Stable low temperature rhombohedral iron trifluoride has been obtained by the fluorination under the pure fluorine gas of iron disilicide. The combination of both unusual fluorination process and precursor avoids to get unhydrated crystalline FeF 3 particles and allows the formation of hierarchized channels of mesoporous/macroporous texture favorable for lithium diffusion. The fluorination mechanism proceeds by temperature steps from the formation, for a fluorination temperature below 200 °C, of an amorphous phase and an intermediate iron difluoride identified mainly by 57 Fe Mössbauer spectroscopy before getting, as soon as a fluorination temperature of 260 °C is reached, the rhombohedral FeF 3 . Both amorphous and crystallized samples display good ability for electrochemical process when used as cathode in lithium-ion battery. The low diameter of rhombohedral structure channels is balanced by an appropriate mesoporous texture and a capacity of 225 mAh.g -1 after 5 cycles for a discharge cut-off of 2.5 V vs. Li + /Li at a current density of C/20 has been obtained and stabilized at 95 mAh.g -1 after 116 cycles.

Choosing the best cathode material used in a Li-ion battery is one of the most crucial issues in achieving higher energy densities, since the energy density is directly correlated to the specific capacity associated with that cathode material. Conversion based cathode materials tend to exhibit substantially high capacities, due to the fact that essentially all the possible oxidation states of the compound during the redox reaction can be used. Transition metal fluorides have recently been investigated as potential cathode materials because of their high electronegativity. Tarascon’s group first demonstrated that these conversion materials, which were Co3O4, CoO, NiO, and FeO, can exhibit high specific capacities of 600–1000 mAh g− 1 along with good cycling properties. However, these metal oxides are only suitable for use as negative electrodes, due to their low conversion potential, which is lower than 1.0 V. Transition metal fluorides have recently been investigated as potential cathode materials because of their high electronegativity. However, the insulating nature of metal fluorides has limited their electrochemical properties for a long time; for that reason a considerable amount of attention has been devoted by Badway et al., to tailoring their nanostructures to overcome poor electronic conductivity. It has been demonstrated that nanocomposite of carbon and metal fluorides such as FeF2, FeF3, and BiF3 may be utilized as cathodes for next generation high energy lithium ion batteries, revealing a high conversion potential and good cycling properties. However there are very few studies on NiF2 based conversion cathode materials in terms of the conversion mechanism and electrochemical properties, due to its poor electrochemical properties compared to other metal fluorides. In this study, we have chosen to insure some electronic conductivity into NiF2 by partial conversion of nanoparticles of nickel into NiF2 through fluorination under pure molecular fluorine. For such purpose, Ni nanoparticles of median diameter of 50 nm were purchased and fluorinated at temperatures ranged in between room temperature and 450°C. It was then possible to get different proportions of NiF2 shell onto Ni core. The structure of Ni-doped NiF2 was investigated by X-ray diffraction (XRD) and the texture by transmission electron microscopy (MET). Two fluorination mechanisms appears depending on the fluorination temperatures. For temperatures lower than 300°C, the initial spherical shape is preserved but only 73% of Ni can be converted into NiF2. Increasing the fluorination temperature leads to damages into the spherical shape. However these damages can be some preferential way for lithium diffusion. Total conversion of Ni into NiF2 can then be obtained Structural data obtained by XRD combined with the weight uptake evolution got at the end of the fluorination show first the progressive formation of a NiF2 shell which appears as amorphous until 250°C i.e. a weight percentage of NiF2 lower than 24%. Then, the proportion of NiF2 and Ni phases becomes only the same by XRD and weight uptake as soon as the NiF2 phase is perfectly crystallized i.e. for fluorination temperature higher than 400°C. The coherence lengths of each phase evolve with the fluorination temperature and will be discussed owing to the different crystallographic planes of each phase. Electrochemical performances of the various samples partially or totally converted have been performed in different electrolytes chosen for their different solubility of LiF but also at different cycling temperatures and current densities. No carbon additive have been necessary in electrode formulation of partially converted nickel nanoparticles and the consecutive performances will be compared to those of totally converted nickel nanoparticles mixed with classical amount of carbon nanofiber additive. The electrochemical performances are different owing to the electrode composition. For low current density (C/20), when only core-shell particles got by fluorination at 200°C are used together with PDVF as binder, the discharge capacities at the first cycle appears as high but only few of it can be recovered at the next oxidation step. So as soon as the second cycle the capacities are half the theoretical value of NiF2 (554 mAh/g). When totally fluorinated nickel particle must be used as electrode material, some carbon additive is needed for conductive purpose. The first discharge capacity is lower than the one of core shell particle but the performance is easily maintained upon cycle.

We report a multistep synthesis of anhydrous NiF 2 nanoparticles using Single Walled Carbon Nanotubes (SWCNT), Multi Walled Carbon Nanotubes (MWCNT) and cellulose as sacrificial supports. The synthesis method includes precipitation at the surface of carbonaceous supports, followed by gas-solid treatments under air and pure molecular fluorine. The as-prepared nanoparticles exhibit different particle size distributions depending on the initial support from~20 nm with SWCNT up tõ 60 nm with cellulose. The carbonaceous matrices act as sacrificial supports and are totally removed under fluorine gas. The influence of carbonaceous supports on the as-obtained nanoparticles is reported and discussed. Since SWCNT, MWCNT and cellulose exhibit different surface chemistries (nature of chemical sites, size related reactivity, …), they drive the formation of nanoparticles with different characteristics. Moreover, intermediate materials Ni(OH) 2 /C and NiO/C have been also studied and the opportunity to obtain promising structures like NiO/ graphene nanocomposites by this method has been demonstrated.