Electrochemical Behaviour of High Specific Capacity Spinel LixMn2-YTiyO4 As Cathode Material for Li-Ion Batteries

Abstract

The demand for high capacity batteries is increasing rapidly with the upcoming energetic needs of an ever-increasing population, especially in the transportation sector. Inexpensive electric vehicles with long range and high durability are also essential for decarbonisation targets. Spinel-type LixMn2O4 cathode materials have been proposed as an alternative to the more expensive LiCoO2 cathode materials for Li-ion batteries. However, the spinel structure distorts when operated below 3.0 V vs. Li|Li+ due to undesired Jahn-Teller effect [1], lowering its stability and precluding its deployment and commercialisation. Adding Ti to replace a ‘y’ amount of Mn in the spinel, i.e. LixMn2-yTiyO4, has been proposed to stabilise the spinel structure below 3.0 V vs. Li|Li+, allowing a specific capacity of 308 mAh g-1 (for y=1) thanks to existence of both Mn4+|Mn3+ and Mn3+|Mn2+ transitions [2]. However, stability issues persist, and specific capacities decrease with the amount of electrochemically inactive Ti. The aim of this work is to optimise the stoichiometry of LixMn2-yTiyO4 by varying the amount of Ti used in a modified sol-gel method [2]; and to study their electrochemical performance by electrochemical impedance spectroscopy (EIS), cyclic voltammetry and charge/discharge cycling with hopes to understand underlying degradation processes during cycling, and propose suitable strategies to diminish their effects on Li-ion batteries. Additionally, cathode materials were characterised by XRD and XPS before and after cycling, along with Raman, SEM and BET surface area.XRD and Raman confirmed a successful synthesis of spinel-type LixMn2-yTiyO4 cathode materials. Traces of TiO2/rutile were observed only in the XRD patterns for stoichiometry y=1 (LiMnTiO4). In all other cases, the shape and intensity of the peaks in the XRD pattern confirmed a correct crystalline structure. SEM showed that nanometric particles (ca. 100 nm) were synthesised in all cases. XPS showed a decrease in the intensity of the Mn 2p and Ti 2p cores lines before and after 20 cycles, possible due to leaching or loss of active material during cycling.The stoichiometry LiMn1.8Ti0.2O4 showed the best initial specific capacity, ca. 215 mAh g-1 as seen in Fig. 1a. A pronounced decrease of the specific capacity during cycling was evident in all cases when the operating potential window was 4.6 – 2.0 V vs. Li|Li+. However, when the potential window was narrowed to 4.6 – 3.0 V vs. Li|Li+, the cathode’s initial specific capacity was halved, although stability improved considerable during 50 cycles. This is a clear indication of undesired structure modifications when the transition Mn3+|Mn2+ occurs, which can be seen the voltammogram in Fig. 1(b) and it is confirmed by EIS spectra, shown in Fig. 1(c), by a considerable increase of the total impedance only after 5 cycles.