip (°C) E MN a (eV) s MN −1 (kHz) T-Pristine +116 0.49 a 51.2 T-750 −47 0.30 58.3 Sint-Pristine −49 0.28 56.9 Sint-750 −49 0.28 57.5 a Calculated from eqn (11). 11680 |J. Mater. Chem. A,2023,11, 11675–11683 This journal is © The Royal Society of Chemistry 2023 Journal of Materials Chemistry A Paper Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online s¼s0eEa KBT(14) Both Sint-Pristine and Sint-750 show identical activation energies (0.43 eV) and comparable bulk ionic conductivities at 25 and 70 °C, as presented in Table 2 and Fig. S2.† We can therefore conclude that heat treating the powder prior to pelletizing and sintering shows no considerable improvement in electrochemical performance. This result demonstrates that the high temperatures used during the sintering processing directly promotes proton, hydroxide, and carbonate removal. However, based on the signicant increase of local Li + mobility observed from the pristine to heat treated powder, shown in Fig. 7a, heat treating LLZO at 750 °C is expected to have a strong inuence in cases in which a sintering step is not performed, as for composite electrolytes or coldsintered pellets in which ion transport is highly sensitive to the interfacial properties of grains. To assess the possible impact of the heat treatment on the Li ion transport at the bulk and the outermost surface of LLZO grains, T-Pristine and T-750 garnets were added to the PEO-LiTFSI mixture to prepare garnet-rich solid composite electrolytes (90 wt% LLZO). The membranes were cast in PTFE dishes and carefully removed aer drying in a vacuum oven. Although the sample preparation was the same (as detailed in the Experimental section), the membranes show clearly different mechanical properties as shown in Fig. 8a and c. While the composite made with TPristine is highly fragile and non-self-standing, the one containing T-750 has high integrity and is self-standing and exible. The SEM images (Fig. 8b and d) demonstrate that such better mechanical properties could be related to the garnet–polymer interphase. In the composite made of T-750, the garnet particles are uniformly coated with the polymer matrix, whereas in the composite with T-Pristine, many garnet particles are not covered by PEO, which could be attributed to a weak garnet– polymer interaction as a result of the presence of surface group impurities, causing a phase segregation and thus leading to low mechanical properties. Despite the non-self-standing nature of the composite with T-Pristine, the impedance spectra of these membranes were recorded (Fig. S3†) and the ionic conductivities obtained were 4.3 ×10 −6 and 1.6 ×10 −5 Scm −1 at 70 °C for composites with T-Pristine and T-750 respectively. The higher ionic conductivity of the latter can be attributed to its more homogenous morphology and better PEO–LLZO interface. The low ionic conductivity of garnet-rich composites is not unprecedented and has been previously reported and explained. 50,51 Detailed characterization of the garnet-rich composites and determining their Li-ion pathways are out of the scope of this article and therefore will be the subject of our future work. It is well known that different dopants and concentrations have an important effect on the Li conductivities of LLZO, 24,30 although it has also been reported that the protonation and formation of LiOH and carbonates are present in garnets regardless of the dopant used. 14,17 The impact of heating on the chemical and thermodynamic processes reported here for a Nb sample are therefore expected to be dopant-independent as mainly surface effects are involved. However, to conrm this hypothesis, additional NMR experiments were performed on commercial Al-doped LLZO. Fig. S4†clearly shows the LiOH formed on the Al-LLZO surface, the adsorbed water molecules and LLZO protonation through a H + /Li exchange reaction. The comparison of the 1 H NMR spectra before and aer the heat treatment at 750 °C (Fig. S5†) indicates the total removal of LiOH and water molecules at the surface, and a signicant reduction of protons inside the LLZO structure. The considerable reduction of the 7Li T 1 relaxation times upon heat treatment shown in Fig. S6†is attributed to the LiOH and Li 2 CO 3 decomposition as well as to the increased Li dynamics upon garnet deprotonation. This conrms that heat treatment at 750 °C is sufficient for complete removal of protons and secondary phases on the Al-LLZO surface. Moreover, like for Nb-LLZO, the heat treatment has a strong impact on the 7 Li NMR linewidths, decreasing from 607 to 305 Hz upon heating at 750 °C in agreement with faster local Li dynamics (Fig. S7†). Concluding remarks In this work, the impact of heat treatment on the LLZO structure, the secondary products present on its surface, and the Li dynamics inside the garnet was investigated. The characterization of the pristine garnet shows a signicant degree of LLZO protonation as well as the presence of water, LiOH and carbonates at the surface. Heat treatment at 350 °C was found Table 2 Bulk ionic conductivity and activation energy of Sint-Pristine and Sint-750 at 25 and 70 °C obtained by using eqn (13) and (14) Sample s 25 °C (S cm −1 )s 70 °C (S cm −1 )E a (eV) Sint-Pristine 3.7 ×10 −4 3.5 ×10 −3 0.43 Sint-750 5.1 ×10 −4 4.2 ×10 −3 0.43 Fig. 8 Photo and SEM micrographs of garnet-rich composite polymer electrolytes containing 90 wt% (a and b) T-Pristine and (c and d) T-750. This journal is © The Royal Society of Chemistry 2023 J. Mater. Chem. A,2023,11,11675–11683 | 11681 Paper Journal of Materials Chemistry A Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online effective to completely eliminate the LiOH at the surface and reduce the number of protons inside the LLZO structure and water at the surface. However, it doesn't suffice to remove carbonates and cause the phase transition to a highly Li conductive cubic phase. Total elimination of carbonates is achieved only aer heat treating at 750 °C. Multiple phase transitions from a low temperature cubic phase to a tetragonal phase and later to a highly conductive cubic phase are involved in this process and together with Li 2 CO 3 removal are responsible for a signicant increase in the local Li mobility inside the garnet structure. In the second part of this work, the impact of heat treatment at 750 °C as a pre-treatment step in the preparation of sintered pellets was studied. It was observed that pre-treatment of LLZO before sintering is not effective and has no big impact on the Li dynamics and ionic conductivities of sintered pellets. However, since the local mobility of the garnet aer heat treatment at 750 °C is alike the sintered pellets and the secondary phases at the surface are removed, we expect this heat-treatment to greatly enhance the ionic conductivity of LLZO in other systems such as ceramic-polymer composite electrolytes in which garnet llers are not sintered. This improvement could be achieved both by faster Li motion in the bulk of LLZO and by better processability of the garnet-rich composite solid electrolytes. These composite electrolytes will be thoroughly investigated in our future work. Considering the similar observations on the impact of heat treatment on the interfacial and Li transport properties in Nb and Al-doped LLZO, we believe that the general conclusion of this work could be applied to LLZO garnets, regardless of their doping composition. Author contributions PG: investigation, formal analysis, writing-original dra, and writing-review & editing. AP: investigation and writing-review & editing. KG: investigation. GA, PL, and DS: writing-review & editing. JL: investigation, supervision, and writing-review & editing. Conflicts of interest There are no conicts to declare. Acknowledgements P. G. as a part of the DESTINY PhD programme acknowledges funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie Actions COFUND –Grant Agreement No: 945357. This work was supported by the “Ministerio de Ciencia e Innovaci´ on/ Agencia Estatal de Investigaci´ on”, under the project grant TED2021-129663B-C52. Notes and references 1 J.-M. Tarascon and M. Armand, Nature, 2001, 414, 359–367. 2 B. Scrosati and J. Garche, J. Power Sources, 2010, 195, 2419– 2430. 3 M. Forsyth, L. Porcarelli, X. Wang, N. Goujon and D. Mecerreyes, Acc. Chem. Res., 2019, 52, 686–694. 4 M. Armand and J.-M. Tarascon, Nature, 2008, 451, 652–657. 5 T. Famprikis, P. Canepa, J. A. Dawson, M. S. Islam and C. Masquelier, Nat. Mater., 2019, 18, 1278–1291. 6 Q. Zhao, S. Stalin, C. Z. Zhao and L. A. Archer, Nat. Rev. Mater., 2020, 5, 229–252. 7 Y. Zheng, Y. Yao, J. Ou, M. Li, D. Luo, H. Dou, Z. Li, K. Amine, A. Yu and Z. Chen, Chem. Soc. Rev., 2020, 49, 8790–8839. 8 J. Li, C. Ma, M. Chi, C. Liang and N. J. Dudney, Adv. Energy Mater., 2015, 5, 1401408. 9 V. Thangadurai, S. Narayanan and D. Pinzaru, Chem. Soc. Rev., 2014, 43, 4714–4727. 10 J. Awaka, A. Takashima, K. Kataoka, N. Kijima, Y. Idemoto and J. Akimoto, Chem. Lett., 2011, 40,60–62. 11 S. Toda, K. Ishiguro, Y. Shimonishi, A. Hirano, Y. Takeda, O. Yamamoto and N. Imanishi, Solid State Ionics, 2013, 233, 102–106. 12 G. Larraz, A. Orera and M. L. Sanju´ an, J. Mater. Chem. A, 2013, 1, 11419–11428. 13 C. Deviannapoorani, S. Ramakumar, N. Janani and R. Murugan, Solid State Ionics, 2015, 283, 123–130. 14 J. Leng, H. Wang, H. Liang, Z. Xiao, S. Wang, Z. Zhang and Z. Tang, ACS Appl. Energy Mater., 2022, 5, 5108–5116. 15 H. Yamada, T. Ito, S. P. Kammampata and V. Thangadurai, ACS Appl. Mater. Interfaces, 2020, 12, 36119–36127. 16 A. Boulant, J. F. Bardeau, A. Jouanneaux, J. Emery, J. Y. Buzare and O. Bohnke, Dalt. Trans., 2010, 39, 3968– 3975. 17 A. Shara, S. Yu, M. Naguib, M. Lee, C. Ma, H. M. Meyer, J. Nanda, M. Chi, D. J. Siegel and J. Sakamoto, J. Mater. Chem. A, 2017, 5, 13475–13487. 18 L. Dhivya, K. Karthik, S. Ramakumar and R. Murugan, RSC Adv., 2015, 5, 96042–96051. 19 A. Shara, E. Kazyak, A. L. Davis, S. Yu, T. Thompson, D. J. Siegel, N. P. Dasgupta and J. Sakamoto, Chem. Mater., 2017, 29, 7961–7968. 20 Y. Li, X. Chen, A. Dolocan, Z. Cui, S. Xin, L. Xue, H. Xu, K. Park and J. B. Goodenough, J. Am. Chem. Soc., 2018, 140, 6448–6455. 21 F. Han, J. Yue, C. Chen, N. Zhao, X. Fan, Z. Ma, T. Gao, F. Wang, X. Guo and C. Wang, Joule, 2018, 2, 497–508. 22 Y. Zhang, J. Meng, K. Chen, Q. Wu, X. Wu and C. Li, ACS Appl. Mater. Interfaces, 2020, 12, 33729–33739. 23 M. Cai, Y. Lu, J. Su, Y. Ruan, C. Chen, B. V. R. Chowdari and Z. Wen, ACS Appl. Mater. Interfaces, 2019, 11, 35030–35038. 24 P. Zhao, Y. Xiang, Y. Wen, M. Li, X. Zhu, S. Zhao, Z. Jin, H. Ming and G. Cao, J. Eur. Ceram. Soc., 2018, 38, 5454–5462. 25 K. Ishiguro, Y. Nakata, M. Matsui, I. Uechi, Y. Takeda, O. Yamamoto and N. Imanishi, J. Electrochem. Soc., 2013, 160, A1690. 26 I. McClelland, H. El-Shinawi, S. G. Booth, A. Regoutz, J. Clough, S. Altus, E. J. Cussen, P. J. Baker and S. A. Cussen, Chem. Mater., 2022, 34, 5054–5064. 11682 |J. Mater. Chem. A,2023,11, 11675–11683 This journal is © The Royal Society of Chemistry 2023 Journal of Materials Chemistry A Paper Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online 27 Z. Cao, W. Wu, Y. Li, J. Zhao, W. He, J. Liu, H. Zhang and G. Li, Ionics, 2020, 26, 4247–4256. 28 Y. Zhang, D. Luo, W. Luo, S. Du, Y. Deng and J. Deng, Electrochim. Acta, 2020, 359, 136965. 29 H. Geng, K. Chen, D. Yi, A. Mei, M. Huang, Y. Lin and C. Nan, Rare Met. Mater. Eng., 2016, 45, 612–616. 30 Y. Meesala, Y. K. Liao, A. Jena, N. H. Yang, W. K. Pang, S. F. Hu, H. Chang, C. E. Liu, S. C. Liao, J. M. Chen, X. Guo and R. S. Liu, J. Mater. Chem. A, 2019, 7, 8589–8601. 31 F. Tietz, T. Wegener, M. T. Gerhards, M. Giarola and G. Mariotto, Solid State Ionics, 2013, 230,77–82. 32 T. Thompson, J. Wolfenstine, J. L. Allen, M. Johannes, A. Huq, I. N. David and J. Sakamoto, J. Mater. Chem. A, 2014, 2, 13431–13436. 33 C. Bernuy-Lopez, W. Manalastas, J. M. Lopez Del Amo, A. Aguadero, F. Aguesse and J. A. Kilner, Chem. Mater., 2014, 26, 3610–3617. 34 P. Ranque, J. Zag´ orski, S. Devaraj, F. Aguesse and J. M. L´ opez del Amo, J. Mater. Chem. A, 2021, 9, 17812–17820. 35 G. Larraz, A. Orera, J. Sanz, I. Sobrados, V. Diez-G´ omez and M. L. Sanju´ an, J. Mater. Chem. A, 2015, 3, 5683–5691. 36 M. Leskes, A. J. Moore, G. R. Goward and C. P. Grey, J. Phys. Chem. C, 2013, 117, 26929–26939. 37 T. Liu, Z. Liu, G. Kim, J. T. Frith, N. Garcia-Araez and C. P. Grey, Angew. Chem., Int. Ed., 2017, 56, 16057–16062. 38 F. Nardelli, S. Borsacchi, L. Calucci, E. Carignani, F. Martini and M. Geppi, Rend. Lincei Sci. Fis. Nat., 2020, 31, 999–1010. 39 P. Posch, S. Lunghammer, A. Wilkening, K. Hogrefe and H. M. R. Wilkening, JPhys Energy, 2023, 5, 15001. 40 A. Kuhn, M. Wilkening and P. Heitjans, Solid State Ionics, 2009, 180, 302–307. 41 J. R. Hendrickson and P. J. Bray, J. Magn. Reson., 1973, 9, 341–357. 42 M. Wilkening, D. Bork, S. Indris and P. Heitjans, Phys. Chem. Chem. Phys., 2002, 4, 3246–3251. 43 J. S. Waugh and E. I. Fedin, Sov. phys., Solid state, 1963, 4, 1633–1636. 44 O. Bohnke, J. Emery, A. Veron, J. L. Fourquet, J. Y. Buzare, P. Florian and D. Massiot, Solid State Ionics, 1998, 109,25– 34. 45 V. Epp, ¨ O. Gün, H. J. Deiseroth and M. Wilkening, J. Phys. Chem. Lett., 2013, 4, 2118–2123. 46 A. Kuhn, S. Narayanan, L. Spencer, G. Goward, V. Thangadurai and M. Wilkening, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83,1–11. 47 O. Bohnke, J. Emery and J.-L. Fourquet, Solid State Ionics, 2003, 158, 119–132. 48 M. Wilkening and P. Heitjans, ChemPhysChem, 2012, 13,53– 65. 49 H. Buschmann, J. Dölle, S. Berendts, A. Kuhn, P. Bottke, M. Wilkening, P. Heitjans, A. Senyshyn, H. Ehrenberg, A. Lotnyk, V. Duppel, L. Kienle and J. Janek, Phys. Chem. Chem. Phys., 2011, 13, 19378–19392. 50 A. I. Waidha, T. Ferber, M. Donzelli, N. Hosseinpourkahvaz, V. Vanita, K. Dirnberger, S. Ludwigs, R. Hausbrand, W. Jaegermann and O. Clemens, ACS Appl. Mater. Interfaces, 2021, 13, 31111–31128. 51 V. Vanita, A. I. Waidha, S. Yadav, J. J. Schneider and O. Clemens, Int. J. Appl. Ceram. Technol., 2023, 20, 236–250. This journal is © The Royal Society of Chemistry 2023 J. Mater. Chem. A,2023,11, 11675–11683 | 11683 Paper Journal of Materials Chemistry A Open Access Article. Published on 28 April 2023. Downloaded on 10/20/2024 10:17:52 PM. This article is licensed under a Creative Commons Attribution-NonCommercial 3.0 Unported Licence. View Article Online Investigating the role of interphases in composite electrolytes by solid-state NMR 62 7.2 Publication #2: Title: Influence of the LLZO–PEO interface on the microand macro-scale properties of composite polymer electrolytes for solid-state batteries Authors: Pedram Ghorbanzade, Grazia Accardo, Kerman Gómez, Pedro López-Aranguren, Shanmukaraj Devaraj, Carlos Miguel Costa, Senentxu Lanceros-Mendez, and Juan Miguel López del Amo Journal: Materials Today Energy Journal impact factor: 9.0 Journal quartile: Q1 (Materials Science) Influence of the LLZOePEO interface on the microand macro-scale properties of composite polymer electrolytes for solid-state batteries Pedram Ghorbanzade a , b , c , Grazia Accardo a , Kerman Gomez a , Pedro L opez-Aranguren a , Shanmukaraj Devaraj a , Carlos Miguel Costa d , Senentxu Lanceros-Mendez d , e , f , Juan Miguel L opez del Amo a , * a Centre for Cooperative Research on Alternative Energies (CIC energiGUNE), Basque Research and Technology Alliance (BRTA), Alava Technology Park, Albert Einstein 48, 01510 Vitoria-Gasteiz, Spain b University of Basque Country (UPV/EHU), Barrio Sarriena, s/n, 48940 Leioa, Spain c ALISTORE-European Research Institute, 80039 Amiens, France d Center of Physics Universities of Minho and Porto (CFUM-UP) and Laboratory of Physics for Materials and Emergent Technologies, LapMET, University of Minho, 4710-057 Braga, Portugal e BCMaterials, Basque Center for Materials, Applications and Nanostructures, UPV/EHU Science Park, 48940 Leioa, Spain f Ikerbasque Basque Foundation for Science, 48009 Bilbao, Spain article info Article history: Received 25 August 2023 Received in revised form 9 October 2023 Accepted 24 October 2023 Available online 31 October 2023 Keywords: Solid-state electrolytes Garnet-rich composite electrolytes Hybrid electrolytes Garnetepolymer interface abstract Li 7 La 3 Zr 2 O 12 (LLZO) garnets offering high ionic conductivity and electrochemical stability are among the most promising ceramic materials for lithium metal solid-state batteries. Although their application in composite polymer electrolytes (CPEs) with poly(ethylene oxide) (PEO) has been widely studied, their surface chemistry which is influenced by their hygroscopic nature is often neglected. This work reports on how the thermal treatment and the consequent elimination of secondary phases at the LLZOePEO interface impacts the microstructure of the garnet-rich CPEs, which in turn affects their mechanical and ion-transport properties. It is shown that LLZO heat treatment restricts local polymer chain motions, indicating reinforcement of PEO/LLZO interactions which enhances the mechanical strength and homogeneity of the CPEs. These micro-scale modifications of CPEs eventually increase their ionic conductivity and improve the solid electrolyte/lithium metal interface. Thus, the Li þ exchange at the PEO/ LLZO interface was studied by using 7 Lie 7 Li exchange spectroscopyenuclear magnetic resonance , and it was observed that after LLZO heat treatment, the interfacial Li þ exchange has significantly decreased, in line with a lower LiOH content. This result confirms the role of LiOH as an intermediate in the Li þ exchange reaction and that surface chemistry plays a more important role in the Li þ exchange than the local Li mobilities in the individual phases. ©2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Introduction Current Li-ion batteries (LIBs) based on liquid electrolytes and intercalation electrodes offer nowadays high gravimetric and volumetric energy densities above 260 Wh/kg and 730 Wh/L at cell level, approaching their theoretical energy density [1,2]. However, the flammable nature and low thermal stability of their organic solvents raise serious safety concerns, particularly for large-scale applications such as electric vehicles [3,4]. Therefore, to move toward safer batteries, solid-state electrolytes are promising alternatives to substitute the flammable organic electrolytes in LIBs [5,6]. In addition, solid electrolytes could arguably mitigate the Li dendrite growth, and in this way, they could potentially enable the use of Li metal as the anode, which would significantly increase the gravimetric energy density of the cell [7,8]. Furthermore, solid electrolytes enable the application of bipolar electrodes, thus decreasing the mass and volume of the package and increasing the cell energy density [5]. Solid electrolytes are typically divided into three groups: [9] i) polymer electrolytes, having simple processability and an acceptable interface with the electrodes; ii) inorganic electrolytes, having high thermal, mechanical, and electrochemical stability as well as high ionic conductivity; iii) composite polymer *Corresponding author. E-mail address: [email protected] (J.M. L opez del Amo). Contents lists available at ScienceDirect Materials Today Energy journal homepage: www.journals.elsevier.com/materials-today-energy/ https://doi.org/10.1016/j.mtener.2023.101448 2468-6069/©2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Materials Today Energy 38 (2023) 101448 electrolytes (CPEs); combining the previous two, to ideally combine the high conductivity of inorganics with the better interfacial properties of polymers. poly(ethylene oxide) (PEO) is the most widely used polymer in CPEs [10]. However, other polymers such as polyethylene carbonate and polycaprolactone have also been successfully used [11]. Lithium bis(trifluoromethanesulfonyl)imide (LiTFSI) with its bulky anion structure is the most common Li salt in studying CPEs, although lithium perchlorate (LiClO 4 ) and lithium trifluoromethanesulfonate (LiTF) have also been investigated [12e14]. Among Li þ conductors, Li 7 La 3 Zr 2 O 12 , (LLZO) garnets are the top candidates, because they offer a very high ionic conductivity at room temperature [15], can effectively increase the electrochemical stability of the composites [16], and improve the interface with Li metal [17]. Therefore, this study focuses on CPEs made of PEO/ LiTFSI and LLZO. Most studies on CPEs focus on ceramic-in-polymer systems with a low fraction of the inorganic phase. It has been shown that the addition of a small amount of filler successfully improves the ionic conductivity of the CPE. These improvements are usually associated with the modification of the polymer phase by decreasing its crystallinity [18e20]. However, Wieczorek et al. [21] pointed out that the Lewis acid-base nature of the interactions between the polyether host and the inorganic fillers affects the Li þ conductivity. Further, Wang et al. [22] compared the effect of Li-conducting (active) fillers with non-conducting (passive) fillers and attributed the improvement of the ionic conductivity to the filler/polymer interface. Zaman et al. [23] reported similar observations in a PEO/LLZO system with 15 vol% LLZO. Nevertheless, in all cases, the inorganic active filler does not contribute to the ion transport, due to the filler fraction being below its percolation threshold and no continuous network is formed. Consequently, no long-range Li-ion conduction occurs through the ceramic phase [16]. An approach toward improving LLZO contribution in ionic conductivity is increasing its content, and moving from ceramic-in-polymer to polymer-in-ceramic electrolytes. These composites, which have recently attracted huge attention, are very interesting from a fundamental point of view, as larger fractions of LLZO increase the ceramicepolymer interfaces and the Li þ transport pathway is more complex. Furthermore, with the incorporation of a higher amount of LLZO, the mechanical and electrochemical stability of the electrolyte can be improved [16,18,24]. One of the major drawbacks of LLZO is the very fast H þ /Li exchange reaction after exposure to humidity, forming a layer of LiOH which turns into Li 2 CO 3 upon reaction with CO 2 [25e27]. Due to their hygroscopic nature, in most cases, the LLZO surface is covered by secondary phases such as LiOH and Li 2 CO 3 . In our previous study [28], we showed that a heat treatment at 750  C is highly effective in removing these secondary phases and improves the processability of garnet-rich composites. In addition, it was shown that this heat treatment significantly enhances Li þ dynamics in the LLZO, which could reflect in their ionic conductivity. In this work, we investigate in more detail the impact of the LLZO heat treatment on the morphological, mechanical, and Li transport properties of garnet-rich CPEs, as well as the impact of heat treatment on the Li þ exchange at PEO/LLZO interface. Finally, we assess the electrochemical properties and interface with Li metal of the symmetric cells to demonstrate the suitable performance of the optimized solid electrolytes. 2. Experimental section Commercial LLZO (Li 6.75 La 3 Zr 1.75 Nb 0.25 O 12 ) purchased from Toshima was used in this work. The powders were first ball-milled to reduce the particle size. Part of the powders was placed in a furnace at 750  C for 12 h to remove the decomposition products from their surface and obtain heat-treated LLZO. A soft ball-milling procedure was used for the preparation of the garnet-rich composite electrolytes. Poly(ethylene-oxide) (PEO, 110 6 g/mol, Sigma Aldrich) was mixed with lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Solvionic), LLZO, and acetonitrile (ACN, Sigma Aldrich). The EO:Li ratio of 20:1 was kept constant in all samples. The mixture was ball-milled for 25 min in 5-min intervals with a frequency of 350 rpm, and then cast in 4 mm Teflon evaporating dishes. The solvent was evaporated at room temperature under continuous argon flow overnight. The membranes were later dried at 50 C under dynamic vacuum to ensure complete solvent evaporation. Finally, solid membrane electrolytes with a thickness of 120 m m were carefully removed from the evaporating dish using a tweezer. Except for the ball-milling which was performed inside air-tight jars, all the preparation and handling of the membranes were done in an Ar-filled glove box. The composites made of 80 wt% pristine and heat-treated LLZO are referred to as CP80 and CA80 hereinafter. The micrographs were obtained using a scanning electron microscope (SEM Quanta FEG 250) by applying an acceleration voltage of 10 kV and utilizing an Everhart-Thornley detector. Particle size distribution was analyzed using a Mastersizer 3000 (Malvern Panalytical). The stressestrain curves of the CPEs were evaluated at room temperature with a TST350 tensile device from Linkam Scientific Instruments at a strain rate of 15 m m/s. For each sample, the mechanical curves were tested in triplicate with dimensions of 10 70.1 mm. Young modulus was obtained through the tangent method in the elastic region at 3% of deformation. All magic angle spinning nuclear magnetic resonance (MAS NMR) spectra were recorded using a Bruker Avance III 500 spectrometer, a 2.5 mm probe and a MAS frequency of 20 kHz. 1 H and 7 Li chemical shifts were referenced to bulk water and 0.1 M LiCl aqueous solution, respectively. 7 Li 1D spectra were recorded using single excitation pulses of 2.4 m s and a relaxation delay of 250 s. For the 1 H 1D spectra, Hahn echo experiments were used in which a p / 2 pulse (1.9 m s) is followed by a p pulse (3.8 m s), and the recycle delay is set to 5s. 1 He 7 Li Heteronuclear correlation (HETCOR) experiments were recorded applying a cross polarization (CP) 1 He 7 Li magnetization transfer step of 1 ms. 7 Lie 7 Li exchange spectroscopy (EXSY) correlations were obtained using standard three p /2 pulse experiments of 1.9 m s. The combined CP-EXSY experiment shown in Fig. 5d was performed by using an initial 1 He 6 Li CP step followed by a 6 Lie 6 Li EXSY, with a mixing time of 1.6 ms and relaxation delay of 3 s. The electrochemical impedance spectra of the CPEs were recorded using a Biologic VPM3 with AC in the frequency range of 1 MHze1 Hz at an excitation voltage of 10 mV. The cells were prepared with stainless steeleblocking electrodes and the spectra were recorded at the temperature range of 30e80  Cwith 10  C intervals. The ionic conductivities were calculated using equation (1). s ¼d A$R(1) where dis the electrolyte thickness, A is its surface area and R is the electrolyte resistance, obtained from the impedance spectrum. Cyclic voltammetry was conducted using a Biologic VPM3 in a potential range of 0.5 to 2.5 V vs. Li  /Li þ , at a scan rate of 10 mV/s for 10 scans. The measurements were performed on Li metal/CPE/ stainless steel cells at 70  C and the cells were conditioned for 2 h at this temperature prior to measurement. P. Ghorbanzade, G. Accardo, K. Gomez et al. Materials Today Energy 38 (2023) 101448 2 The Li plating/stripping tests were performed on Li symmetrical cells using a Maccor Series 4000 at 70  C and a current density of 40 m A/cm 2 . The cycling was stopped after ~500 h, to collect the electrochemical impedance spectra, and resumed at 80 m A/cm 2 . The impedance spectra of Li symmetrical cells both before and after cycling was recorded at 70  C and with the same parameters mentioned above. The equivalent circuit model used for fitting the impedance data consists of a resistance element for external resistance in series with two R//CPE blocks, corresponding to the bulk and electrode/electrolyte interface resistance. 3. Results and discussions To begin with, the impact of heat treatment on the particle size distribution of the ball-milled LLZO powders was studied. As demonstrated in Fig. S.1, it was observed that overall, the LLZO particle size and agglomeration were decreased, which could be attributed to the elimination of LiOH and Li 2 CO 3 layers upon heat treatment. In the next step, garnet-rich composites with 80 wt% pristine and heat-treated LLZO were prepared to investigate the impact of LLZO heat-treatment at 750  C as a pre-processing step on different properties of CPEs. To understand the morphology and garnetepolymer interface in each sample, SEM analysis was performed. The micrograph of CP80 (Fig. 1a and c) shows that LLZO particles are agglomerated, and the two phases (polymer and ceramic) are segregated. On the other hand, in the case of CA80 (Fig. 1b and d), LLZO particles are uniformly dispersed in the polymer matrix. Almost all LLZO particles are homogenously covered by PEO and the polymer distribution is rather uniform. The improved LLZO wetting by the polymer implies stronger interactions between PEO and LLZO after the heat treatment. From the comparison of the SEM images, a better LLZOePEO interface in the membrane with heat-treated LLZO is evident, likely due to the removal of LiOH and Li 2 CO 3 from the LLZO surface upon heat treatment. When the garnet fraction is increased to 90 wt%, the heat treatment and the improvement of the interface will be even more relevant, due to the increase of the solid-solid interfaces. Moreover, with a fraction as high as 90 wt% LLZO, the heat treatment becomes imperative for the successful processing of selfstanding membranes [28]. In solid-state batteries, the evaluation of the mechanical properties of solid electrolytes is essential as they affect battery safety and performance [29,30]. Therefore, tensile tests were performed to evaluate the mechanical strength of CPEs and the LLZOefree PEO/LiTFSI membranes described in Fig. 1.Fig. 2 shows the characteristic mechanical response for solid polymer electrolytes, where a typical thermoplastic behavior is observed, with two characteristic regions, the elastic, and the plastic regimes. It can be observed that the addition of LLZO improves the mechanical properties of the membranes, in the sense that the ceramic particles act as mechanical reinforcement. Table 1 presents the Young's modulus, yield stress, and elongation at break of each sample. PEO/ LiTFSI membrane shows the lowest yield stress (0.41 MPa) but the highest elongation at break. CA80 on the other hand, possesses the highest yield stress (0.91 MPa), with a Young's modulus similar to CP80 (~35 MPa) but considerably higher than the one of PEO/LiTFSI (2.1 MPa), indicating that the addition of LLZO increases the stiffness of the membranes. CP80 shows a significantly lower elongation at break (42.7 %) compared to CA80 and PEO/LiTFSI (109.8 and 83.7%), which is in line with its phase-segregated microstructure in which LLZO particles are agglomerated and PEO matrix is not uniformly distributed. The lower yield stress of CP80 can also be explained by the poor PEO/LLZO interactions, in agreement with SEM micrographs. On the contrary, in CA80 the homogenous distribution of PEO and the stronger PEO/LLZO interactions (wettability of the LLZO by the polymer) noticeably increase the yield stress and the elongation at break. Overall, the mechanical properties of CA80 are suitable for solid-state batteries [31]. In any case, Fig. 1. SEM micrographs of a, c) CP80 and b, d) CA80. P. Ghorbanzade, G. Accardo, K. Gomez et al. Materials Today Energy 38 (2023) 101448 3 Investigating the role of interphases in composite electrolytes by solid-state NMR 72 7.3 Publication #3: Title: A Vision for LLZO Carbonate Formation: Perspectives on Surface Treatment Approaches and Characterization Techniques Authors: Pedram Ghorbanzade, Pedro López-Aranguren, and Juan Miguel López del Amo Journal: ChemElectroChem Journal impact factor: 3.5 Journal quartile: Q2 (Electrochemistry) A Vision for LLZO Carbonate Formation: Perspectives on Surface Treatment Approaches and Characterization Techniques Pedram Ghorbanzade,*[a, b, c] Pedro López-Aranguren,[a] and Juan Miguel López del Amo*[a] Li7La3Zr2O12 garnets are among the most promising materials for solid-state electrolytes thanks to their high electrochemical stability and ionic conductivity. However, they are unstable in air and easily undergo a Li+/H+exchange reaction, forming a lithiophobic and poorly Li-conducting surface layer composed of LiOH and Li2CO3upon further reaction with CO2. Despite a simple reaction mechanism, protonation of LLZO is a complex phenomenon. The thickness and composition of the secondary phases on the surface depend on many parameters both in the LLZO properties and in the storage atmosphere. Due to this complexity, proposing a universal procedure for removing the secondary phases is not simple. However, understanding the role of different parameters and rational selection of characterization techniques makes it possible to find the optimal conditions for removing the LiOH/Li2CO3layer and further processing each material. This work compares some of the main approaches for cleaning the LLZO surface and describes their advantages and shortcomings. Additionally, the strengths and limitations of the most common characterization techniques for analyzing the (de)protonation of LLZO (XRD, XPS, Raman, NMR, TGA) are explained to help researchers design their experiments more accurately. 1. Introduction Among many different solid-state electrolytes (SSEs), garnet type LLZO (Li7La3Zr2O12) is one of the top candidates, thanks to its broad electrochemical stability window which makes it compatible with high-voltage cathodes,[1] high ionic conductivity (up to 1×103Scm1),[2,3] and chemical stability of its interface with Li metal.[4,5] LLZO appears in tetragonal and cubic phases, with only the latter offering high ionic conductivity. To stabilize the cubic phase at room temperature and achieve higher ionic conductivities, doping and aliovalent substitution of Zr4+is a common approach.[6,7] The reactivity of LLZO towards the humid air and the formation of LiOH and Li2CO3on its surface is a well-known but non-trivial process, with several parameters playing a role in the content and composition of this layer. This work reviews the affecting parameters in two groups of (i) LLZO properties and (ii) exposure conditions, and discusses the common methods for removing the LiOH/Li2CO3layer. As understanding and following the effects of (de)protonation mechanisms of LLZO requires adequate characterization tools, an overview of the usual characterization techniques used in studying the (de)protonation of LLZO and their limitations is presented to better design the experiments. 2. Protonation and Affecting Parameters Garnets were initially considered to be stable in air.[8] However, many works later revealed that LLZO is highly hygroscopic and can react with moisture to form LiOH through a Li+/H+ exchange,[9,10] according to reaction 1.[11,12] Li7La3Zr2O12 þx H2O!Li7xHxLa3Zr2O12 þx LiOH (1) LiOH can then react with CO2to form Li2CO3according to reaction 2.[11,12] 2 LiOH þCO2!Li2CO3þH2O(2) Combining experimental and DFT calculations, Sharafi et al. investigated the reaction mechanism and found out that the direct hydration and carbonation of LLZO are thermodynamically unfavorable and that the LiOH can be formed only through the Li+/H+exchange.[11] Interestingly, this exchange reaction can experimentally go up to 60%, without changing the crystal structure or phase transition.[13] In fact, protonation can stabilize the cubic phase at room temperature.[14,15] However, the ionic conductivity of this low-temperature cubic phase is much lower than the highly Li-conductive phase.[16] In addition, it was shown that inserted protons are immobile [a] P. Ghorbanzade, Dr. P. López-Aranguren, Dr. J. M. López del Amo Centre for Cooperative Research on Alternative Energies (CIC energiGUNE) Basque Research and Technology Alliance (BRTA) Albert Einstein 48, 01510 Vitoria-Gasteiz, Spain E-mail: [email protected] [email protected] [b] P. Ghorbanzade University of Basque Country (UPV/EHU) Barrio Sarriena, s/n, 48940 Leioa, Spain [c] P. Ghorbanzade ALISTORE-European Research Institute 80039 Amiens, France © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 80/86] 1 ChemElectroChem 2024,11, e202400136 (1 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem www.chemelectrochem.org Concept doi.org/10.1002/celc.202400136 below 400 K, and do not compensate for the loss in Li+ conductivity.[13,17] In general, protonation of LLZO strongly depends on: 2.1. LLZO Properties 2.1.1. Form/Geometry LLZO is generally pelletized and sintered at temperatures above 1100°C to achieve a dense pellet and reduce the grain boundaries. Although the facile and rather low-cost preparation of LLZO pellets makes them ideal for research studies, other forms of LLZO solid electrolytes such as LLZO sheets, thin films, or ceramic-polymer composite electrolytes should not be neglected. In fact, due to a higher surface area in these systems, the protonation would be more pronounced and the surface properties of the LLZO will be of higher importance.[18,19] 2.1.2. Pellet Density and Grain Boundaries An approach to reduce the LLZO protonation is maximizing the pellet density to reduce the interaction with moisture by decreasing the effective surface area, and the ratio of grain boundaries which can accommodate the LiOH and Li2CO3.[20,21] Selecting the proper type and content of dopants, crucibles, and sintering additives is important to obtain a highly dense pellet.[20,22–25] For instance, LiF can be used as a sintering aid, to locate at the surface and the grain boundaries of the LLZO pellet, to prevent the diffusion of H2O and CO2into the pellet, thus protecting the LLZO.[26] 2.2. Exposure Conditions As expected from reactions 1 and 2, depending on the relative humidity (RH) and the amount of CO2in the atmosphere, the ratio of LiOH to Li2CO3in the decomposition layer can vary.[27] Considering that the direct reaction of LLZO with CO2is minimal, the LiOH is the intermediate for Li2CO3formation, thus relative humidity is the key parameter to control. Although the CO2content is an important parameter which defines the ratio of Li2CO3to LiOH in the surface layer, it is very challenging to precisely control, particularly in dry rooms. Therefore, minimizing the relative humidity is more crucial from a practical point of view. It has been reported that exposing LLZO to dry air (RH �0.5%) instead of ambient air (RH�50%), significantly reduces the formation of Li2CO3.[11] It has also been shown that the growth of the contamination layer continues when extending the exposure time.[11] This implies that the storage conditions and the time of exposure are highly decisive for the handling. Nevertheless, complete protection of LLZO from Li2CO3formation is extremely challenging because even inside an argonfilled glovebox, traces of H2O (<0.5 ppm) and CO2(<5 ppm) can form a 4–5 nm layer of LiOH/Li2CO3on the LLZO surface in less than 30 mins.[28,29] Considering the longer storage time of commercial garnets, they might contain a higher content of secondary phases compared to powders freshly calcined in the lab. While this will Pedram Ghorbanzade obtained his B.S degree in Polymer Engineering from the University of Tehran in 2018. He then started his Erasmus Mundus Master studies in Materials for Energy Storage and Conversion (MESC+). Soon after, he joined CIC EnergiGUNE to start his doctoral research within Marie Sklodowska-Curie COFUND DESTINY European Doctorate programme under the supervision of Dr. López del Amo. He is currently a visiting researcher at the Grey group (University of Cambridge). His research mainly focuses on the design of composite polymer-ceramic electrolytes and their characterization by solid state NMR. Dr. Pedro López-Aranguren is responsible for the Ceramic Electrolyte research line at CIC EnergiGUNE in Vitoria-Gasteiz, Spain. He obtained his Ph.D. in materials science from the UAB in Barcelona, Spain, in 2014. In 2015 he was awarded with a Marie Curie fellowship. He contributed to the EU ECOSTORE project, gaining unique expertise in electrochemistry and on the new-generation solid-state batteries. His main research focuses on addressing the challenges and bottlenecks in oxides, sulfides, phosphates, halides and composite electrolytes for advancing solid-state battery technology. Dr Juan Miguel López del Amo completed his Ph.D. at the Chemistry Department of the Free University of Berlin in 2006. His thesis was dedicated to the development and applications of solid-state NMR for the physicochemical characterization of solid compounds. From 2007 to 2012, he joined the FMP center of Berlin (Leibniz Institute) and the Helmholtz Institute of Structural Biology (Munich), where his research was primarily focused on biosolid-state NMR. Since 2012, he is the head of the Solid State NMR department at CIC energiGUNE, where he investigates electrodes and solid electrolyte materials. Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 81/86] 1 ChemElectroChem 2024,11, e202400136 (2 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License not make any huge difference for the pellets or green bodies sintered at high temperatures,[30] it must be seriously considered for applications such as manufacturing composite electrolytes or cold-sintering in which high sintering temperatures are not involved.[31] (see figure 1). It should also be mentioned that the Li+/H+exchange of LLZO happens not only when exposed to humid air, but also when in contact with solvents. Kun et al. showed that the Li+/H+exchange is more intense in protic solvents with higher acidic character, nevertheless, it does occur in aprotic solvents such as ACN and c-Hex.[17] This must be considered in wet-processing methods of LLZO solid electrolytes. 3. Surface Cleaning The presence of LiOH and Li2CO3on the surface of LLZO may have a negative impact on the properties and performance of LLZO electrolytes, particularly on the interfacial resistance with the Li metal and the cell cyclability.[32] This is mainly due to the lithiophobic and poorly-conductive nature of Li2CO3which needs to be removed.[12,32] A successful “cleaning” of the LLZO surface will result in the reduction of the area specific resistance (down to 2 Ωcm2), which can in turn increase the critical current density (>0.3 mAcm2) and enhance the cycling performance of the cell.[32] Despite the higher vulnerability of LLZO powders to protonation and their higher surface area, removing the secondary phases from the LLZO surface is usually neglected in the case of LLZO-polymer composite electrolytes. Nevertheless, it was shown that the removal of LiOH/Li2CO3can effectively improve the LLZO-PEO interactions and the wetting, resulting in improved mechanical properties, ionic conductivity, and cyclability.[33] In addition, it is reported that a polymer-LLZO interface free of Li2CO3can also increase the Li transference number, modify the Li+transport pathways, and suppress the Li dendrite growth.[12] Herein different methods to “clean” the LLZO surface are listed and their advantages and shortcomings are discussed. 3.1. Mechanical Polishing Mechanical polishing, which can be done in dry or wet form, is a simple and low-cost approach to remove the LiOH/Li2CO3 layer from the surface of the LLZO. Upon polishing, the LLZO/Li metal interfacial resistance will decrease significantly.[32,34] However, it has the following shortcomings. (i) can only be applied to pellets, not powders. (ii) not easy to scale up (iii) cannot completely remove the LiOH and Li2CO3, as some will remain in the grain boundaries.[26] (iv) cannot deprotonate and recover the LLZO. 3.2. Acid Treatment Another low-cost and simple method to remove the Li2CO3, is through a chemical reaction with acids such as HF, HCL, and citric acid. According to reactions 3 and 4, the Li2CO3will decompose to form a layer of LiCl and LiF, coating the LLZO surface.[35,36] Li2CO3þ2 HF !2 LiF þCO2þH2O(3) Li2CO3þ2 HCl !2 LiCl þCO2þH2O(4) The acids can also react with LiOH to form Li salts, according to reactions 5 and 6.[37,38] LiOH þ2 HF !2 LiF þH2O(5) LiOH þ2 HCl !2 LiCl þH2O(6) LiF and LiCl are quite lithiophilic and as a result, they decrease the interfacial resistance with the Li metal.[35,39] In addition, the coating layer can passivate the LLZO and prevent its rehydration.[26,40] Considering the low cost of acid treatment and the passivation of LLZO which allows its exposure to air for some time, this approach seems attractive for the manufacturing scale up. However, it comes with the following drawbacks: (i) A long treatment time can corrode the LLZO.[36] (ii) Although the Li2CO3is removed from the surface, the produced H2O can further protonate the LLZO by Li+/H+substitution.[41] (iii) The LLZO is not recovered and the Li+of the Li2CO3turns into lowconducting LiCl or LiF, rather than forming back the highly conducting LLZO. (iv) Due to the possible extra protonation and waste of Li, this approach is not suitable for LLZO powders, where a high surface area and high content of Li2CO3exist.[12] 3.3. Heat Treatment A very straightforward approach to eliminate the secondary phases is to heat the LLZO at temperatures between 500– Figure 1. illustration of Li2CO3content in freshly calcined and commercial LLZO powders, and the impact of different processing methods on the final composition. Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 82/86] 1 ChemElectroChem 2024,11, e202400136 (3 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 900°C. By doing this, all the secondary phases including the LiOH, the adsorbed water, protons inside the LLZO, and Li2CO3 will be gradually removed at different temperatures, following reactions 7–10.[14,30] Li7xHxLa3Zr2O12:nH2O!Li7xHxLa3Zr2O12 þn H2O(7) Li7xHxLa3Zr2O12 !Li7xLa3Zr2O12x 2þx=2 H2O(8) Li7xHxLa3Zr2O12 þx LiOH !Li7La3Zr2O12 þx H2O(9) Li7xLa3Zr2O12x 2þx 2Li2CO3!Li7La3Zr2O12 þx 2CO2(10) The recovery of LLZO through the decomposition of Li2CO3 (equation 8) is a very important feature of this approach, making it especially suitable for cases with high surface area and high content of Li2CO3, such as LLZO powders and thin films. In addition, heat treatment is the only method that eliminates the protons with low mobility from the LLZO structure, resulting in a decrease in the activation energy for Li jumps in LLZO.[30] These faster dynamics could lead to higher ionic conductivities, especially at low temperatures at which protons have lower mobility. Despite all the mentioned advantages, heat treatment comes with the risk of formation of the Li deficient La2Zr2O7 phase (pyrochlore) and Li loss, which can occur at temperatures above 600°C,[42] as depicted in reaction 11.[43] Li7La3Zr2O12 $La2Zr2O7þLi2O(11) Nevertheless, at sufficiently high temperatures, LLZO can be restored either through reaction 9 in reverse direction,[44] or through reaction of the pyrochlore with Li2CO3according to reaction 12.[6,43] Li2CO3þLa2Zr2O7!Li7La3Zr2O12 þCO2(12) The temperature, duration, and the atmosphere of the heat treatment can strongly influence the formation of pyrochlore, hence the ionic conductivity of the LLZO.[42,45] It is reported that a shorter dwell time and higher heating rates can efficiently suppress the LLZO decomposition to achieve a fully lithiated LLZO phase.[6,42] Regarding the gas flow, heat treatment in atmospheres containing oxygen is usually preferred, as it prevents both oxygen loss at elevated temperatures, and formation of carbonated species on the surface, ultimately reducing the interfacial resistance.[30,45,46] Unlike the acid treatment (where a passivation layer of LiF or LiCl is formed), after the heat treatment, the “cleaned” surface of LLZO is prone to rehydration and carbonation. Therefore, it is essential to perform the heat treatment at inert atmosphere or vacuum and protect the LLZO afterward in an inert atmosphere. The temperature and duration of the heat treatment necessary to remove LiOH and Li2CO3depend on the extent of protonation, composition of the LiOH/Li2CO3layer, and atmosphere of the heat treatment, thus varying from one system to another. Such complexity explains the discrepancies in the literature regarding the optimal heat treatment temperature/ duration, ranging from 500 to 900°C.[6,32,42,45] As a result, it is extremely difficult to suggest a general heat treatment procedure for cleaning the surface of all types of LLZO electrolytes, without its degradation to pyrochlore. However, knowing the deciding parameters and using the right characterization techniques, it is possible to optimize the heat treatment parameters to achieve LLZO with clean surface without degradation. Overall, we suggest heat treatment as the most effective approach to deprotonate the LLZO and remove the LiOH/Li2CO3layer from its surface. 4. Characterization Techniques For selecting the proper set of characterization techniques to assess the LLZO (de)protonation and the (de)formation of LiOH/ Li2CO3layers; the strength and limitations of different tests, thickness of the decomposition layer, and the specific location of different chemical species (whether on the surface or within the bulk) must be recognized. This careful consideration ensures a comprehensive understanding of the materials’ characteristics and contributes to the accuracy and reliability of the analyses conducted. For instance, a thin layer of LiOH/Li2CO3 will be underestimated when characterized using a bulk technique. Thus, a combination of techniques with different probing depths and sensitivities are recommended and will be here in discussed. 4.1. Thermal Gravimetry Analysis (TGA) TGA is a simple method to get an overview of the extent of protonation. The sample is heated in a controlled atmosphere and the mass loss is correlated to different thermochemical processes such as dehydration, desorption, and decomposition, thus giving a quantitative estimation of different chemical species in the sample, regardless of being on the surface or in bulk.[47,48] TGA has no limitation for sample geometry but is especially useful for powders, as they are more prone to protonation. 4.2. Powder X-ray Diffraction (PXRD) XRD is a powerful technique to identify different phases of LLZO and the presence of pyrochlore, Li2CO3and LiOH.[6,49] As a big advantage, it can also operate in situ to observe the elimination of secondary phases in real time.[42,45] Considering the alteration of the lattice parameter by the Li+/H+substitution, it can also reveal the state of protonation of the LLZO.[17] However, a main drawback of conventional powder-XRD is that in most cases, the LiOH/Li2CO3layer is too thin (<1μm) and infiltrative to obtain a clear diffraction signal. In this case, Grazing Incidence X-Ray Diffraction (GIXRD) is preferred to limit Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 83/86] 1 ChemElectroChem 2024,11, e202400136 (4 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License the penetration depth of the incident beam and avoid the signals coming from the bulk. (see Figure 2). In an interesting work, Zhang et al.[42] reported that using GIXRD, formation of pyrochlore was observed at temperatures as low as 600 °C, while it was not visible by conventional PXRD. Thus, conclusions from conventional XRD must be made with caution, particularly when the surface decomposition layer is too thin. 4.3. X-ray Photoelectron Spectroscopy (XPS) XPS is widely used to analyze the chemical composition of the LLZO surface. Using XPS, not only the Li2CO3, but also the hydrocarbons and graphitized carbons adsorbed on the surface can be observed.[45] However, as a surface characterization technique with limited probing depth, it cannot easily detect the LiOH/Li2CO3located in between the grain boundaries. As an important advantage, in-situ near ambient pressure XPS (NAPXPS) allows the real-time surface analysis with increasing temperature up to at least 500°C under the flow of different gases.[45] Depth profiling XPS is an interesting tool to probe different surface layers in the nanometric range, nevertheless, it is inconvenient for cases where the LiOH/Li2CO3layer is micrometer thick, and for powders in general.[11,50,51] Examples of the mentioned experiments are shown in Figure 3. 4.4. Solid State Nuclear Magnetic Resonance (NMR) Being sensitive to 1H, 7Li, and 6Li, solid state NMR is a powerful but underrated technique in analyzing the chemical evolution of LLZO during de(protonation). It is highly sensitive to protons, thus can show traces of HLLZO, H2O, and LiOH in the LLZO. NMR can also be used to detect the presence of Li2CO3, thanks to its long T1relaxation time (unlike mobile Li atoms of LLZO).[30] Finally, as NMR is an isotope-specific technique, the mobilities of H+and Li+in LLZO can be studied separately, and the negative impact of protonation on the Li dynamics is revealed.[30] (see figure 4). Solid state NMR is the ideal technique for analyzing the protonated phases. Moreover, it is a bulk technique and has no limitation to different forms of LLZO, i.e., powders, pellets, or thin films. 4.5. Raman Spectroscopy Raman spectroscopy is a simple yet powerful technique to investigate the structural properties of LLZO as well as its surface chemistry. CO3 2stretching vibrations shows a strong band at around 1100 cm1, allowing easy detection of Li2CO3. Moreover, using Raman mapping, the uniformity of Li2CO3 distribution can be studied..[52,53] As shown in Figure 5, besides surface mapping, Raman can be used for depth profiling and probe the gradient of CO3 2vibrations along the pellet depth. Due to the sensitivity to Li2CO3and the simplicity of the measurements, Raman is a great technique to study Li2CO3 removal. 4.6. Electrochemical Impedance Spectroscopy (EIS) Li2CO3is highly lithiophobic and forms a poor interface with Li metal.[32,36] This poor interface causes a high resistance which can be observed at low frequencies in the Nyquist plot. Despite the inability of EIS for quantified analysis and its inconvenience for powders, recording an impedance spectrum of a freshly sintered LLZO pellet or thin film as a reference can be useful in realizing the formation of Li2CO3upon storage.[11] In addition, Figure 2. Schematic illustration of setup and probing depth of left) conventional XRD and right) Grazing incidence XRD. Figure 3. a) O 1s XPS spectra of Al-LLZO recorded at 500°C under different gas environments. Reproduced from Ref. [45]. Copyright 2023 The Authors. Published by American Chemical Society, under CC-BY-4.0 license. b,c) Analysis and schematic illustration of XPS depth profiling, reproduced from Ref. [11] with permission from the Royal Society of Chemistry. Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 84/86] 1 ChemElectroChem 2024,11, e202400136 (5 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License EIS conveys an impression of the electrolyte performance. Thus, it can be used to assess how detrimental the LiOH/Li2CO3layer is from a practical point of view. For instance, if LLZO processing in dry or ambient air shows only a minor increase in the bulk and interface resistance, it will be well justified for larger-scale production. 5. Summary and Outlook Considering the reactivity of LLZO towards ambient moisture, protonation is practically unavoidable. However, it can be minimized if the relative humidity and CO2content of the storage atmosphere are precisely controlled. Moreover, special caution in the selection of solvent and the mixing time is required for the functionalization and wet processing of LLZO. Although surface properties are of higher importance in powders, the protonation of LLZO in composite polymer electrolytes is often neglected or underestimated. The surface properties of LLZO particles can greatly impact the LLZO/PEO interactions, modifying microand macro-scale properties of the electrolytes and the Li transport. The consequence of LLZO instability is not only the formation of LiOH/Li2CO3layer on its surface, but also the Li+/ H+exchange which plays a big role in the Li dynamics, especially when protonation is significant. Therefore, besides the LLZO/Li-metal interface resistance, bulk conductivity and the local ion dynamics should be carefully investigated. To fill the gap in the literature, detailed studies on the deprotonation of LLZO are particularly needed for powders and thin films. In addition, mentioning details of LLZO properties (form, doping, relative density) and storage atmosphere (relative humidity, CO2 content, and exposure time) is necessary for easier comparison of the results. Acknowledgements As a part of the DESTINY PhD programme, this publication is acknowledged by funding from the European Union’s Horizon2020 research and innovation programme under the Marie Skłodowska Curie Actions COFUND e Grant Agreement No: 945357. Conflict of Interests The authors declare no conflict of interest. Figure 4. Different NMR experiments showing the impact of heat treatment of LLZO in a) disappearance of protonated phases, b) removal of Li2CO3with long relaxation time c) improving the 7Li local dynamics in LLZO powders. Reproduced from Ref. [30] with permission from the Royal Society of Chemistry. Figure 5. Top) Surface Raman Mapping for Ta-LLZO pellets aged in humid air, dry air and after polishing. Bottom) Raman mapping of Ta-LLZO pellet along its depth after exposure in humid air. Reprint from Ref. [52] with permission. Copyright 2017 The American Ceramic society. Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 85/86] 1 ChemElectroChem 2024,11, e202400136 (6 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Data Availability Statement The data that support the findings of this study are available from The authors. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the authors with the permission of The authors. Keywords: LLZO Garnet ·Lithium Carbonate ·Protonation ·air instability ·Solid electrolytes [1] T. Thompson, S. Yu, L. Williams, R. D. Schmidt, R. Garcia-Mendez, J. Wolfenstine, J. L. Allen, E. Kioupakis, D. J. Siegel, J. Sakamoto, ACS Energy Lett. 2017,2, 462. [2] J.-F. Wu, W. K. Pang, V. K. Peterson, L. Wei, X. Guo, ACS Appl. Mater. Interfaces 2017,9, 12461. [3] X. Tao, L. Yang, J. Liu, Z. Zang, P. Zeng, C. Zou, L. Yi, X. Chen, X. Liu, X. Wang, J. Alloys Compd. 2023,937, 168380. [4] F. Han, Y. Zhu, X. He, Y. Mo, C. Wang, Adv. Energy Mater. 2016,6, 1501590. [5] H. Duan, H. Zheng, Y. Zhou, B. Xu, H. Liu, Solid State Ionics 2018,318, 45. [6] G. Charrad, S. Pradeilles, R. Berthelot, P.-L. Taberna, P. Simon, P. Rozier, Energy Technol. 2023,11, 2300234. [7] Y. Li, J.-T. Han, C.-A. Wang, H. Xie, J. B. Goodenough, J. Mater. Chem. 2012,22, 15357. [8] E. J. Cussen, Chem. Commun. 2006, 412. [9] C. Galven, J.-L. Fourquet, M.-P. Crosnier-Lopez, F. Le Berre, Chem. Mater. 2011,23, 1892. [10] C. Galven, J. Dittmer, E. Suard, F. Le Berre, M.-P. Crosnier-Lopez, Chem. Mater. 2012,24, 3335. [11] A. Sharafi, S. Yu, M. Naguib, M. Lee, C. Ma, H. M. Meyer, J. Nanda, M. Chi, D. J. Siegel, J. Sakamoto, J. Mater. Chem. A 2017,5, 13475. [12] H. Huo, X. Li, Y. Sun, X. Lin, K. Doyle-Davis, J. Liang, X. Gao, R. Li, H. Huang, X. Guo, X. Sun, Nano Energy 2020,73, 104836. [13] X. Liu, Y. Chen, Z. D. Hood, C. Ma, S. Yu, A. Sharafi, H. Wang, K. An, J. Sakamoto, D. J. Siegel, Y. Cheng, N. H. Jalarvo, M. Chi, Energy Environ. Sci. 2019,12, 945. [14] G. Larraz, A. Orera, M. L. Sanjuán, J. Mater. Chem. A 2013,1, 11419. [15] L. Dhivya, K. Karthik, S. Ramakumar, R. Murugan, RSC Adv. 2015,5, 96042. [16] S. Toda, K. Ishiguro, Y. Shimonishi, A. Hirano, Y. Takeda, O. Yamamoto, N. Imanishi, Solid State Ionics 2013,233, 102. [17] R. Kun, F. Langer, M. Delle Piane, S. Ohno, W. G. Zeier, M. Gockeln, L. Colombi Ciacchi, M. Busse, I. Fekete, ACS Appl. Mater. Interfaces 2018, 10, 37188. [18] Z. F. Yow, Y. L. Oh, W. Gu, R. P. Rao, S. Adams, Solid State Ionics 2016, 292, 122. [19] J. Sastre, T.-Y. Lin, A. N. Filippin, A. Priebe, E. Avancini, J. Michler, A. N. Tiwari, Y. E. Romanyuk, S. Buecheler, ACS Appl. Energ. Mater. 2019,2, 8511. [20] X. Chen, Z. Guan, F. Chu, Z. Xue, F. Wu, Y. Yu, InfoMat 2022,4, e12248. [21] K. J. Kim, M. Balaish, M. Wadaguchi, L. Kong, J. L. M. Rupp, Adv. Energy Mater. 2021,11, 2002689. [22] X. Zeng, A. J. Martinolich, K. A. See, K. T. Faber, J. Energy Storage 2020, 27, 101128. [23] L. H. Abrha, T. T. Hagos, Y. Nikodimos, H. K. Bezabh, G. B. Berhe, T. M. Hagos, C.-J. Huang, W. A. Tegegne, S.-K. Jiang, H. H. Weldeyohannes, S.- H. Wu, W.-N. Su, B. J. Hwang, ACS Appl. Mater. Interfaces 2020,12, 25709. [24] W. Xia, B. Xu, H. Duan, Y. Guo, H. Kang, H. Li, H. Liu, ACS Appl. Mater. Interfaces 2016,8, 5335. [25] W. Jeong, S. S. Park, J. Yun, H. R. Shin, J. Moon, J.-W. Lee, Energy Storage Mater. 2023,54, 543. [26] Y. Li, B. Xu, H. Xu, H. Duan, X. Lü, S. Xin, W. Zhou, L. Xue, G. Fu, A. Manthiram, J. B. Goodenough, Angew. Chem. Int. Ed. 2017,56, 753. [27] M. Rosen, R. Ye, M. Mann, S. Lobe, M. Finsterbusch, O. Guillon, D. Fattakhova-Rohlfing, J. Mater. Chem. A 2021,9, 4831. [28] H. Yamada, T. Ito, S. P. Kammampata, V. Thangadurai, ACS Appl. Mater. Interfaces 2020,12, 36119. [29] Y. Zhu, J. G. Connell, S. Tepavcevic, P. Zapol, R. Garcia-Mendez, N. J. Taylor, J. Sakamoto, B. J. Ingram, L. A. Curtiss, J. W. Freeland, D. D. Fong, N. M. Markovic, Adv. Energy Mater. 2019,9, 1803440. [30] P. Ghorbanzade, A. Pesce, K. Gómez, G. Accardo, S. Devaraj, P. LópezAranguren, J. M. L. del Amo, J. Mater. Chem. A 2023. [31] J. Guo, R. Floyd, S. Lowum, J.-P. Maria, T. Herisson de Beauvoir, J.-H. Seo, C. A. Randall, Annu. Rev. Mater. Res. 2019,49, 275. [32] A. Sharafi, E. Kazyak, A. L. Davis, S. Yu, T. Thompson, D. J. Siegel, N. P. Dasgupta, J. Sakamoto, Chem. Mater. 2017,29, 7961. [33] P. Ghorbanzade, G. Accardo, K. Gomez, P. López-Aranguren, S. Devaraj, C. M. Costa, S. Lanceros-Mendez, J. M. López del Amo, Mater. Today Energy 2023,38, 101448. [34] L. Cheng, E. J. Crumlin, W. Chen, R. Qiao, H. Hou, S. F. Lux, V. Zorba, R. Russo, R. Kostecki, Z. Liu, K. Persson, W. Yang, J. Cabana, T. Richardson, G. Chen, M. Doeff, Phys. Chem. Chem. Phys. 2014,16, 18294. [35] Y. Ruan, Y. Lu, Y. Li, C. Zheng, J. Su, J. Jin, T. Xiu, Z. Song, M. E. Badding, Z. Wen, Adv. Funct. Mater. 2021,31, 2007815. [36] H. Huo, Y. Chen, N. Zhao, X. Lin, J. Luo, X. Yang, Y. Liu, X. Guo, X. Sun, Nano Energy 2019,61, 119. [37] W. M. Seong, Y. Kim, A. Manthiram, Chem. Mater. 2020,32, 9479. [38] K. Kanamura, in Encyclopedia of Electrochemical Power Sources (Ed.: J. Garche), Elsevier, Amsterdam, 2009, pp. 27–39. [39] A. Orue, M. Arrese-Igor, R. Cid, X. Júdez, N. Gómez, J. M. L. del Amo, W. Manalastas, M. Srinivasan, C. Rojviriya, M. Armand, F. Aguesse, P. LópezAranguren, J. Mater. Chem. A 2022,10, 2352. [40] M. Jia, Z. Bi, C. Shi, N. Zhao, X. Guo, J. Power Sources 2021,486, 229363. [41] S. Kim, J.-S. Kim, L. Miara, Y. Wang, S.-K. Jung, S. Y. Park, Z. Song, H. Kim, M. Badding, J. Chang, V. Roev, G. Yoon, R. Kim, J.-H. Kim, K. Yoon, D. Im, K. Kang, Nat. Commun. 2022,13, 1883. [42] H. Zhang, G. Paggiaro, F. Okur, J. Huwiler, C. Cancellieri, L. P. H. Jeurgens, D. Chernyshov, W. van Beek, M. V. Kovalenko, K. V. Kravchyk, ACS Appl. Energ. Mater. 2023,6, 6972. [43] J. Cai, B. Polzin, L. Fan, L. Yin, Y. Liang, X. Li, Q. Liu, S. E. Trask, Y. Liu, Y. Ren, X. Meng, Z. Chen, Mater. Today Energy 2021,20, 100669. [44] C. Deviannapoorani, S. Ramakumar, N. Janani, R. Murugan, Solid State Ionics 2015,283, 123. [45] S. Vema, F. N. Sayed, S. Nagendran, B. Karagoz, C. Sternemann, M. Paulus, G. Held, C. P. Grey, ACS Energy Lett. 2023,8, 3476. [46] Y. Kim, I. Waluyo, A. Hunt, B. Yildiz, Adv. Energy Mater. 2022,12, 2102741. [47] W. M. Jones, J. B. Tapia, R. R. Tuttle, M. M. Reynolds, Langmuir 2020,36, 3903. [48] J. Cai, D. Xu, Z. Dong, X. Yu, Y. Yang, S. W. Banks, A. V. Bridgwater, Renewable Sustainable Energy Rev. 2018,82, 2705. [49] L. Fallarino, U. N. Chishti, A. Pesce, G. Accardo, A. Rafique, M. CasasCabanas, P. López-Aranguren, Chem. Commun. 2023,59, 12346. [50] J. Leng, H. Wang, H. Liang, Z. Xiao, S. Wang, Z. Zhang, Z. Tang, ACS Appl. Energ. Mater. 2022,5, 5108. [51] R. Dubey, J. Sastre, C. Cancellieri, F. Okur, A. Forster, L. Pompizii, A. Priebe, Y. E. Romanyuk, L. P. H. Jeurgens, M. V. Kovalenko, K. V. Kravchyk, Adv. Energy Mater. 2021,11, 2102086. [52] W. Xia, B. Xu, H. Duan, X. Tang, Y. Guo, H. Kang, H. Li, H. Liu, J. Am. Ceram. Soc. 2017,100, 2832. [53] Y. Li, X. Chen, A. Dolocan, Z. Cui, S. Xin, L. Xue, H. Xu, K. Park, J. B. Goodenough, J. Am. Chem. Soc. 2018,140, 6448. Manuscript received: February 9, 2024 Revised manuscript received: March 14, 2024 Version of record online: April 15, 2024 Wiley VCH Freitag, 24.05.2024 2411 / 347887 [S. 86/86] 1 ChemElectroChem 2024,11, e202400136 (7 of 7) © 2024 The Authors. ChemElectroChem published by Wiley-VCH GmbH ChemElectroChem Concept doi.org/10.1002/celc.202400136 21960216, 2024, 11, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/celc.202400136 by Readcube (Labtiva Inc.), Wiley Online Library on [24/06/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Investigating the role of interphases in composite electrolytes by solid-state NMR 80 7.4 Publication #4: Title: Transport Properties and Local Ions Dynamics in LATP-Based Hybrid Solid Electrolytes Authors: Nicola Boaretto, Pedram Ghorbanzade, Haritz Perez-Furundarena, Leire Meabe, Juan Miguel López del Amo, Isuru E. Gunathilaka, Maria Forsyth, Jörg Schuhmacher, Andreas Roters, Sergey Krachkovskiy, Abdelbast Guerfi, Michel Armand, and María Martinez-Ibañez Journal: Small Journal impact factor: 13 Journal quartile: Q1 (Materials Science) RESEARCH ARTICLE www.small-journal.com Transport Properties and Local Ions Dynamics in LATP-Based Hybrid Solid Electrolytes Nicola Boaretto,* Pedram Ghorbanzade, Haritz Perez-Furundarena, Leire Meabe, Juan Miguel López del Amo, Isuru E. Gunathilaka, Maria Forsyth, Jörg Schuhmacher, Andreas Roters, Sergey Krachkovskiy, Abdelbast Guerfi, Michel Armand, and María Martinez-Ibañez* Hybrid solid electrolytes (HSEs), namely mixtures of polymer and inorganic electrolytes, have supposedly improved properties with respect to inorganic and polymer electrolytes. In practice, HSEs often show ionic conductivity below expectations, as the high interface resistance limits the contribution of inorganic electrolyte particles to the charge transport process. In this study, the transport properties of a series of HSEs containing Li(1+x)AlxTi(2–x)(PO4)3 (LATP) as Li+-conducting filler are analyzed. The occurrence of Li+exchange across the two phases is proved by isotope exchange experiment, coupled with 6Li/7Li nuclear magnetic resonance (NMR), and by 2D 6Li exchange spectroscopy (EXSY), which gives a time constant for Li+exchange of about 50 ms at 60 °C. Electrochemical impedance spectroscopy (EIS) distinguishes a short-range and a long-range conductivity, the latter decreasing with LATP concentration. LATP particles contribute to the overall conductivity only at high temperatures and at high LATP concentrations. Pulsed field gradient (PFG)-NMR suggests a selective decrease of the anions’ diffusivity at high temperatures, translating into a marginal increase of the Li+transference number. Although the transport properties are only marginally affected, addition of moderate amounts of LATP to polymer electrolytes enhances their mechanical properties, thus improving the plating/stripping performance and processability. N. Boaretto, P. Ghorbanzade, H. Perez-Furundarena, L. Meabe, J. M. López del Amo, M. Armand, M. Martinez-Ibañez Centre for Cooperative Research on Alternative Energies CIC energiGUNE Basque Research and Technology Alliance (BRTA) Alava Technology Park Albert Einstein 48, Vitoria-Gasteiz 01510, Spain E-mail: [email protected]; [email protected] The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/smll.202305769 © 2023 The Authors. Small published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. DOI: 10.1002/smll.202305769 1. Introduction Lithium-ion batteries (LIBs) are the principal element for electrochemical energy storage in portable electronics and automotive applications. LIBs are characterized by high energy density (260 Wh kg−1, 700 Wh L−1at cell level),[1] which is, however, still insufficient to meet the targets of driving range in full battery electric vehicles (>300 Wh kg−1,>800 Wh L−1at cell level).[2] Higher values of energy density can be obtained by substituting the conventional graphite anode with a high-capacity lithium metal anode (372 mAh g−1vs 3860 mAh g−1for graphite and lithium metal, respectively). This, in turn, leads to serious safety concerns, owing to the growth of lithium dendrites, which can cause short circuits and thermal runaway, and are particularly dangerous in combination with common flammable liquid battery electrolytes.[3] A possible solution to this issue involves the adoption of nonflammable solid-state electrolytes (SSEs), such as inorganic ceramic P. Ghorbanzade University of Basque Country (UPV/EHU) Barrio Sarriena, s/n, Leioa 48940, Spain P. Ghorbanzade ALISTORE-European Research Institute, CNRS, Hub de l’Energie Amiens 80039, France I. E. Gunathilaka, M. Forsyth Institute for Frontier Materials Deakin University Geelong, VIC 3217, Australia M. Forsyth Ikerbasque Basque Foundation for Science Bilbao 48013, Spain M. Forsyth POLYMAT University of the Basque Country UPV/EHU Joxe Mari Korta Center Donostia-San Sebastián 200018, Spain Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (1 of 16) www.advancedsciencenews.com www.small-journal.com Figure 5. a) Interface capacitance, divided by the electrodes area. Empty dots are multiplied for the volume fraction of LATP; b) time constants t1(full dots) and 𝜏2(empty dots), calculated by multiplying Cint for R1andR2, respectively. For HSE-05, 𝜎eff is largely superimposed to 𝜎tot, indicating negligible LATP contribution in the whole temperature range. With HSE-10, 𝜎eff is close to 𝜎tot at low temperatures, but 𝜎tot increases faster than 𝜎eff with the temperature, indicating increasing LATP contribution at high temperatures to the long-range conductivity. In HSE-20, 𝜎eff is lower than 𝜎tot in the whole temperature range, and the spread between the two values increases with temperature. The lower value of 𝜎eff, with respect to 𝜎tot, is possibly caused by an underestimation of 𝜎PE,20, but the increasing spread is clearly caused by the decrease of the interface resistance. Altogether, the analysis reveals that LATP contribution to the conductivity increases with the temperature, and that this effect becomes more intense with increasing LATP content. This can be easily observed by comparing the conductivity ratio 𝜌=𝜎tot/𝜎loc (Figure 4c). The conductivity ratio is practically constant in HSE05 at 𝜌≈0.9, but it increases with the temperature in both HSE10 and HSE-20. The increase is also more evident in HSE-20, for which 𝜌increases from 0.6 at 25 °C to 0.8 at 70 °C. Further details on the interface resistance can be obtained through the analysis of R2, and of the parallel capacitance (Cint). As noted above, R2 increases with LATP concentration (Figure 4c). More specifically, R2 is roughly proportional to the volumefractionofLATP(Figure4d).ThedependenceofR2 with the temperature is of Arrhenius type, and the activation energy is ≈0.55 eV, constant with LATP concentration, thus very close to the activation energy of the inorganic phase. R2 can be estimated by combining Equations (4) and (6). The resulting formula is as follows R2MG =9𝜎LATPfLATP 2𝜎PE (1−fLATP) 1 𝜎LATP +2𝜎PE +2(𝜎LATP −𝜎PE)fLATP L A(7) The values of R2MG, calculated though Equation (7), are also shown in Figure 4d. The calculated values are roughly close to the experimental ones (except for HSE-20, for which 𝜎eff LATP >0), but the Maxwell equation predicts a VTF dependence of R2on the temperature, whereas the experimental dependence, as noted previously, is clearly of Arrhenius type, with activation energy of ≈0.55 eV, constant with LATP concentration. The area-specific resistance R2sp calculated by normalizing R2 on the LATP volume fraction and multiplying for the electrodes area, ranges from 270Ωcm2inHSE-05 to450 Ωcm2in HSE-20,at 25 °C(FigureS5, Supporting Information). At 60 °C, R2sp varies between 20 and 30 Ωcm2. Although R2sp is only indirectly related to the interface resistance between the LATP particles and the polymer matrix, it is quite interesting that the values of R2sp are close to the values of interface resistance previously reported for NASICON-based multilayered model systems.[18,43,55] The interface capacitance, Cint, was calculated from the values of CPE2, R1, and R2 (see Figure S4 in the Supporting Information), with the following equation[56] Cint =(1 R1+1 R2)(𝛼−1 𝛼)Q21 𝛼(8) where Q2and𝛼are the pseudocapacitance and exponent of the CPE2 element in Figure S4 (Supporting Information), respectively. The values of Cint divided by the electrodes area, are reported in Figure 5a. In all samples, Cint is below 1 μFcm −2,confirming that this capacitance is related to an internal interface and not to an electrodes interface. Furthermore, Cint decreases with the LATP concentration, thus further confirming that it is related to the LATP interface. Indeed, by dividing Cint for the LATP volume fraction, the values of Cint fall into a single master curve (Figure 5a), resembling the behavior of R2. Interestingly, Cint increases with temperature at T>40 °C. This may be related to the decreasing interface resistance. By multiplying Cint for R1andR2, two time constants can be obtained, namely 𝜏1and 𝜏2, respectively. The first one is the characteristic charging time of the double layer at the LATP interface, whereas the origin of 𝜏2is more ambiguous, although it is possibly related to the conductivity of LATP and to the interface resistance. The values of 𝜏1and 𝜏2are shown in Figure 5b. 𝜏1 varies between 10−5and 10−6s, and it decreases with increasing LATP concentration, owing to the decreasing values of Cint.𝜏1 decreases initially with increasing temperature, but it is approximately constant at T>40 °C. This behavior, which is related to the increasing values of Cint, is possibly caused by the decrease of the interface resistance and by the increasing rate of Li+transfer across the interface. On the contrary, 𝜏2is practically constant and generally lower than 𝜏1, varying from ≈2∙10−6s at room temperature to ≈5∙10−7sat70°C. The constant values of 𝜏2suggest that R2 is strongly correlated with Cint. Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (8 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 6. Self-diffusion coefficients and transference numbers by PFG-NMR: diffusion coefficients of a) 7Li and b) 19Ffrom20to80°C; c) comparison of the diffusion coefficients at 60 °C; d) ionic conductivity calculated from the diffusion coefficients with the Nernst–Einstein equation; e) Haven ratios between 30 and 70 °C; f) lithium transference numbers, calculated from the diffusion coefficients of 7Li and 19F, between 20 and 70 °C. The red asterisk indicates the transference number calculated by potentiostatic polarization method. 2.2. Solid-State NMR Characterization Information on the individual mobility of anions and cations was obtained by PFG-NMR. The self-diffusion coefficients of 7Li (DLi) and 19F(DF) between 20 and 80 °C are depicted in Figure 6a,b, respectively. Both diffusion coefficients increase with temperature between 10−12 to 10−11 m2s−1, but the values of DFare generally higher than those of DLi, as could be expected for polymer electrolytes, since ethylene oxide units strongly coordinate with Li+ cations. The activation energies of the diffusion coefficients were determined by fitting with the Arrhenius equation. The activation energy is ≈0.3 eV for DLi, practically constant with LATP, and in the case of DFis comprised between 0.4 and 0.3 eV, decreasing with LATP concentration. Interestingly, DFdecreases with LATP concentration, especially at high temperature (Figure 6c), indicating that the mobility of TFSI−anions is negatively affected by the presence of LATP. On the contrary, DLiis almost constant, suggesting that the diffusivity of Li+cations in the LATP particles is close to the diffusivity in the polymer matrix. It must be noted that the fitting of the 7Li signals, to retrieve the diffusion coefficients, was carried out with a single exponential function. For comparison, the fitting was additionally performed with two exponentials, which resulted in quite similar diffusion coefficients. However, since the quality of the fitting was only marginally improved by fitting with a two-exponential function, fitting with one exponential was ultimately preferred. The difficulty in differentiating the diffusion coefficients of the two phases is also related to the timescale of thePFG-NMRexperiment. Thediffusion delay of25 ms issignificantly larger than the characteristic time for ions accumulation at the LATP surface which, as observed by EIS, is in the order of 10−6 s. The diffusion length, given a diffusion coefficient of 5∙10−12 m2 s−1at 60 °C, is slightly lower than 1 μm. This is larger than the average particles radius (≈0.7 μm), thus sufficient for most Li+ within the LATP particles to experience the particles boundary. Although this is not necessarily related to the characteristic time of the Li+exchange across the interface, the results of EIS experiments suggest that a timescale of 25 ms is sufficient to give Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (9 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com a picture of the long-range transport properties of the HSEs. To confirm this assumption, PFG-NMR experiments were repeated at longer diffusion delay of 200 ms, on the samples HSE-00 and HSE-20. The resulting diffusion coefficients (Figure S6, Supporting Information) are close to those measured at 25 ms, except the fluorine diffusion coefficient of fluorine in HSE-20, which is slightly higher than the one measured at 25 ms. In the following analysis, for sake of coherence, the values obtained at 25 ms will be used. The diffusion coefficients of 7Li, as determined by PFGNMR, should be understood as the long-range diffusivity in the HSEs, accounting also for the contribution of the interface resistance. Consequently, the results of the PFG-NMR experiment can be compared with the long-range ionic conductivity, by applying the Nernst-Einstein equation. This allows determining whether there are any ion pairing effects caused by the filler. To calculate the ionic conductivity from the diffusion coefficients, the concentration of charge carriers is needed. The effective charge carrier concentration is not known, but two limiting scenarios can be conceived. In the first case, the total ions concentration is used, includingalso the Li+inthe LATP particles.[53] In thesecondcase, only the salt concentration is considered. In the former case, the Nernst-Einstein equation assumes the following form[57] 𝜎NMR =e2 kBT(CLiDLi +CFDF)(9) With ebeing the elementary charge, kBthe Boltzmann constant, Tthe temperature, and CLi and CFthe concentrations of Li+and TFSI−, respectively. The values of 𝜎NMR , for HSE-00, are close to the experimental values of 𝜎tot (compare Figure 4b and Figure 6d), with a Haven ratio HR(that is, the ratio between 𝜎NMR and 𝜎tot) ranging between 1.2 and 1.3 (Figure 6e). On the contrary, the conductivity of the LATP-containing samples is overestimated, approaching 𝜎NMR of HSE-00 and thus close to the values of 𝜎loc. This is expected as the concentration of cations CLi in Equation (9) considers also the Li+in the inorganic phase, which at low temperatures is essentially confined owing to the high interface resistance. Nonetheless, the Haven ratio decreases with the temperature (Figure 6e), thus confirming that Li+plays increasingly a role in the conduction process. For comparison, we can calculate again the conductivity, by considering only the salt concentration. This is done by substituting CLi and CFwith Csalt in Equation (9). In this case, 𝜎NMR ≈𝜎tot also for the LATPcontaining samples, and the Haven ratio is close to 1 in all samples (Figure S7, Supporting Information). If the values of HSE00 are taken as a reference, lower values of HRare observed, in LATP-containing samples, only at T≥60 °C. In this case, we can assume that the concentration of Li+is possibly underestimated. Despite this, the use of the salt concentration alone, in the Nernst–Einstein equation, gives a good approximation for the calculation of the total conductivity. The Nernst–Einstein equation can also be applied reversely to calculate the salt diffusion coefficient from the ionic conductivity. This can be compared with the average salt diffusion coefficient by PFG-NMR. Again, there is a good agreement between the results of PFG-NMR and of the conductivity measurements, with diffusivity values ranging between 4∙10−12 m2s−1and 8∙10−12 m2s−1and decreasing between HSE-00 and HSE-20 (Figure S8, Supporting Information). The values are close to those obtained by interrupted current method (5–6∙10−12 m2s−1at 60 °C), although in this case no effect of LATP is observed. Altogether, it appears that, at least at high temperatures, the observed decrease of ionic conductivity is caused by a decrease of the anions’ diffusivity. This is attributed to the anions-specific blocking effect exercised by the LATP particles. On the contrary, Li+diffusion is unhindered by LATP particles, at least at high temperatures. The larger Haven ratio at low temperatures is attributed to the high interface resistance blocking the transfer of Li+. Overall, this results in a slight increase of the Li+transference number at high temperatures: The latter was calculated from the diffusivity values[58] t+=DLi DLi +DF (10) It must be noted that the transference number calculated in this way assumes an equal concentration of cations and anions, and thus is possibly underestimated, as the Li+concentration in the LATP particles is not considered. However, as noted above, the use of the salt concentration is a good approximation when considering the long-range charge transport. At room temperature, the four HSEs have similar transference numbers of 0.38 (Figure 6f). However, the values start to diverge at ≈50 °C. At 60 °C, the sample HSE-20 has transference number of over 0.41, whereas HSE-00 has transference number slightly below 0.3. It must be noted that this difference is quite low. Indeed, transference number T+, determined by potentiostatic polarization method, is equal to T+=0.35 in all samples (chronoamperometric profiles and impedance spectra are shown in Figure S9 in the Supporting Information). The value is close to the one determined by PFG-NMR, but no effect is observed due to LATP. Altogether, the experimental results do not confirm a significant increase of the transference number due to LATP. On the other hand, the close values obtained by potentiostatic polarization method and PFG-NMR suggest the absence of ion pairing effects. This is confirmed by solid-state NMR measurements (see later discussion in this section) and is expected owing to the high dissociation degree of LiTFSI.[59] Finally, the lithium transference numbers, obtained by PFGNMR, were combined with the total ionic conductivity, to calculate the cationic conductivity 𝜎Li+(Figure 7). Owing to the increase of the transference number, the cationic conductivity remains constant up to 10% LATP. At 60 °C, the cationic conductivity is ≈1.6∙10−4Scm −1, up to 10% LATP. At 20% LATP, a slight decrease is observed, with 𝜎Li+=1.1∙10−4Scm −1. To summarize, at moderate LATP content, since the decrease of the ionic conductivity is mostly caused by the blocking of anions, it has almost no effect on the cationic conductivity. This suggests that the electrochemical performances should not be compromised by the decrease of the total conductivity, at least at moderate contents of LATP. This, combined with the results of the mechanical measurements, suggests that the overall cycling performance should be enhanced by the addition of moderate amounts of LATP. Electrochemical characterization suggests that, at 60 °C, LATP particles participate in the long-range Li+transport process. To assess this hypothesis, a 6Li–7Li isotope exchange experiment was carried out. This type of experiment, which combines electrochemical 6Li–7Li isotope exchange with NMR, has been used for Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (10 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 7. a) Li+ionic conductivity as a function of temperature, and b) at 60 °C, compared to the total ionic conductivity. probing the participation of active fillers in the conduction process of HSEs.[14,40,41] Isotope substitution of 7Li by 6Li in a HSE-10 electrolyte was forced by cycling galvanostatically the electrolyte between 99% enriched 6Li metal electrodes, as described in the Experimental Section. The possible 7Li–6Li exchange in LATP and the polymer matrix was subsequently assessed by 7Li and 6Li 1D NMR measurements, performed ex situ on the recovered electrolyte. From this experiment, the Li-ions stemming from the LATP particles and from the Lithium salt can be distinguished from the Li-ions coming from the metal foil by solid-state NMR, by recording samples of HSE-10 before and after plating versus 6Li metal. 7Li and 6Li NMR spectra were collected on the pristine HSE-10 membrane, as a reference (Figure 8a,b, respectively). The 7Li spectrum of HSE-10 before plating (HES-10-Pristine in Figure 8a) shows two highly overlapped signals at around −1.3 and −1.1 ppm. From the deconvolution of the 7Li spectra of HSE-10 and its comparison with the one of HSE-00 (Figure S10, Supporting Information), these two signals can be assigned to the Li+ions in the polymer matrix and the LATP, respectively. The comparison of the NMR spectra before and after plating, Figure 8a, clearly shows that, after the 6Li plating, the intensity of the 7Li signal of the HSE-10 sample did almost disappear. The signal reduction clearly involves Li NMR signals from both the LATP and the polymer phases. Parallelly, the 6Li signal increases drastically, giving an intense asymmetric peak, as expected after homogeneous 7Li to 6Li exchange in both the polymer and LATP phases. This result clearly demonstrates that Li+ ions at the polymer and LATP phases must be interchanging. The spectra shown in Figure 8b also shows that the 6Li NMR signal linewidth increases after the plating. This phenomenon, which could be attributed to the increased heterogeneity of the membrane,[44] makes the deconvolution complicated. However, since the alteration of 7Li signal intensities is homogeneous, it can be concluded that the Li+ions in both the polymer and ceramic phases are mobile and are accessible for ion transport. This suggests the occurrence of Li+ion exchange between the polymer matrix and the LATP particles. The results also show that this exchange is not limited to the Li atoms on the surface of LATP, but rather all the Li atoms in the bulk of LATP can be substituted by the ones from the metal foil. 6Li EXSY experiments have been carried out with the pristine HSE-20 sample to unequivocally demonstrate ion exchange in the ceramic-polymer interface, the same way as in our previous works regarding LLZO-PEO electrolytes.[14,47] In these spectra, the diagonal peaks (identical chemical shifts in both dimensions) correspond to regular signals of the identified components of the sample, while the off-diagonal responses (cross-peaks) represent magnetization transfer between the components either via chemical exchange or through dipolar interaction between neighboring spins separated in space by less than 5 Å. Taking into account that natural abundance of 6Li is less than 8%, the probability of observing several 6Li isotopes at a close distance from each other is negligible. Moreover, the low dipolar moment of 6Li would make spin diffusion through dipolar interactions even more unlikely. It means that the appearance of cross-peaks in our spectra clearly demonstrates the presence of dynamic physical exchange processes of Li+ions between the two phases in the HSE. Intensity of cross-peaks depends on a mixing time – time delay in the EXSY experiment, during which magnetization transfer occurs (Figure 8c and Figure S11, Supporting Information). By running the experiment multiple times with different mixing time, one can quantify the exchange rate. In the analyzed sample at 20 °C, the time constant of the exchange is about 300 ms, with the plateau observed at around 1 s, while at 60 °C the constant decreases to 50 ms and the plateau is reached at 400 ms (Figure 8d). Beside probing the participation of the LATP particles in the long-range charge transport, solid-state NMR can be used also to investigate the local ions dynamics, which may be affected by chemical interactions between the ions and polymer matrix and the filler particles. For instance, possible chemical interactions of the anions with the particles surface may give rise to a decrease of the anion’s mobility, prompting an increase of the cation transference number and of the oxidative stability in HSEs.[29] To study the possible TFSI−aggregation at the LATP surface, 19FNMRexperiments were conducted and T1relaxation times were investigated. The longitudinal relaxation times of the nuclear magnetizations in NMR depend strongly on the local mobilities and time fluctuations of the chemical environments. In this line, the attachment of anions to the surface of a filler would result in different anion environments with distinct relaxation times and line shapes.[60] Since in the 19F 1D spectra of all samples (HES-00, HSE-10, and HSE-20) only single lines were observed, Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (11 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 8. Solid-state NMR characterization of LATP-based HSEs. a) 7Li NMR spectra of HSE-10, pristine (black line) and after electrochemically driven isotope exchange (red line); b) 6Li NMR spectra of HSE-10, pristine (black line) and after isotope exchange (red line); c) 6Li 2D EXSY spectra of HSE-20 at 60 °C, at mixing times of 10 ms (red contour) and 600 ms (blue contour); d) normalized intensity of the off-diagonal peaks in the 6Li 2D EXSY spectra of HSE-20 at 20 and 60 °C, as a function of the mixing time; e) 19F NMR saturation recovery experiments of HSEs with different LATP concentrations. Fitting line was obtained considering a single exponential; f) 19F linewidth at variable temperature, with and without LATP. complementary T1saturation-recovery experiments were conducted to examine the possible overlapping of resonances. As shown in Figure 8e, the data could be perfectly fitted in all cases with a single exponential term, and similar 19F relaxation times (T1=0.76 s) were obtained, indicating no anion immobilization upon the addition of LATP. If this were the case, different 19F NMR resonances and/or multiexponential behavior in T1would be expected. To compare the anion dynamics in the HSE-00 and HSE-10 in more detail, variable temperature NMR experiments were performed and the linewidths (FWHM) as a function of temperature were plotted in Figure 8f. The motional narrowing curve shows similar behavior of the 19F signals of the TFSI−anions between these two hybrid electrolyte membranes. For both HSE00 and HSE-10, the motional narrowing is completed at around 20 °C, and the linewidth of the rigid lattice is at least 32 kHz, which would result in jump rates (1/𝜏NMR) above 2∙105s−1at the temperature of inflection point (−31 °C). A similar experiment to follow 7Li local dynamics by NMR was not feasible due to the overlapping between 7Li signals from the polymer and LATP phases. The results agree with the T1relaxation time measurements, and further prove that filler incorporation keeps the anion local dynamics rather unaffected. It must be noted that this result is not in conflict with the previous results of impedance spectroscopy and PFG-NMR experiments. Indeed, impedance spectroscopy showed that the local molar conductivity is not affected by the introduction of LATP, which agrees well with the results of the NMR linewidth experiments. On the other hand, as noted above, the diffusion coefficients determined by PFG-NMR are related to the long-range transport, and thus are partially decoupled from the local mobility, owing to the presence of the high interface resistance. Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (12 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com Figure 9. Plating/stripping profiles of the four HSEs, at 60 °C. Tests were performed at fixed capacity (1 mAh cm−2) and increasing current density. The current density is indicated as C-rate, where 1C (or C/1) is 1 mA cm−2. The red asterisk indicates the minimum C-rate at which voltage instability or short circuit is observed. a) HSE-00; b) HSE-05; c) HSE-10; d) HSE-20. 2.3. Plating/Stripping Test Plating/stripping tests were carried out to test the resistance of the four electrolytes against dendrites growth, and thus to verify the combined effect of LATP on the mechanical and transport properties of the electrolytes. The tests were performed at 60°C, with fixed capacity of 1 mAh cm−2and at increasing current density, to determine the maximum current density at which stable plating/stripping can be achieved. Thus, this test gives an indication of the limiting current density at which the hybrid electrolytes can be cycled. The plating/stripping profiles are shown in Figure 9. HSE-00 shows stable cycling only up to 0.1 mA cm−2(C/10), whereas voltage instabilities (soft shorts) and a clear short circuit are observed at 0.2 mA cm−2 (C/5) and 0.33 mA cm−2(C/3), respectively. HSE-05 and HSE10 show better plating/stripping performance, with stable cycling at 0.2 mA cm−2(C/5) and voltage instabilities starting from 0.33 mA cm−2(C/3). In both cases, no clear short circuit is observed. HSE-20, on the contrary, shows a short circuit already at 0.2 mA cm−2(C/5). The enhanced plating/stripping performance obtained with HSE-05 and HSE-10 results from the combination of improved mechanical properties and retained transport properties, whereas the decrease of the performance at HSE-20 is attributed to the decrease of the ionic conductivity at high LATP concentrations. These results possibly confirm the combined results obtained by mechanical, conductivity and PFG-NMR measurements, which is that addition of moderate amounts of LATP can enhance the cycling performance of the polymer electrolytes. 3. Conclusions The transport properties and local ions dynamics of LATPcontaining HSEs were analyzed by means of impedance spectroscopy and solid-state NMR. The analysis of the admittance/impedance spectra allowed discerning two conductivities, a local conductivity, at timescales so short that no ions accumulation nor Li+transfer across the interphase boundary occur, and a long-range conductivity, at longer timescales, accounting for the contribution of interface resistance and Li+transfer across the particles interface. The former corresponds to the effective conductivity of the HSE, modeled through the Maxwell-Garnett mixing rule, without the contribution of the interface resistance. Overall, this local conductivity is negligibly affected by LATP, owing to the high intrinsic conductivity of the amorphous and plasticized polymer matrix. On the contrary, the long-range conductivity decreases with increasing volume fraction of LATP, owing to the high interface resistance for Li+transfer between the two phases. Up to 10 vol% LATP, the drop in the long-range ionic conductivity can be modeled with the Maxwell-Garnett mixing rule, by considering LATP particles as insulating. Nonetheless, the contribution of LATP particles to the overall charge transport process increases with LATP content and with increasing temperature. Indeed, the additional polarization responsible for the conductivity drop decreases with increasing temperature, especially at high LATP content. In other words, at high temperatures the long-range conductivity approaches the local conductivity. Ions’ diffusivity was studied by PFG-NMR. Results show a slight decrease of the anions’ diffusion coefficient with increasing LATP concentration, whereas the cations’ diffusivity, at least Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (13 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com at high temperatures, is unaffected. This was attributed to the blocking effect of LATP particles towards the anions. Altogether, this results in a moderate increase of the Li+transference number, which compensates for the decrease of the total conductivity up to 10 vol% LATP. Nonetheless, this increase of transference number is small and was not confirmed by potentiostatic polarization method. The occurrence of Li+exchange between the two phases was further proved by 6Li 2D EXSY NMR, which provided an exchange time constant of 50 ms at 60 °C and 20 vol% LATP, and by isotope exchange experiment, coupled with 6Li/7Li NMR, showing almost complete isotope substitution in the electrolyte after galvanostatic cycling between 6Li-enriched Li discs. Finally, relaxometry experiments and variable temperature NMR experiments indicate negligible variation in the local ions’ dynamics, confirming that the local motion is unhindered by the LATP particles and that no anions aggregation occurs on the LATP surface. Regarding a possible application of the HSEs in Li metal batteries, results of PFG-NMR and conductivity measurements suggest that the transport properties are not affected by addition of moderate amounts of LATP (up to 10% LATP). Mechanical measurements, on the other hand, indicate a clear enhancement of the toughness and Young modulus of the membranes with LATP concentration. The combined results suggest that HSEs with low concentration of LATP (up to 10 vol%) should have improved processability and resistance to dendrites growth. Indeed, this last conclusion was later confirmed by plating/stripping tests in Li||Li symmetric cells. 4. Experimental Section Hybrid Solid Electrolytes Preparation:LATP-based HSEs were prepared by mixing micrometer-sized LATP particles (Schott) with an ethylene oxidebased polymer matrix. LiTFSI (solvionic, 99.9%) was used as lithium salt. The composition of the polymer phase was kept constant, with a mole ratio of ethylene oxide units to Li+(EO:Li+)fixedatEO:Li +=16, whereas the content of LATP was varied between 0 and 20 vol% (0, 5, 10 and 20 vol%, corresponding to 0, 12, 22, and 39 wt% of the total HSE mass). The polymer electrolyte matrix was composed of an ethylene oxide-propylene oxide copolymer (p(EO-PO), Mw≈800 000), poly(ethylene glycol)dimethyl ether (PEGDME, Mn≈500, Sigma Aldrich) as plasticizer, and poly(ethylene glycol)diacrylate (PEGDA, Mn≈700, Sigma Aldrich) as cross-linker. PEGDME with molecular weight of 500 g mol−1was chosen, over lower molecular weight analogues, to enhance the thermal stability of the electrolytes. All HSE components were dried under high vacuum before use, and all preparation steps were carried out in an argon-filled glove box. First, the polymer matrix components, LiTFSI, and the LATP were mixed overnight in acetonitrile (ACN) under vigorous stirring. In a typical preparation, 24 mL of ACN, 1.26 g of p(EO-PO) (25.3 wt% with respect to the total mass of the polymer electrolyte phase), 1.38 g of LiTFSI (27.7 wt%), 0.9 g of PEGDA (18.1 wt%), and 1.44 g of PEGDME (28.9 wt%) were used. Second, AIBN (≈20 mg) was added to the solution. The solution was then milled with a Micro Pulverisette 7 premium planetary (Fritsch). Hermetically sealed 45 mL zirconium flasks were used, which were loaded and sealed in argon atmosphere with 5 mm diameter ZrO2beads (70 g). The total wet milling time was 20 min at 250 rpm, interrupted by a 10 min pause to avoid overheating of the system. The homogeneous solutions were cast on a Mylar film, with a target dry thickness of 140 μm. The ACN was first evaporated for 3 h at room temperature. Then, the temperature was increased to 70°C for one hour (half vacuum was applied) for the cross-linking step, forming an interpenetrating acrylic polymer network. Prior to characterization, the membranes were dried under vacuum overnight. The casting was carried out on a minicoater (TOC sheen) in a glove box, under Ar atmosphere, and the drying steps were carried out in a heated glovebox antechamber. The membranes were finally stored in the same glovebox for further use. The membranes density was measured by weighing discs with diameter of 16 mm and by measuring their thickness with a micrometer. Characterization Methods:The morphology of the HSEs was studied by SEM, with a FEI Quanta 250. Cross sections were prepared by cutting the membranes at room temperature in an argon glovebox. Images were collected with a voltage acceleration of 10 keV and with a backscattered electron detector (BSED) and with a secondary electron detector. Thermogravimetric analysis (TGA) was carried out under argon (60 mL min−1), from room temperature up to 600 °C, at a heating rate of 10 °Cmin −1, with a TGA 209 F1 Libra (Netzsch). Differential scanning calorimetry measurements (DSC) were performed with a DSC 2500 differential calorimeter (TA Instruments). The measurements were carried out by placing samples of 5–10 mg in sealed aluminum pans under argon atmosphere, in the temperature range between −80 and 100 °C, and with a heating rate of 2°Cmin −1. Each sample was cycled twice between −80 and 100 °C, and the second heating scan was used for the analysis. The mechanical properties of the hybrid electrolytes were characterized by tensile test, with a single column universal testing machine (Instrom, 34SC-5). The static load cell (100 N 2519 Series S-beam) had a displacement speed of 20 cm min−1. The samples had an approximate length of 4 cm and a width of 10 mm, whereas the separation between the tensile clamps was of ≈10 mm. Ionic conductivity measurements were performed with a Solartron 1260A Impedance/Gain-Phase Analyzer, in the frequency range between 32 MHz and 1 Hz (20 points per decade), with a signal amplitude of 20 mV, and in the temperature range between 25 and 70 °C(with10°C step). The measurements were carried out by placing the membranes in coin cells CR2032, with three stainless steel plates of 0.5 mm thickness. The temperature was controlled with a Binder KB23 Cooling incubator. The ionic conductivity was calculated with the following formula 𝜎i=1 Ri L A(11) where 𝜎iis the local or long-range conductivity, Riis the correspondent resistance, as determined by EIS, Ais the electrodes surface area, and Lis the membrane thickness. The latter was measured with a digital micrometer, after the experiment. The measurement was repeated on three different cells for each composition, and the values of resistance and conductivity used in the analysis are the average values of the three measurements. Lithium transference number T+and salt restricted diffusion coefficient Dres were measured in Li||Li coin cells, on a Biologic VMP3 potentiostat, at 60 °C. The lithium transference number was determined by potentiostatic polarization method, by combining a chronoamperometry with the measurement of the impedance spectra, collected before and after the chronoamperometry. For the chronoamperometry, a constant voltage of ±10 mV was applied for a duration of 20 min, and the resulting current was registered with a frequency of 10 points s−1during the first minute, and of one point per second during the rest of the chronoamperometry. The impedance spectra were collected in the frequency range between 1 MHz and 100 mHz and with potential amplitude of 10 mV. The experiment was repeated on three cells per sample and six times on each cell, by alternating positive and negative potentiostatic polarization. The cells were allowed to relax for one hour after each chronoamperometric step, and for 10 min after the EIS. The transference number was then calculated using the usual formalism[61] T+ 1=ISS I0(ΔV−Rint,0I0) (ΔV−Rint,SSI0)(12) where I0and ISS are the initial and steady-state current, respectively, ΔV is the applied potential and Rint,0 and Rint,SS are the interface resistances Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (14 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com before and after the chronoamperometric step, respectively. The values of T+were further controlled with the following equation[62] T+ 2=Rb,0 (ΔV∕ISS −Rint,SS)(13) where Rb,0 is the bulk resistance before the chronoamperometry. The two calculation methods gave similar results, and the transference numbers reported in the results section are the ones calculated with Equation (12). The salt restricted diffusion coefficient, Dres, was determined using the following equation[63] Dres =− BL2 𝜋2(14) where Bis the slope of the natural logarithm of the cell voltage versus time, during the rest step following the chronoamperometry, and Lis the membrane thickness. Magic angle spinning NMR spectroscopy (MAS NMR) experiments were performed on hybrid solid electrolytes using a Bruker Avance III 500 spectrometer equipped with a 2.5 mm probe. The MAS frequency was set to 20 kHz for all measurements except variable temperature studies that were carried out in static samples. 7Li and 6Li chemical shifts were referenced to a 1 m LiCl water solution and 19F was referenced indirectly to solid LiF resonating at −204 ppm. 1D experiments were recorded using single excitation pulses (2.4, 3, and 3.7 μs) and relaxation delays of 10, 3, and8sfor7Li, 6Li, and 19F respectively. These spectra were recorded for qualitative purposes. Saturation recovery experiments were performed to obtain the 19F relaxation times with the relaxation delay varying from 1 ms to 180 s. Variable temperature experiments were conducted in the temperature range of −80 to 110 °C in a static probe, and the 19F 1D experiments were recorded using a single 3 μs excitation pulse with a single scan. In the Li isotope substitution experiment, a HSE-10 electrolyte with a natural 6Li abundance (7.5%) was used as electrolyte in a symmetric Li||Li cell, with 99% enriched 6Li metal. A constant current density of 32 μAcm −2 was applied to the cell, for 20 h, to force isotope substitution in the electrolyte. The charge passed was calculated to be sufficient to displace all the Li atoms contained in the electrolyte. After the plating, the cell was opened in the glovebox and the electrolyte membrane was carefully removed and packed into a rotor to perform the 7Li and 6Li 1D NMR measurements. Pulsed-field gradient nuclear magnetic resonance (PFG-NMR) was used to determine the diffusion coefficients of 7Li and 19F for all the electrolytes at variable temperature (ranging from 80 to 20 °C) using a Bruker Avance III 300 MHz wide bore NMR spectrometer equipped with a 5 mm diff50 probe. Stimulated Echo was used for all nuclei diffusion measurements. Typical diffusion time for both nuclei was 25 ms, gradient pulse duration was 2 ms. The maximum field strength was 7.05 T on a log scale. 7Li and 19F PFG-NMR experiments were repeated for 0 and 20 vol% LATP containing samples to verify whether the obtained diffusion coefficients depend on diffusion time. Bruker Avance NEO 500 MHz wide bore NMR spectrometer equipped with double resonance (7Li/19F) 8 mm Diff 50 probe has been used for these experiments. Diffusion time was set to 200 ms, while gradient delay, and the maximum field strength were kept the same as in previous experiments: 2 ms and 7.05 T, respectively. The same Bruker Avance NEO 500 MHz wide bore NMR spectrometer but with 4 mm MAS probe (maximum spinning speed of 15 kHz) has been used to carry out 6Li–6Li 2D EXSY experiments for 20 vol% LATP sample at 20 and 60 °C. A standard three-pulse sequence with mixing times from 10 ms to 1.6 s and a relaxation delay of 80 s was applied for quantitative characterization of cation exchange between LATP particles and polymer matrix. 16 scans were collected for each of the 64 data points in indirect dimension with the total acquisition time of 23 h per spectrum. Lithium plating/stripping tests were carried out in symmetrical Li||Li coin cells, at 60 °C. Before each measurement, the cells were kept at 70 °C for 12 h, to ensure stabilization of the lithium interface. The cells were cycled galvanostatically with a fixed plating/stripping capacity of 1mAhcm −2, and with progressively increasing current densities of 0.05, 0.1, 0.2, 0.33, 0.5, 1 mA cm−2(C/10, C/5, C/3, C/2, C/1), plus one final control cycle at 0.1 mA cm−2(C/10). One cycle was performed at each current density. The measurements were performed in a Neware battery tester. Supporting Information Supporting Information is available from the Wiley Online Library or from the author. Acknowledgements The authors are grateful to the European Commission for the support of the work performed within SAFELiMOVE. The project has received funding from the European Union’s Horizon 2020 research and innovation programme under Grant agreement no. 875189. The information and views set out on this paper do not necessarily reflect the official opinion of the European Commission. Neither the European Union institutions and bodies nor any person acting on their behalf, may be held responsible for the use of the information contained therein. P.G. as a part of the DESTINY PhD programme acknowledges funding from the European Union’s Horizon2020 research and innovation programme under the Marie Skłodowska-Curie Actions COFUND – Grant Agreement No: 945357. The authors acknowledge Maria Etxebarria and Maria Campos Terron for collecting the cross-section SEM images. Conflict of Interest The authors declare no conflict of interest. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. Keywords composite polymer electrolytes, conduction mechanism, hybrid solid electrolytes, LATP, NASICON, solid-state batteries, transport properties Received: July 10, 2023 Revised: September 18, 2023 Published online: October 24, 2023 [1] R. Schmuch, R. Wagner, G. Hörpel, T. Placke, M. Winter, Nat. Energy 2018,3, 267. [2] D. Andre, H. Hain, P. Lamp, F. Maglia, B. Stiaszny, J. Mater. Chem. A 2017,5, 17174. [3] Y. Shen, Y. Zhang, S. Han, J. Wang, Z. Peng, L. Chen, Joule 2018,2, 1674. [4] M.Balaish,J.C.Gonzalez-Rosillo,K.J.Kim,Y.Zhu,Z.D.Hood,J.L. M. Rupp, Nat. Energy 2021,6, 227. [5] M. A. Cabañero Martínez, N. Boaretto, A. J. Naylor, F. Alcaide, G. D. Salian, F. Palombarini, E. Ayerbe, M. Borras, M. Casas-Cabanas, M. Casas-Cabanas, Adv. Energy Mater. 2022,12, 2201264. [6] N. Boaretto, I. Garbayo, S. Valiyaveettil-Sobhanraj, A. Quintela, C. Li, M. Casas-Cabanas, F. Aguesse, J. Power Sources 2021,502, 229919. [7] V. Di Noto, S. Lavina, G. A. Giffin, E. Negro, B. Scrosati, Electrochim. Acta 2011,57,4. Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (15 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License www.advancedsciencenews.com www.small-journal.com [8] J. Przyluski, M. Siekierski, W. Wieczorek, Electrochim. Acta 1995,40, 2101. [9] W. Wieczorek, Z. Florjanczyk, J. R. Stevens, Electrochim. Acta 1995, 40, 2251. [10] B. Scrosati, F. Croce, G. B. Appetecchi, L. Persi, Nature 1998,394, 456. [11] D. Golodnitsky, G. Ardel, E. Strauss, E. Peled, Y. Lareah, Y. Rosenberg, J. Electrochem. Soc. 1997,144, 3484. [12] Y. Dai, Y. Wang, S. G. Greenbaum, S. A. Bajue, D. Golodnitsky, G. Ardel, E. Strauss, E. Peled, Electrochim. Acta 1998,43, 1557. [13] M. Keller, A. Varzi, S. Passerini, J. Power Sources 2018,392, 206. [14] J. Zagórski, J. M. López Del Amo, M. J. Cordill, F. Aguesse, L. Buannic, A. Llordés, ACS Appl. Energy Mater. 2019,2, 1734. [15] Z. Li, H.-M. Huang, J.-K. Zhu, J.-F. Wu, H. Yang, L. Wei, X. Guo, ACS Appl. Mater. Interfaces 2019,11, 784. [16] D. Brogioli, F. Langer, R. Kun, F. La Mantia, ACS Appl. Mater. Interfaces 2019,11, 11999. [17] M. R. Bonilla, F. A. García Daza, P. Ranque, F. Aguesse, J. Carrasco, E. Akhmatskaya, ACS Appl. Mater. Interfaces 2021,13, 30653. [18] E. Trevisanello, T. Ates, S. Passerini, F. H. Richter, J. Janek, J. Electrochem. Soc. 2022,169, 110547. [19] P. Ranque, J. Zagórski, G. Accardo, A. Orue Mendizabal, J. M. López Del Amo, N. Boaretto, M. Martinez-Ibañez, H. Arrou-Vignod, F. Aguesse, M. Armand, S. Devaraj, Inorganics 2022,10, 81. [20] J. Zheng, P. Wang, H. Liu, Y.-Y. Hu, ACS Appl. Energy Mater. 2019,2, 1452. [21] M. Liu, S. Zhang, E. R. H. Van Eck, C. Wang, S. Ganapathy, M. Wagemaker, Nat. Nanotechnol. 2022,17, 959. [22] J. Pcharski, Solid State Ion 1988,28–30, 979. [23] K. Nairn, Solid State Ion. 1999,121, 115. [24] Y. Inda, T. Katoh, M. Baba, J. Power Sources 2007,174, 741. [25] H. Zhai, P. Xu, M. Ning, Q. Cheng, J. Mandal, Y. Yang, Nano Lett. 2017,17, 3182. [26] C. Wang, Y. Yang, X. Liu, H. Zhong, H. Xu, Z. Xu, H. Shao, F. Ding, ACS Appl. Mater. Interfaces 2017,9, 13694. [27] D. Li, L. Chen, T. Wang, L.-Z. Fan, ACS Appl. Mater. Interfaces 2018, 10, 7069. [28] M. Liu, Z. Cheng, S. Ganapathy, C. Wang, L. A. Haverkate, M. Tulodziecki, S. Unnikrishnan, M. Wagemaker, ACS Energy Lett. 2019, 4, 2336. [29] N. Wu, P.-H. Chien, Y. Li, A. Dolocan, H. Xu, B. Xu, N. S. Grundish, H. Jin, Y.-Y. Hu, J. B. Goodenough, J. Am. Chem. Soc. 2020,142, 2497. [30] Y. Liang, N. Chen, F. Li, R. Chen, ACS Appl. Mater. Interfaces 2022,14, 47822. [31] Y. Jin, X. Zong, X. Zhang, Z. Jia, H. Xie, Y. Xiong, Energy Storage Mater. 2022,49, 433. [32] G. Foran, A. Mery, M. Bertrand, S. Rousselot, D. Lepage, D. AyméPerrot, M. Dollé, ACS Appl. Mater. Interfaces 2022,14, 43226. [33] A. Paolella, W. Zhu, D. Campanella, S. Kaboli, Z. Feng, A. Vijh, Curr. Opin. Electrochem. 2022,36, 101108. [34] L. Meabe, J. Zagórski, D. Mecerreyes, F. Aguesse, A. Llordes, ChemPhysChem 2022,23, 202200296. [35] F. J. Simon, M. Hanauer, A. Henss, F. H. Richter, J. Janek, ACS Appl. Mater. Interfaces 2019,11, 42186. [36] N. Riphaus, B. Stiaszny, H. Beyer, S. Indris, H. A. Gasteiger, S. J. Sedlmaier, J. Electrochem. Soc. 2019,166, A975. [37] W. Wang, E. Yi, A. J. Fici, R. M. Laine, J. Kieffer, J. Phys. Chem. C 2017, 121, 2563. [38] X. Ban, W. Zhang, N. Chen, C. Sun, J. Phys. Chem. C 2018,122, 9852. [39] Z. Yao, K. Zhu, X. Li, J. Zhang, J. Chen, J. Wang, K. Yan, J. Liu, Electrochim. Acta 2022,404, 139769. [40] Y.-J. Wang, Y.i P an, J Polym Sci B Polym Phys 2005,43, 743. [41] I. V. Belova, G. E. Murch, J. Phys. Chem. Solids 2005,66, 722. [42] S. Yang, Z. Zhang, L. Shen, P. Chen, Z. Gu, M. Chang, Y. Zhao, H. He, X. Yao, J. Power Sources 2022,518, 230756. [43] J. A. Isaac, L. R. Mangani, D. Devaux, R. Bouchet, ACS Appl. Mater. Interfaces 2022,14, 13158. [44] J. Zheng, M. Tang, Y.-Y. Hu, Angew. Chem., Int. Ed. 2016,55, 12538. [45] T. Yang, J. Zheng, Q. Cheng, Y.-Y. Hu, C. K. Chan, ACS Appl. Mater. Interfaces 2017,9, 21773. [46] J. Zheng, Y.-Y. Hu, ACS Appl. Mater. Interfaces 2018,10, 4113. [47] P. Ranque, J. Zagórski, S. Devaraj, F. Aguesse, J. M. López Del Amo, J. Mater. Chem. A 2021,9, 17812. [48] V. Di Noto, J. Phys. Chem. B 2002,106, 11139. [49] P. Barai, K. Higa, V. Srinivasan, Phys. Chem. Chem. Phys. PCCP 2017, 19, 20493. [50] N. Boaretto, L. Meabe, M. Martinez-Ibañez, M. Armand, H. Zhang, J. Electrochem. Soc. 2020,167, 070524. [51] S. Kitajima, F. Bertasi, K. Vezzù, E. Negro, Y. Tominaga, V. Di Noto, Phys.Chem.Chem.Phys.2013,15, 16626. [52] A. Lasia, Electrochemical Impedance Spectroscopy and Its Applications, Springer, New York 2014. [53] J. Jamnik, J. R. Kalnin, E. A. Kotomin, J. Maier, Phys. Chem. Chem. Phys. PCCP 2006,8, 1310. [54] R. F. Samsinger, M. Letz, J. Schuhmacher, M. Schneider, A. Roters, D. Kienemund, H. Maune, A. Kwade, J. Electrochem. Soc. 2020,167, 140510. [55] W. E. Tenhaeff, K. A. Perry, N. J. Dudney, J. Electrochem. Soc. 2012, 159, A2118. [56] G. J. Brug, A. L. G. Van Den Eeden, M. Sluyters-Rehbach, J. H. Sluyters, J. Electroanal. Chem. 1984,176, 275. [57] N. A. Stolwijk, J. Kösters, M. Wiencierz, M. Schönhoff, Electrochim. Acta 2013,102, 451. [58] G. Foran, N. Verdier, D. Lepage, C. Malveau, N. Dupré, M. Dollé, Polymers 2021,13, 1207. [59] N. Boaretto, C. Joost, M. Seyfried, K. Vezzù, V. Di Noto, J. Power Sources 2016,325, 427. [60] A. Mejía, S. Devaraj, J. Guzmán, J. M. Lopez Del Amo, N. García, T. Rojo, M. Armand, P. Tiemblo, J. Power Sources 2016,306, 772. [61] J. Evans, C. A. Vincent, P. G. Bruce, Polymer 1987,28, 2324. [62] M. Watanabe, Solid State Ion 1988,28–30, 911. [63] Y. Ma, M. Doyle, T. F. Fuller, M. M. Doeff, L. C. De Jonghe, J. Newman, J. Electrochem. Soc. 1995,142, 1859. Small 2024,20, 2305769 © 2023 The Authors. Small published by Wiley-VCH GmbH 2305769 (16 of 16) 16136829, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202305769 by Universidad Del Pais Vasco, Wiley Online Library on [10/03/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Investigating the role of interphases in composite electrolytes by solid-state NMR 97 7.5 Publication #5: Title: Unveiling the Reactivity and the Li-Ion Exchange at the PEO-Li6PS5Cl Interphase: Insights from Solid-State NMR Authors: Pedram Ghorbanzade, Arianna Pesce, Michel Armand, Kerman Gómez, Shanmukaraj Devaraj, Pedro López-Aranguren, and Juan Miguel López del Amo Journal: Small Structures Journal impact factor: 13.9 Journal quartile: Q1 (Materials Science, Multidisciplinary) quantify different decomposition products in the interphase, it is evident that the total amount of these products is significantly higher in the composite made with Br-doped LPSCl. Comparing the 7 Li– 7 Li 2D EXSY NMR experiments in these two samples indicated that the chemical species in the interphase restrict the Li ion exchange between the two phases, and do not play an intermediate role in the exchange. In general, the interphase species are found to be undesirable. Thus, to limit the interfacial reaction one should target either the Li affinity of the polymer by decreasing its complexing ability, or the disorder of the LPSCl crystal structure. The addition of side chains to the PEO or reducing the number and content of dopants are among possible approaches to minimize the interfacial reaction. Considering the observed Li exchange between the two phases, we believe that if the interfacial reaction is prevented and the activation energy for the Li exchange is reduced, a continuous Li pathway through both phases could be formed, and the objective of combining the high ionic conductivity of the LPSCl with the processability of polymers could be achieved. 4. Experimental Section Solid State NMR Experiments: All NMR experiments were recorded using a Brucker Avance III 500 MHz spectrometer. Samples were packed inside 2.5 mm rotors and magic angle spinning speed was fixed at 20 kHz in all experiments. The rotors were spun in a nitrogen atmosphere, to avoid possible reactions with moisture during the experiment. The 7 Li, 6 Li, and 31 P spectra were recorded using one pulse excitation and pulse lengths of 2.4, 3, and 2.5 μs for 7 Li, 6 Li, and 31 P respectively. For the 19 F and 1 H experiments, rotor-synchronized Hahn-echo pulse sequences were used with the 90° pulse length set to 3 and 2.1 μs for 19 F and 1 H, respectively. The mixing times for the 7 Li– 7 Li and 6 Li– 6 Li EXSY experiments were set to 400 ms and the contact time of 1000 and 1200 μs were respectively selected for the 1 H– 7 Li and 1 H– 31 P CP HETCOR experiments. The 1 H, 19 F, 31 P, 7 Li, and 6 Li 1D spectra were recorded with recycle delays of 5, 5, 30, 60, and 120 s, respectively (unless mentioned otherwise), while the recycle delay was set to 3 s for 1 H– 31 P and 1 H– 7 Li CP experiments, and 10 s for the 19 F– 7 Li HETCOR. The signal deconvolutions of 6 Li 1D spectra were performed using DM-Fit program. [46] Powder X-ray Diffraction: X-ray diffraction of the synthetized powder was carried out at BL04-MSPD beamline at ALBA synchrotron. The powders were packed in a 0.7 mm diameter borosilicate glass capillary. The pattern was recorded at 15 keV. LPSClBr and LPSCl Synthesis: Stoichiometric amounts of Li 2 S, P 2 S 5 , LiCl, and LiBr were ball-milled in a Pulverisette 7 for 20 h to obtain LPSClBr according to reaction 1. 4.2 Li2SþP2S5þ2 LiCl þ0.8 LiBr !2Li 5.6PS4.6ClBr0.4 (1) The composition and the crystal structure of the LPSClBr was confirmed by synchrotron XRD experiments as shown in Figure S8, Supporting Information. Commercial Li 6 PS 5 Cl was purchased from NEI corporation and was used with no further modification. Composite Polymer Electrolyte Preparation: Composite electrolytes were prepared by mixing 70 wt% poly(ethylene-oxide) (PEO, 1 106 g mol 1 , Sigma Aldrich), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Solvionic) (EO:Li =20), and 30 wt% argyrodite (LPSCl or LPSClBr) powders in airtight grinding bowls inside an Ar-filled glovebox. The ball-milling was conducted using a Planetary Micro Mill Pulverisette 7 outside the glovebox for 30 min (six intervals of 5 min at 500 rpm with 10 min pauses in between to prevent overheating). After being transferred back to the glovebox, the mixture was pressed into 6 mm pellets using a hydraulic press applying 1.5 tons of weight. Each pellet was placed between two plastic films, sealed inside a coffee bag, and subsequently hot-pressed at 80 °C for 2 min outside the glovebox. The coffee bags were then transferred and opened inside the glovebox. The PEO-argyrodite self-standing films appeared visually homogeneous and had a thickness of ≈70 μm. This method avoids uncertainties regarding possible interface modification by solvent while ensuring sufficient contact between the two phases to form the interface. Both LPSCl and LPSClBr composites were prepared following the same procedure to allow for fair comparison. Additionally, the reproducibility of the composites was verified by solidstate NMR experiments performed on different batches of composites. Digital photographs of the different processing steps of the PEO-LPSCl composite are shown in Figure S9, Supporting Information. Solid State NMR Sample Preparation: The composite membranes were cut into thin ribbon-like shapes and packed inside 2.5 mm NMR rotors. After the first set of measurements, the same rotor was placed inside a glovebox antechamber under dynamic vacuum and heated at 70 °C overnight to promote more reactions at the interphase. After cooling down to room temperature, the sample was transferred back to the spectrometer for further NMR experiments. For the time evolution experiments, the same rotor was stored inside the glovebox after the first set of measurements and the tests were repeated after 10 days. Repeating the experiments on the same samples after heating or a period of time allows us to clearly observe signal evolutions and eliminates other parameters that could cause uncertainty in the conclusions. Supporting Information Supporting Information is available from the Wiley Online Library or from the author. Acknowledgements As a part of the DESTINY PhD programme, this publication is acknowledged by funding from the European Union’s Horizon2020 research and innovation programme under the Marie Skłodowska-Curie Actions COFUND (Grant agreement #945357). The authors also acknowledge Rosalia Cid for fruitful discussion. The XRD experiments were performed at BL04-MSPD. beamline at ALBA Synchrotron with the collaboration of ALBA. This project has received funding from the European Union through the H2020 program under Grant agreement number 875028 (SUBLIME Project). Conflict of Interest The authors declare no conflict of interest. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. Keywords argyrodite ion conductors, batteries, composite polymer electrolytes, interphases, solid state NMR Received: March 26, 2024 Revised: July 12, 2024 Published online: July 28, 2024 www.advancedsciencenews.com www.small-structures.com Small Struct. 2024,5, 2400139 2400139 (7 of 8) © 2024 The Author(s). Small Structures published by Wiley-VCH GmbH 26884062, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/sstr.202400139 by Readcube (Labtiva Inc.), Wiley Online Library on [09/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License [1] S. Li, S.-Q. Zhang, L. Shen, Q. Liu, J.-B. Ma, W. Lv, Y.-B. He, Q.-H. Yang, Adv. Sci. 2020,7, 1903088. [2] T. Yang, C. Wang, W. Zhang, Y. Xia, H. Huang, Y. Gan, X. He, X. Xia, X. Tao, J. Zhang, J. Energy Chem. 2023,84, 189. [3] W. Liu, S. W. Lee, D. Lin, F. Shi, S. Wang, A. D. Sendek, Y. Cui, Nat. Energy 2017,2, 17035. [4] H. Liang, L. Wang, A. Wang, Y. Song, Y. Wu, Y. Yang, X. He, NanoMicro Lett. 2023,15, 42. [5] S. Boulineau, M. Courty, J.-M. Tarascon, V. Viallet, Solid State Ionics 2012,221,1. [6] Y. Liu, H. Su, Y. Zhong, X. Wang, X. Xia, C. Gu, J. Tu, Adv. Funct. Mater. 2022,32, 2207978. [7] P. Adeli, J. D. Bazak, K. H. Park, I. Kochetkov, A. Huq, G. R. Goward, L. F. Nazar, Angew. Chem., Int. Ed. 2019,58, 8681. [8] A. Tron, R. Hamid, N. Zhang, A. Paolella, P. Wulfert-Holzmann, V. Kolotygin, P. López-Aranguren, A. Beutl, J. Energy Storage 2023, 66, 107480. [9] X. Chen, Z. Guan, F. Chu, Z. Xue, F. Wu, Y. Yu, InfoMat 2022,4, e12248. [10] A. Golov, J. Carrasco, ACS Appl. Mater. Interfaces 2021,13, 43734. [11] Y. Zhou, C. Doerrer, J. Kasemchainan, P. G. Bruce, M. Pasta, L. J. Hardwick, Batteries Supercaps 2020,3, 647. [12] A. Orue Mendizabal, M. Cheddadi, A. Tron, A. Beutl, P. López-Aranguren, ACS Appl. Energy Mater. 2023,6, 11030. [13] F. J. Simon, M. Hanauer, A. Henss, F. H. Richter, J. Janek, ACS Appl. Mater. Interfaces 2019,11, 42186. [14] F. J. Simon, M. Hanauer, F. H. Richter, J. Janek, ACS Appl. Mater. Interfaces 2020,12, 11713. [15] H. Huo, M. Jiang, B. Mogwitz, J. Sann, Y. Yusim, T.-T. Zuo, Y. Moryson, P. Minnmann, F. H. Richter, C. Veer Singh, J. Janek, Angew. Chem. 2023,135, e202218044. [16] J. Zhang, C. Zheng, J. Lou, Y. Xia, C. Liang, H. Huang, Y. Gan, X. Tao, W. Zhang, J. Power Sources 2019,412, 78. [17] J. Keeler, Understanding NMR Spectroscopy, 2nd ed., John Wiley and Sons, Chichester, UK 2010. [18] A. Pines, M. G. Gibby, J. S. Waugh, J. Chem. Phys. 1972,56, 1776. [19] X. Feng, P.-H. Chien, Y. Wang, S. Patel, P. Wang, H. Liu, M. Immediato-Scuotto, Y.-Y. Hu, Energy Storage Mater. 2020, 30, 67. [20] D. H. S. Tan, E. A. Wu, H. Nguyen, Z. Chen, M. A. T. Marple, J.-M. Doux, X. Wang, H. Yang, A. Banerjee, Y. S. Meng, ACS Energy Lett. 2019,4, 2418. [21] F. P. Nkosi, M. Valvo, J. Mindemark, N. A. Dzulkurnain, G. Hernández, A. Mahun, S. Abbrent, J. Brus, L. Kobera, K. Edström, ACS Appl. Energy Mater. 2021,4, 2531. [22] J. Zagórski, J. M. López del Amo, M. J. Cordill, F. Aguesse, L. Buannic, A. Llordés, ACS Appl. Energy Mater. 2019,2, 1734. [23] D. Spencer Jolly, D. L. R. Melvin, I. D. R. Stephens, R. H. Brugge, S. D. Pu, J. Bu, Z. Ning, G. O. Hartley, P. Adamson, P. S. Grant, A. Aguadero, P. G. Bruce, Inorganics 2022,10, 60. [24] O. Pecher, J. Carretero-González, K. J. Griffith, C. P. Grey, Chem. Mater. 2017,29, 213. [25] D. S. Hall, U. Werner-Zwanziger, J. R. Dahn, J. Electrochem. Soc. 2017, 164, A2171. [26] A. L. Michan, M. Leskes, C. P. Grey, Chem. Mater. 2016,28, 385. [27] W. Yu, Z. Yu, Y. Cui, Z. Bao, ACS Energy Lett. 2022,7, 3270. [28] M. Armand, Adv. Mater. 1990,2, 278. [29] Z. Xue, D. He, X. Xie, J. Mater. Chem. A 2015,3, 19218. [30] H. Stöffler, T. Zinkevich, M. Yavuz, A. Senyshyn, J. Kulisch, P. Hartmann, T. Adermann, S. Randau, F. H. Richter, J. Janek, S. Indris, H. Ehrenberg, J. Phys. Chem. C 2018,122, 15954. [31] E. C. Self, P.-H. Chien, L. F. O’Donnell, D. Morales, J. Liu, T. Brahmbhatt, S. Greenbaum, J. Nanda, Mater. Today Phys. 2021, 21, 100478. [32] G. Foran, N. Verdier, D. Lepage, C. Malveau, N. Dupré, M. Dollé, Polymers 2021,13, 1207. [33] H.-J. Deiseroth, S.-T. Kong, H. Eckert, J. Vannahme, C. Reiner, T. Zaiß, M. Schlosser, Angew. Chem. Int. Ed. 2008,47, 755. [34] I. Hanghofer, M. Brinek, S. L. Eisbacher, B. Bitschnau, M. Volck, V. Hennige, I. Hanzu, D. Rettenwander, H. M. R. Wilkening, Phys. Chem. Chem. Phys. 2019,21, 8489. [35] E. C. Self, Z. D. Hood, T. Brahmbhatt, F. M. Delnick, H. M. I. Meyer, G. Yang, J. L. M. Rupp, J. Nanda, Chem. Mater. 2020,32, 8789. [36] R. Koerver, F. Walther, I. Aygün, J. Sann, C. Dietrich, W. G. Zeier, J. Janek, J. Mater. Chem. A 2017,5, 22750. [37] N. Riphaus, B. Stiaszny, H. Beyer, S. Indris, H. A. Gasteiger, S. J. Sedlmaier, J. Electrochem. Soc. 2019,166, A975. [38] M. Sumita, Y. Tanaka, T. Ohno, J. Phys. Chem. C 2017,121, 9698. [39] T. K. Schwietert, V. A. Arszelewska, C. Wang, C. Yu, A. Vasileiadis, N. J. J. de Klerk, J. Hageman, T. Hupfer, I. Kerkamm, Y. Xu, E. van der Maas, E. M. Kelder, S. Ganapathy, M. Wagemaker, Nat. Mater. 2020,19, 428. [40] M. Murakami, K. Shimoda, S. Shiotani, A. Mitsui, K. Ohara, Y. Onodera, H. Arai, Y. Uchimoto, Z. Ogumi, J. Phys. Chem. C 2015,119, 24248. [41] M. Ernst, B. Meier, in Solid State NMR of Polymers (Eds: I. Ando, T. Asakura), Elsevier Science, Amsterdam 1998. [42] P. Ghorbanzade, G. Accardo, K. Gomez, P. López-Aranguren, S. Devaraj, C. M. Costa, S. Lanceros-Mendez, J. M. López del Amo, Mater. Today Energy 2023,38, 101448. [43] P. Ranque, J. Zagórski, S. Devaraj, F. Aguesse, J. M. L. del Amo, J. Mater. Chem. A 2021,9, 17812. [44] R. Chometon, M. Deschamps, R. Dugas, E. Quemin, B. Hennequart, M. Deschamps, J.-M. Tarascon, C. Laberty-Robert, ACS Appl. Mater. Interfaces 2023,15, 58794. [45] M. Liu, S. Zhang, E. R. H. van Eck, C. Wang, S. Ganapathy, M. Wagemaker, Nat. Nanotechnol. 2022,17, 959. [46] D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B. Alonso, J.-O. Durand, B. Bujoli, Z. Gan, G. Hoatson, Magn. Reson. Chem. 2002,40, 70. www.advancedsciencenews.com www.small-structures.com Small Struct. 2024,5, 2400139 2400139 (8 of 8) © 2024 The Author(s). Small Structures published by Wiley-VCH GmbH 26884062, 2024, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/sstr.202400139 by Readcube (Labtiva Inc.), Wiley Online Library on [09/10/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Investigating the role of interphases in composite electrolytes by solid-state NMR 106 7.6 Publication #6: * The manuscript is submitted to the Journal of Materials Chemistry A and is under review. Title: Insights into the Compatibility and Interphases of LPSCl Argyrodites and Zr-based Halide Electrolytes for All-Solid-State Batteries Authors: Pedram Ghorbanzade, Arianna Pesce, Kerman Gómez, Pedro López-Aranguren, and Juan Miguel López del Amo Journal: --- Journal impact factor: --- Journal quartile: --- ARTICLE Please do not adjust margins Please do not adjust margins Received 00th January 20xx, Accepted 00th January 20xx DOI: 10.1039/x0xx00000x Insights into the Compatibility and Interphases of Li6PS5Cl Argyrodites and Zr-based Halide (LGZC) Electrolytes for All-SolidState Batteries Pedram Ghorbanzade a,b,c, Arianna Pesce a, Kerman Gomez a, Pedro López-Aranguren a and Juan Miguel López del Amo *a Cost-effective Li2.1Ga0.1Zr0.9Cl6 (LGZC) halides with high ionic conductivity (> 0.4 mS/cm at 25°C) are considered promising solid electrolytes for beyond Li-ion batteries. However, they suffer from low stability against Li metal, forming a solid electrolyte interphase (SEI) that causes continuous degradation, and limits their long-term cyclability. Incorporating Li6PS5Cl (LPSCl) electrolytes as an interlayer between LGZC and Li metal is a common approach to address this issue and improve cyclability. Nevertheless, very few studies assessed the compatibility of halides and sulfides, reporting contradicting results. Thus, this work used solid-state NMR and impedance spectroscopy to investigate interfacial reactions and compatibility between these phases. The results show that LPSCl and LGZC are chemically incompatible, and their interfacial reactions involve their decomposition to species with slower ionic mobility, decreasing the cell performance. Although their interface reaches stability with time, the decomposition reactions are accelerated at high temperatures. 6Li-6Li EXSY experiments demonstrate spontaneous Li-ion exchange between LGZC and LPSCl, despite their incompatibility. Interestingly the decomposed products participate in this exchange, explaining why these bilayer systems can successfully enhance the cell performance. Our study sheds light on the complex interfacial interaction between halides and sulfides, providing insights for a more optimal design of solid electrolytes for the new generation of electrochemical devices. Introduction All-solid-state battery (ASSB) is considered a promising technology able to increase both the energy density and safety of the device.1,2 In contrast to conventional lithium-ion batteries which use a liquid electrolyte and a separator, ASSBs employ a solid separator that performs both functions. Thanks to their high mechanical properties, solid electrolytes could enable the use of Li metal as the anode which offers a considerably higher theoretical energy density than traditional graphite anodes.3,4 In addition, since solid electrolytes do not contain any volatile and flammable organic solvents, they could address some safety issues of lithium-ion batteries.5 Sulfides and halides are highly promising and fast-emerging groups of solid electrolytes for ASSBs. They both offer a high room-temperature ionic conductivity, allowing rapid charge and discharge.6,7 In addition, they are both soft and easy to process, and unlike oxides, they do not require any high-temperature sintering.8 This important feature makes sulfides and halides particularly well-suited for large-scale manufacturing and commercialization8,9. The argyrodite type Li6PS5X (X=Cl, Br, I) are among the most promising sulfide electrolytes, owing to their high ionic conductivity, attributed to their highly disordered crystal structure. The room-temperature ionic conductivity of Liargyrodites reaches 7 mS.cm-1,10 close to the Li mobility in liquid electrolytes. Within this family, Li6PS5Cl is particularly popular as it offers a more stable interface with Li metal.11 Halides have the general formula of Li3MX6 (M=trivalent metals, X=halogen element), with Li3InCl6 and Li3YCl6 being among the most common compositions, offering ionic conductivities as high as 1.49 mS.cm-1.12 Substitution of elements with different ionic radii or valences is a typical approach to alter the properties of the halide by inducing disorder and vacancy in the cation/anion sublattices.13 For instance, the partial or total substitution of trivalent metal with tetravalent zirconium (Zr) leads to Li3-xZrxIn1-xCl6 or Li2ZrCl6 which are highly conducting, more tolerant to humidity, and more cost-effective.13 Although halide electrolytes offer a great oxidative stability potential (reaching 6 V vs Li+/Li in some compositions) and are compatible with most high-voltage cathode materials like LiNixMnyCo1-x-yO2 (NMC), their reductive stability potential is rather high,14 causing their degradation in contact with Li metal. This reaction, which was observed in several studies,15,16 involves the reduction of the transition metal to its reduced state according to reaction 1.14 a. Centre for Cooperative Research on Alternative Energies (CIC energiGUNE) Basque Research and Technology Alliance (BRTA), Albert Einstein 48, 01510 Vitoria-Gasteiz, Spain. E-mail: [email protected] b. University of Basque Country (UPV/EHU), Barrio Sarriena, s/n, 48940 Leioa, Spain. c. ALISTORE-European Research Institute, 80039 Amiens, France. Supplementary Information available: See DOI: 10.1039/x0xx00000x ARTICLE Journal Name 2 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Li  MCl  + 3 Li → 6 LiCl + M  (M = In, Y) reaction (1) This decomposition reaction may cause a significant increase in the solid electrolyte/Li metal interface resistance,16 due to the poor ionic conductivity of the byproducts present at the interphase. As reported in several studies, the interphase species have mixed ionic and electronic conductivity, which causes the reaction to proceed for several days before reaching stability.16,17 Common approaches applied to deal with this issue include modifying the halide chemistry,17 using Li-metal alloys instead of pure Li metal,9,18 or incorporating interlayers.19,20 These thin interlayers, which physically separate the halide and Li metal, have been found to effectively prevent interfacial reactions and improve cell cyclability.21 Keeping the interface resistance at a minimum requires the interlayer to be as thin and conductive as possible. Thus, sulfide argyrodite electrolytes, with their high ionic conductivity and ease of processability, are proposed and investigated in various studies. It must be noted that argyrodites are also unstable against the Li metal.22,23 However, the products of their degradation reaction, typically Li2S, LiCl, and Li3P form a stable solid electrolyte interphase (SEI) that passivates the electrolyte from further decomposition, thus allowing long cycling of the cell.21,23,24 Ji et al.20 demonstrated that incorporating an argyrodite LPSCl buffer layer prevents the undesired reduction of Li3YCl6 and significantly improves the cell cyclability, enabling a stable plating/stripping over 1000 h at 0.2 mA/cm2. In the same line, Wang et al.23 showed that the LPSCl interlayer stabilizes the solid electrolyte/Li metal interface and remarkably decreases the cell overpotential during plating/stripping. Zhang et al.21 reported a similar enhancement in a Li2ZrCl6-based system. This multilayer design enabled ASSBs with high coulombic efficiency and superior cycling behavior.21 While investigating the performance improvement of these multilayer systems has been the subject of several studies, little effort has been devoted to investigating the chemical compatibility and the interphases between the halides and argyrodites. Moreover, the few available works on the literature report contradicting results. While Janek et al.18 claim that the interfacial reactions between halides and argyrodites are negligible, Tarascon et al.25 report their incompatibility and question the hetero-structural cell design, in which these phases come into contact. Kwak et al.26 also demonstrated that Li2ZrCl6 and LPSCl are incompatible, but these interfacial reactions are driven electrochemically at high temperatures. Therefore, further research on this topic and clarification of the reactivity is required. In addition, a deeper understanding of the fundamentals of ion transport between these phases and its controlling parameters is necessary for designing cells with improved performance. This work focuses on the interaction of halide and argyrodite electrolytes, aiming to investigate the role of the interphase species in the local Li+ exchange between halides and argyrodites. Understanding the transport properties of the interphase species is crucial in all cell configurations in which these materials are in contact. This also applies to the relevant works from Kim et al.27 and Ye et al.28 in which they proposed coating the cathode materials with a thin halide layer to prevent the oxidation of the argyrodite electrolytes vs high-voltage cathodes. In a previous study, solid-state NMR proved to be an effective tool for investigating the compatibility of electrolyte materials and examining the composition and properties of interphase species.29 However, this method requires the formation and presence of a sufficiently high amount of interphases. Thus, a composite electrolyte design was used as a model system to increase the effective surface area and to generate more interphase species compared to a bilayer configuration. The same approach is applied in the current work as it enables a more in-depth investigation of the interfacial chemistry between halides and argyrodites, even though these materials are often used in bilayer configurations in practical applications. The central role of solid-state NMR in this study stems from its sensitivity to the chemical environment, making it ideal for distinguishing between different lithium environments in halide and sulfide phases. Moreover, the interfacial Li-ion exchange between the two phases can be studied by applying Exchange Spectroscopy (EXSY) NMR experiments.30,31 These twodimensional NMR experiments allow for probing the exchange processes between two different sites within a certain time frame. Such local short-range interfacial exchanges are critical in composite or multilayer solid electrolytes as they could impact the long-range ion transport properties, ultimately affecting the overall cell performance. In addition to EXSY, monitoring the spectrum evolution over time can provide important insights into potential decomposition reactions arising from phase incompatibility. For instance, in the case of decomposition reactions, a decrease in the signal intensity and the emergence of new signals corresponding to decomposition products is expected. These spectral changes provide a direct indication of chemical instability at the interphase. The interfacial reactions can also influence the full width at half-maximum (FWHM) of the NMR signals, commonly referred to as signal linewidth. The linewidth generally correlates with local ion dynamics32 and can reveal how the interface interactions impact the local ion motions in individual phases. Combining these measurements with EXSY forms a more complete picture of both chemical and dynamic processes occurring at the halide-argyrodite interface, leading to a more comprehensive understanding of the underlying mechanisms that govern short and long-range ionic conductivity in solid-state electrolytes. The halide material investigated in this work corresponds to the chemical formula Li2.1Ga0.1Zr0.9Cl6 (LGZC). Using Zr as the transition metal makes this halide more sustainable and lowers the cost compared to Yor In-based halides. However, the bulk ionic conductivity of the dopant-free Li2ZrCl6 is considerably lower than its Yand In-based competitors. This issue is commonly addressed by the addition of aliovalent dopants such as Ga or Fe. For instance, Kwak et al.33 reported a maximum ionic conductivity of ∼ 1 mS.cm-1 upon 25% doping with Fe3+. Similarly, we observed that 10% Ga doping has increased the room temperature ionic conductivity of the halide up to 0.44 mS.cm-1, compared to 0.103 mS.cm-1 previously reported for Journal Name ARTICLE This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 3 Please do not adjust margins Please do not adjust margins dopant-free LZC. 34 This improvement could be attributed to the induced disorder in the halide structure, as well as the increased Li content. Experimental section Materials and synthesis procedure The LGZC halide materials were synthesized by mixing stoichiometric amounts of LiCl, GaCl 3 , and ZrCl 4 and applying a soft ball-milling process at 600 rpm for 8 cycles of 40 minutes, including 10 minutes of pause after every cycle to prevent overheating. The crystal structure and the composition of the synthesized halide material were confirmed by synchrotron XRD analysis performed at the Alba synchrotron. These structural analyses are presented in Figure S.1 and show no traces of the precursors and a crystal structure corresponding to the doped Li 2 ZrCl 6 . Commercial Li 6 PS 5 Cl argyrodites (LPSCl) were purchased from NEI manufacturing. The analysis of their XRD pattern shown in Figure S.2 confirms the phase purity. The synthesized halides and commercial Li 6 PS 5 Cl argyrodites were mixed in a 3:1 ratio and were hand-milled for 5 mins. The mixture was pressed at 700 MPa to obtain dense pellets with 6 mm diameter. For the NMR measurements, the pellets were crushed into powders to fill the NMR rotor. The main purpose of the pressing step is to ensure good contact between the halide and argyrodite particles and facilitate the interactions and formation of interphases. To observe the impact of temperature on the interfacial reactions, the rotors were heated at 70°C in a vacuum atmosphere. All the materials were handled and processed in an Ar-filled glovebox to avoid moisture absorption and degradation. Solid State NMR Magic angle spinning nuclear magnetic resonance (MAS NMR) spectra were recorded utilizing a Bruker Avance III 500 spectrometer, a 2.5 mm probe, and a MAS frequency of 20 kHz. 1 H and 6 Li chemical shifts were referenced to 0.1 M LiCl aqueous solution. 6 Li 1D NMR spectra were recorded using single excitation π/2 pulses of 3 µs and a relaxation delay of 30 s. The 1 H 1D NMR spectra were recorded by the Hahnecho pulse sequence in which the π/2 and π pulses are 2.1 and 4.2 µs, and the relaxation delay was set to 5 seconds. The 6 Li6 Li and 7 Li7 Li EXSY NMR experiments were recorded by standard three π/2 pulses of 3 and 2.4 µs for 6 Li and 7 Li respectively, with mixing time being set to 256 ms for both experiments. Electrochemical Impedance Spectroscopy (EIS) The Electrochemical Impedance spectroscopy (EIS) experiments were performed on a multilayer pellet in which LGZC is sandwiched between two LPSCl layers and carbon-coated aluminum disks were pressed together with the powders to improve the interface contact. The resulting pellets were placed in a Swagelok cell with stainless steel plungers without further modification. The impedance measurements were conducted using a Solartron 1260 FRA module, applying a bias voltage of 20 mV from 32 MHz to 10 Hz. Finally, the spectra were analyzed using ZView® software from Scribner. All impedance spectra were recorded at room temperature. Distribution of relaxation times (DRT) analyses were applied to the recorded impedance spectra, using the pyDRTtools in the MATLAB toolbox to resolve the overlapping transport processes in the EIS frequency domain. 35 PyDRT is a Python-based graphic user interface (GUI) based on Bayesian ridge regression (also known as Tikhonov regularization). 36,37 The suitability of the EIS spectra for the DRT analyses was verified by the Kramer-Kronig validity test method using the Lin-KK software developed by Karlsruhe Institute of Technology (KIT). 38 Results and discussion NMR Characterization of pure LGZC In the first step and before the characterization of the composite electrolytes, the LGZC halides were analyzed by solid-state NMR. This characterization is conducted firstly to confirm the absence of secondary species over storage time, but also to use as a reference when analyzing the NMR spectrum of the composite electrolytes. The 6 Li spectrum displayed in Figure 1.a shows a strong, narrow signal at approximately -0.85 ppm, with a linewidth of 2.8 Hz, corresponding to the highly mobile Li-ions in the halide structure. Notably, a minor shoulder appears at -0.9 ppm, which may be attributed to protonated phases of the halide, as indicated by measurements on an aged halide sample (Figure S.3). It is challenging to identify these species solely based on NMR experiments. However, it has been previously reported that Li 3 InCl 6 can turn into Li 3 InCl 6 .xH 2 O upon exposure to moisture. 39,40 Thus, the formation of a similar compound such as Li 2.1 Gr 0.1 Zr 0.9 Cl 6 .xH 2 O could be expected. The broader nature of this shoulder signal suggests reduced Li + mobility, aligning with this interpretation. However, the low intensity of the shoulder and the narrow linewidth of the main signal imply that protonation is minimal, indicating that the synthesized halides are in good condition. To further validate this, the 1 H spectrum of the halide was recorded and analyzed. As seen in Figure 1.b, the 1 H spectrum reveals several broad signals. Yet, the low Figure 1 a) 6 Li and b) 1 H NMR spectra of LGZC halide electrolyte ARTICLE Journal Name 4 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins intensity of these signals, despite the high sensitivity of NMR to the 1 H nucleus, confirms that protonation is indeed minimal. Moreover, the stark contrast between the 1 H spectra of the fresh and aged halide samples presented in Figure S.4 supports the conclusion that the synthesized LGZC powders are adequately dry. It must be noted that high sensitivity to 1 H -owing to the high natural abundance and gyromagnetic ratio of 1 Hmakes solidstate NMR an excellent tool for monitoring the protonated phases. Since the protonated phases might be amorphous or too thin, they might not be easily detected by other characterization techniques such as powder X-ray Diffraction (PXRD). This is similar to the case of LLZO garnets where protonated phases are sometimes underestimated due to inaccurate material characterization. 41 This is highly important because the protonated phases or the decomposition products of the protonation are known to be undesirable, decrease the ionic conductivity of the halides, and in general, negatively impact their performance. 39,40 Thus, further research into the protonation and deprotonation mechanisms of halide materials is needed. NMR Characterization of LPSCl/LGZC Composites In the next step, halide/sulfide composite electrolytes were prepared by hand milling and pressing, as described in the experimental section. Subsequently, NMR measurements were conducted to investigate their potential interfacial reactivity along with transport properties of the formed species. In the 6 Li NMR spectrum of the composite, shown in Figure 2, two sets of signals at positive (1.8 to 0.8 ppm) and negative (-0.5 to -1.3 ppm) chemical shifts are observed which are attributed to the sulfide and halide phases respectively. While the 6 Li signals of the LPSCl sulfide at 1.5 ppm and its shoulder at 1.4 ppm are similar to its pure form, the less intense signal at 1.1 ppm only appears at the composite electrolyte, already suggesting the reactivity between sulfide and halide. This interphase could be due to the interface reactions between the halide and sulfide, possibly involving their decomposition. Similarly, the signal at - 1.1 ppm is also attributed to the products of halide decomposition, as it was absent in the pure halide 6 Li spectrum, shown in Figure 1. The assignment of the signals at 1.1 and -1.1 ppm to the interfacial reaction products was further validated by heating the samples overnight and subsequently repeating the NMR measurements. The heating is expected to facilitate the reactions between the LGZC and LPSCl, increasing the reaction products. Indeed, the evolution of the 6 Li NMR spectra demonstrated in Figure 3 reveals a marked increase in the intensity of the Li signals at 1.1 and -1.1 ppm after heating, confirming that these signals are directly correlated with the extent of the interfacial reactions. Additionally, this increase in signal intensity is accompanied by a corresponding decrease in the intensity of the Li signals at 1.4 and -0.85 ppm, indicating that the observed reactions involve the decomposition of both the halide and sulfide phases. This spectrum evolution provides strong evidence for the chemical reactions occurring at the interface and highlights the dynamic nature of the interphase as the reaction progresses. As shown in Figure S.5, this spectrum evolution was also accompanied by an increase in the 7 Li T 1 relaxation time. This indicates that the interphase species have slower local Li dynamics compared to the main argyrodite and halide phases. Thus, these reactions are expected to be detrimental to the composite electrolyte's performance, and they must be minimized. Figure 2 6 Li NMR spectrum of LPSCl/LGZC composite electrolyte. Figure 3 6 Li spectrum of pristine and heat-treated LPSCl-LGZC composite. Journal Name ARTICLE This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 5 Please do not adjust margins Please do not adjust margins EIS and DRT analysis To better understand the impact of this reactivity on the cell performance, EIS experiments were conducted and the evolution of the impedance spectra with time was monitored. As demonstrated in Figure 4, the resistance of the multilayer electrolyte with blocking electrodes increases with time but reaches stability in less than 9 hours. This increase in resistance can be attributed to the instability of the two phases and the formation of less conductive phases through interfacial reactions between the LGZC and LPSCl, as was suggested by NMR results. Since the NMR experiments revealed the impact of high temperatures on the interfacial reactions between LPSCl and LGZC, the multilayer pellets were heated at 70°C for 64 hours and returned to room temperature for impedance measurements. The results show a significant increase in the cell impedance, implying interfacial reactions and decomposition of highly conductive halide and sulfide phases. This can be further validated by the distribution of relaxation times (DRT) analysis (Figure 4.b) which shows different ion dynamics and their corresponding timescale. In the DRT spectrum of the multilayer pellets, 3 main resistive processes are observed labeled as R1, R2, and R3, which are typically assigned to ion transport at bulk, grain boundaries, and passivation layer/electrolyte interface respectively. 42,43 Comparing the DRT spectra at 0.5 and 18 h reveals a minor increase in the resistance for the medium-frequency R3 process (1-6 kHz) while the high-frequency R1 peak (5-13 MHz) exhibits almost no change during this period. In contrast, after heating, the DRT spectrum shows a clear change in the high-frequency region and a shift of the R1 peak to lower frequencies. This shift suggests a degradation in the bulk ion transport properties, consistent with the partial decomposition of the highly conducting halide and sulfide phases, as confirmed by NMR analysis. Moreover, the substantial increase in resistance at low-frequency R3 after heating can be explained by the formation of decomposition products at the interphase, which exhibit slower ion dynamics. These observations confirm that higher temperatures facilitate the interfacial reactions between LPSCl and LGZC, leading to the formation of secondary phases with slower dynamics, as previously suggested by NMR. Figure 5 a) 6 Li6 Li and b) 7 Li7 Li EXSY NMR spectra of LPSCl-LGZC composite electrolytes Figure 4 a) Nyquist plot and b) DRT analysis of LPSCl/LGZC composite electrolytes and their evolution with time and temperature ARTICLE Journal Name 6 | J. Name., 2012, 00, 1-3 This journal is © The Royal Society of Chemistry 20xx Please do not adjust margins Please do not adjust margins Finally, while it might seem that these reactions have less impact on the R2 peak, with careful analysis, an increase in the intensity of this peak after heating is observed. Again, this is explained by the decomposition of the main phase, negatively affecting the ion transport in grain boundaries. Overall, the DRT and EIS analyses align well with the NMR findings, providing a comprehensive picture of the interfacial degradation and the resulting impact on ion transport processes. Interfacial Li Exchange In the next step, the role of these species in the interfacial Li exchange between the two phases is investigated by EXchange SpectroscopY (EXSY) experiments, presented in Figure 5. This two-dimensional NMR experiment allows for observing the magnetization transfer between two different sites within a limited period, called mixing time. The double Fourier transformation generates a 2D graph typically demonstrated with contour levels. The signals appearing on the diagonal line correspond to the Li ions that remained in their initial environment. In contrast, the off-diagonal signals, also called cross-peaks, represent the Li ions that have exchanged their environment. If the interphase species blocked the interfacial ion exchange between the halide and argyrodite, no cross-peak could be observed. Similarly, if the exchange is too slow and its time constant is longer than the mixing time, no exchange would be visible. The results from our 6Li-6Li EXSY measurements demonstrated in Figure 5.a show clear exchange between the argyrodite signals (1.1 and 0.8 ppm) and the main halide signal (-0.85 ppm). Additionally, a weak cross-peak between the two halide signals at -0.85 and -1.1 ppm is observed. In theory, cross-peaks might also appear due to spin diffusion between two neighboring nuclei, a phenomenon that strongly depends on the internuclear distance. However, considering the natural abundance of the 6Li (around 7%), the spin diffusion between two neighboring 6Li is rather unlikely and cannot cause such clear signals. Thus, the observed cross-peaks unambiguously represent the physical exchange between the halide and argyrodite. This implies that even if the interphase species have slower dynamics, they are not blocking the Li-ion exchange between the halide and sulfides. To better observe the exchange between the minor phases, i.e. decomposed sulfide (1.1 ppm) and decomposed halide (-1.1 ppm), 7Li-7Li experiments (Figure 5.b) were conducted. Thanks to the higher natural abundance and gyromagnetic ratio of 7Li, this experiment offers a higher signal intensity, although this comes at the cost of decreased resolution. Despite the rather low intensity of the cross-peak signals, the results show an exchange between decomposed argyrodite (1.1 ppm) and decomposed halide (-1.1 ppm). Combining this observation with the previously mentioned exchanges between the rest of the signals, it can be concluded that these minor species indeed participate in the complex ion dynamics of the system. This contrasts with the LPSCl and PEO-LiTFSI composite, in which the decomposition products at the interphase were found to impede the interfacial exchange.29 This interfacial Li exchange and its rate are of high importance, as they can impact and control the ion transport mechanism at a larger scale. For instance, in a halide-argyrodite multilayer setup, if this exchange does not occur fast enough, the interphase could be the bottleneck for the ion transport, and the overall ionic conductivity of the cell is compromised. Concluding Remarks In this study it was shown that LGZC and LPSCl are chemically incompatible, driving forward an interfacial reaction that causes their partial decomposition into less conductive phases. While precise identification of these decomposition products remains challenging and requires further investigation, it was shown that these products are detrimental to cell performance and decrease the ionic conductivity of the cells. Both NMR and Impedance spectroscopy revealed that despite their occurrence, these interfacial reactions stabilize rather rapidly. Nevertheless, the reactions may continue to progress at higher temperatures, indicating that thermal conditions could exacerbate interfacial degradation. In addition, using EXSY NMR experiments, the interfacial Li+ exchange between LGZC and LPSCl was observed. It was shown that this exchange, which is essential for fast ion transport through the interphase, directly involves the decomposed phases present at the interphase. It must be emphasized that this study in composite design involving heating highlights the interfacial reactions for a better fundamental understanding of the system. In a bilayer design which is the setup of choice in practical applications, the contact area between the phases is considerably smaller, thus the extent of these reactions is expected to be lower. Finally, while the results presented in this work confirm the interfacial reactions between the LGZC and LPSCl solid electrolytes, it must be noted that the interactions between halides and argyrodites highly depend on their chemistries and these conclusions cannot be extended to the whole family of halides with different chemical compositions. However, the methods used in this study can be applied to other systems with relative ease to determine the compatibility of different halides and argyrodites. Author contributions PG: investigation, formal analysis, writing-original draft, and writingreview & editing. AP: investigation and writing-review & editing. KG: investigation. PL: writing-review & editing. JL: investigation, supervision, and writing-review & editing. Conflicts of interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Journal Name ARTICLE This journal is © The Royal Society of Chemistry 20xx J. Name., 2013, 00, 1-3 | 7 Please do not adjust margins Please do not adjust margins Data availability The data that support the findings of this study are available on request from the corresponding author. Acknowledgments As a part of the DESTINY PhD programme, this publication is acknowledged by funding from the European Union's Horizon2020 research and innovation programme under the Marie Skłodowska-Curie Actions COFUND (Grant Agreement #945357). These experiments were performed at BL04-MSPD beamline at ALBA Synchrotron with the collaboration of ALBA staff. This project has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No 101069726 (SEATBELT Project). Notes and references 1 Q. Wang, Y. Zhou, X. Wang, H. Guo, S. Gong, Z. Yao, F. Wu, J. Wang, S. Ganapathy, X. Bai, B. Li, C. Zhao, J. Janek and M. Wagemaker, Nat Commun, 2024, 15, 1050. 2 T. Famprikis, P. Canepa, J. A. Dawson, M. S. Islam and C. Masquelier, Nat. Mater., 2019, 18, 1278–1291. 3 S. Kim, J.-S. Kim, L. Miara, Y. Wang, S.-K. Jung, S. Y. Park, Z. Song, H. Kim, M. Badding, J. Chang, V. Roev, G. Yoon, R. Kim, J.-H. Kim, K. Yoon, D. Im and K. Kang, Nat Commun, 2022, 13, 1883. 4 T. Krauskopf, F. H. Richter, W. G. Zeier and J. Janek, Chem. Rev., 2020, 120, 7745–7794. 5 Y. Chen, Y. Kang, Y. Zhao, L. Wang, J. Liu, Y. Li, Z. Liang, X. He, X. Li, N. Tavajohi and B. Li, Journal of Energy Chemistry, 2021, 59, 83– 99. 6 C. Wang, J. Liang, J. T. Kim and X. Sun, Science Advances, 2022, 8, eadc9516. 7 C. Li, S. Zhang, X. Miao, C. Wang, C. Wang, Z. Zhang, R. Wang and L. Yin, Batteries & Supercaps, 2022, 5, e202100288. 8 H. Huo and J. Janek, National Science Review, 2023, 10, nwad098. 9 T. Asano, A. Sakai, S. Ouchi, M. Sakaida, A. Miyazaki and S. Hasegawa, Advanced Materials, 2018, 30, 1803075. 10 S. Boulineau, J.-M. Tarascon, J.-B. Leriche and V. Viallet, Solid State Ionics, 2013, 242, 45–48. 11 L. Zhou, K.-H. Park, X. Sun, F. Lalère, T. Adermann, P. Hartmann and L. F. Nazar, ACS Energy Lett., 2019, 4, 265–270. 12 X. Li, J. Liang, J. Luo, M. N. Banis, C. Wang, W. Li, S. Deng, C. Yu, F. Zhao, Y. Hu, T.-K. Sham, L. Zhang, S. Zhao, S. Lu, H. Huang, R. Li, K. R. Adair and X. Sun, Energy Environ. Sci., 2019, 12, 2665–2671. 13 K. Tuo, C. Sun and S. Liu, Electrochem. Energy Rev., 2023, 6, 17. 14 B. Tao, D. Zhong, H. Li, G. Wang and H. Chang, Chemical Science, 2023, 14, 8693–8722. 15 S. Wang, Q. Bai, A. M. Nolan, Y. Liu, S. Gong, Q. Sun and Y. Mo, Angewandte Chemie International Edition, 2019, 58, 8039–8043. 16 L. M. Riegger, R. Schlem, J. Sann, W. G. Zeier and J. Janek, Angewandte Chemie International Edition, 2021, 60, 6718–6723. 17 P. Lannelongue, S. Lindberg, E. Gonzalo, A. Golov, F. Bonilla, J. M. Lopez del Amo, T. Marchandier, A. Tron, J. Carrasco and P. LópezAranguren, Energy Storage Materials, 2024, 72, 103733. 18 C. Rosenbach, F. Walther, J. Ruhl, M. Hartmann, T. A. Hendriks, S. Ohno, J. Janek and W. G. Zeier, Advanced Energy Materials, 2023, 13, 2203673. 19 X. Shi, Z. Zeng, M. Sun, B. Huang, H. Zhang, W. Luo, Y. Huang, Y. Du and C. Yan, Nano Lett., 2021, 21, 9325–9331. 20 W. Ji, D. Zheng, X. Zhang, T. Ding and D. Qu, J. Mater. Chem. A, 2021, 9, 15012–15018. 21 H. Zhang, Z. Yu, J. Cheng, H. Chen, X. Huang and B. Tian, Chinese Chemical Letters, 2023, 34, 108228. 22 A. Golov and J. Carrasco, ACS Appl Mater Interfaces, 2021, 13, 43734–43745. 23 C. Wang, J. Liang, J. Luo, J. Liu, X. Li, F. Zhao, R. Li, H. Huang, S. Zhao, L. Zhang, J. Wang and X. Sun, Science Advances, 2021, 7, eabh1896. 24 S. Wenzel, S. J. Sedlmaier, C. Dietrich, W. G. Zeier and J. Janek, Solid State Ionics, 2018, 318, 102–112. 25 T. Koç, F. Marchini, G. Rousse, R. Dugas and J.-M. Tarascon, ACS Appl. Energy Mater., 2021, 4, 13575–13585. 26 H. Kwak, J.-S. Kim, D. Han, J. S. Kim, J. Park, G. Kwon, S.-M. Bak, U. Heo, C. Park, H.-W. Lee, K.-W. Nam, D.-H. Seo and Y. S. Jung, Nat Commun, 2023, 14, 2459. 27 J. S. Kim, S. Jung, H. Kwak, Y. Han, S. Kim, J. Lim, Y. M. Lee and Y. S. Jung, Energy Storage Materials, 2023, 55, 193–204. 28 Q. Ye, X. Li, W. Zhang, Y. Xia, X. He, H. Huang, Y. Gan, X. Xia and J. Zhang, ACS Appl. Mater. Interfaces, 2023, 15, 18878–18888. 29 P. Ghorbanzade, A. Pesce, M. Armand, K. Gómez, S. Devaraj, P. López-Aranguren and J. M. López del Amo, Small Structures, 2024, 5, 2400139. 30 P. Ghorbanzade, G. Accardo, K. Gomez, P. López-Aranguren, S. Devaraj, C. M. Costa, S. Lanceros-Mendez and J. M. López del Amo, Materials Today Energy, 2023, 38, 101448. 31 P. Ranque, J. Zagórski, S. Devaraj, F. Aguesse and J. M. L. del Amo, J. Mater. Chem. A, 2021, 9, 17812–17820. 32 F. Nardelli, S. Borsacchi, L. Calucci, E. Carignani, F. Martini and M. Geppi, Rend. Fis. Acc. Lincei, 2020, 31, 999–1010. 33 H. Kwak, D. Han, J. Lyoo, J. Park, S. H. Jung, Y. Han, G. Kwon, H. Kim, S.-T. Hong, K.-W. Nam and Y. S. Jung, Advanced Energy Materials, 2021, 11, 2003190. 34 X. Luo, Y. Zhong, X. Wang, X. Xia, C. Gu and J. Tu, ACS Appl. Mater. Interfaces, 2022, 14, 49839–49846. 35 T. H. Wan, M. Saccoccio, C. Chen and F. Ciucci, Electrochimica Acta, 2015, 184, 483–499. 36 A. Maradesa, B. Py, T. H. Wan, M. B. Effat and F. Ciucci, J. Electrochem. Soc., 2023, 170, 030502. 37 M. Saccoccio, T. H. Wan, C. Chen and F. Ciucci, Electrochimica Acta, 2014, 147, 470–482. 38 M. Schönleber, D. Klotz and E. Ivers-Tiffée, Electrochimica Acta, 2014, 131, 20–27. 39 W. Li, J. Liang, M. Li, K. R. Adair, X. Li, Y. Hu, Q. Xiao, R. Feng, R. Li, L. Zhang, S. Lu, H. Huang, S. Zhao, T.-K. Sham and X. Sun, Chem. Mater., 2020, 32, 7019–7027. 40 S. Wang, X. Xu, C. Cui, C. Zeng, J. Liang, J. Fu, R. Zhang, T. Zhai and H. Li, Adv Funct Materials, 2022, 32, 2108805.