Na-ion Batteries. Laure Monconduit

also competitive in terms of cost, performance and safety and is already developed to be introduced rapidly into the market.

I.2. From the electrodes to the electrolyte for NIBs

      I.2.1. Positive electrodes

      The first two chapters are dedicated to the positive and negative electrodes for a NIB. Among the materials intensively studied as prospective positive electrode for NIBs are polyanionics, layered oxides, Prussian blue analogues and organic compounds.

With their high theoretical capacity and ionic mobility, layered transition metal oxides are likely the most promising positive electrodes. It is interesting to mention that the alkali deintercalation/intercalation reactions in layered compounds were studied for the first time in Na_xCoO₂ (Braconnier et al. 1981; Delmas et al. 1981) before the demonstration of lithiation/delithiation reactions in the 2D lithium layered oxides. This structure presents a huge versatility of stackings and compositions by playing with the composition in 3d transition metals and in sodium that gives the potential to modulate their electrochemical properties. Their performance and the remaining challenges are detailed in Chapter 1.

      Among the promising positive electrode materials for NIBs, although achieving lower specific capacity than the layered oxides, polyanionic compounds represent an important class of materials. Indeed, they are characterized by high operating voltage, excellent chemical stability and life span. A large panel of structures and compositions exist, with as for layered oxides the possibility to tailor significantly the operating voltage, and thus the energy density, by varying the nature of the involved redox couple. Today, the most attractive materials are vanadium-containing phosphates, characterized by two different structures, Na₃V₂(PO₄)₃ and Na₃V₂(PO₄)₂F₃. Chapter 2 shows the versatility of these families of compounds, highlighting their differences, their interesting properties and the challenges for their development in the next generations of NIBs.

      I.2.2. Negative electrodes

      Carbon materials present great interest as negative electrodes because of their low cost and environmental benignity, as well as their excellent electrochemical stability. In LIBs, graphite is the standard negative electrode. It exhibits a low operating voltage and 370 mAh/g specific capacity based on reversible Li deintercalation and reintercalation from/in between the graphene layers. Huge efforts were devoted to using it in NIBs (Stevens and Dahn 2001). Nevertheless, due to the thermodynamic instability of stage 1 Na-GIC formation, graphite was rapidly abandoned as the negative electrode. More recently, it was shown that when used in linear ether solvents, co-intercalation of ether molecules was observed and beneficial for Na⁺ intercalation in between graphene layers even though it does not lead to acceptable cycles performance (Kim et al. 2015).

      As the utilization of graphite as the negative electrode for NIBs failed, non-graphitic carbon materials such as hard carbon and soft carbon have been extensively investigated as alternatives (Dahbi et al. 2014). Chapter 3 discusses the main synthesis routes used to prepare hard carbons and the impact of both the nature of the precursor and synthesis conditions on the hard carbon features. Relationships between these synthesis conditions and the electrochemical properties are discussed in details, with a specific focus on the sodium storage mechanism, still a debate in literature.

Metal oxides have also been intensively studied as potential alternatives to carbon. Due to their low operation voltage as well as low cost and environmental benignity, Ti-based oxide compounds have been widely studied. The conversion reaction compounds based on metal oxides, and extended to sulfides, phosphides, fluorides, etc., have also recently attracted attention because of their high specific capacity involving the exchange of more than one electron per transition metal. These conversion type materials MX_n (M active (X) or inactive (Fe, Co, Ni, Cu, Mn, Mo) vs. Na and X = O, S, P, Ge, Sn, Sb, etc.) will be described in Chapter 4. High theoretical capacities, higher than those obtained for carbon-based materials, are also expected for Na-rich alloy phases from the phase diagram of most of p-block elements with Na. Silicon, which is the most studied alloy material for LIB due to the extremely high theoretical specific capacity, has been shown to be quite inactive versus Na. On the contrary, the 14 and 15 group elements with Ge, Sn, P and Sb can show excellent electrochemical behavior, sometimes better than in LIBs, even though they all present a huge volume expansion when alloyed with Na.

      I.2.3. Electrolytes and the solid electrolyte interphase

      The electrolytes are essential for the proper functioning of any battery technology, with an important focus to minimize interface electrolyte/electrode reactions and enhance both performance and safety. First studies surveyed electrolytes prepared using classical alkyl carbonates solvents and mixtures, in combination with different Na salts (Ponrouch et al. 2012). Their viscosity, ionic conductivity, thermal and electrochemical stability were evaluated to establish some intrinsic trends to identify the first electrolyte formulations with the widest range of applicability for NIBs.

      More recently, IL, defined as room temperature molten salts and composed mainly of organic cations and (in)organic anions and presenting a huge versatility of structural variations, were proposed as major alternatives for the development of optimized electrolytes. Indeed, IL offer unique physical and chemical properties associated with low volatility that make them extremely interesting for the development of electrolytes with higher electrochemical and thermal stability. Chapter 5 will give the major trends for the IL-based electrolytes developed for NIBs.

The concept of solid electrolyte interphase (SEI), which is formed from the contact of the electrode with the electrolyte, was first introduced by Peled (1979) and then widely accepted in the LIB community, before being logically extended to NIBs as the nature of the electrodes and electrolyte are antecedent. Chapter 6 describes the SEI formation in NIBs, which has to be passivating for the electrode surface and stable upon cycling to get good cycling performance. Although the SEI layer and its formation have been intensely investigated during the last decades, even in LIBs it is still considered to be misunderstood being strongly dependant on a series of parameters such as the composition of the electrolyte, the nature of the electrodes and the cycling conditions. Moreover, regarding NIBs, to date there are only a few reports addressing the fundamental aspects of SEI chemistry. Recent works showed than even though the knowledge coming from LIBs is helpful for understanding the electrolyte/electrode interfaces modifications, a direct transfer of knowledge from LIBs to NIBs is not guaranteed. Indeed, many parameters have to be taken into account: the higher standard potential of Na⁺/Na than Li⁺/Li (~0.33 V), the higher solubility of carbonates from the SEI in NIBs, the lower desolvation energy of Na⁺, the larger coordination shell of Na⁺, the higher water reactivity of Na⁺, etc. (Eshetua et al. 2019). These different features tailoring the nature and stability of the SEI in NIBs will be described in detail in Chapter 6.

I.3. Future commercialization of NIBs

      Although LIBs are unquestionably the leader for electrochemical energy storage, their costs remain high and dependent on the demand that continuously and exponentially increases. In contrast, owing to the abundance of sodium, the price of NIBs could be preserved even with a large demand for energy storage. Furthermore, due to the proximity of their chemistry, the technology developed for LIBs should be promptly transferable to NIBs and easily upscaled. Widespread commercialization and mass production