Chemical, structural, and electrochemical characterization of 5 V spinel and complex layered oxide cathodes of lithium ion batteries

Lithium ion batteries have revolutionized the portable electronics market since their commercialization first by Sony Corporation in 1990. They are also being intensively pursued for electric and hybrid electric vehicle applications. Commercial lithium ion cells are currently made largely with the layered LiCoO₂ cathode. However, only 50% of the theoretical capacity of LiCoO₂ can be utilized in practical cells due to the chemical and structural instabilities at deep charge as well as safety concerns. These drawbacks together with the high cost and toxicity of Co have created enormous interest in alternative cathodes. In this regard, spinel LiMn₂O₄ has been investigated widely as Mn is inexpensive and environmentally benign. However, LiMn₂O₄ exhibits severe capacity fade on cycling, particularly at elevated temperatures. With an aim to overcome the capacity fading problems, several cationic substitutions to give LiMn[subscript 2-y]M[subscript y]O₄ (M = Cr, Fe, Co, Ni, and Cu) have been pursued in the literature. Among the cation-substituted systems, LiMn[subscript 1.5]Ni[subscript 0.5]O₄ has become attractive as it shows a high capacity of ~ 130 mAh/g (theoretical capacity: 147 mAh/g) at around 4.7 V. With an aim to improve the electrochemical performance of the 5 V LiMn[subscript 1.5]Ni[subscript 0.5]O₄ spinel oxide, various cation-substituted LiMn[subscript 1.5-y]Ni[subscript 0.5-z]M[subscript y+z]O₄ (M = Li, Mg, Fe, Co, and Zn) spinel oxides have been investigated by chemical lithium extraction. The cation-substituted LiMn[subscript 1.5-y]Ni[subscript 0.5-z]M[subscript y+z]O₄ spinel oxides exhibit better cyclability and rate capability in the 5 V region compared to the unsubstituted LiMn[subscript 1.5]Ni[subscript 0.5]O₄ cathodes although the degree of manganese dissolution does not vary significantly. The better electrochemical properties of LiMn[subscript 1.5-y]Ni]subscript 0.5-z]M[subscript y+z]O₄ are found to be due to a smaller lattice parameter difference among the three cubic phases formed during the chargedischarge process. In addition, while the spinel Li[subscript 1-x]Mn[subscript 1.58]Ni[subscript 0.42]O₄ was chemically stable, the spinel Li[subscript 1-x]Co₂O₄ was found to exhibit both proton insertion and oxygen loss at deep lithium extraction due to the chemical instability arising from a overlap of the Co[superscript 3+/4+]:3d band on the top of the O[superscript 2-]:2p band. The irreversible oxygen loss during the first charge and the consequent reversible capacities of the solid solutions between Li[Li[subscript 1/3]Mn[subscript 2/3]]O₂ and Li[Co[subscript 1-y]Ni[subscript y]]O₂ has been found to be determined by the amount of lithium in the transition metal layer of the O3 type layered structure. The lithium content in the transition metal layer is, however, sensitively influenced by the tendency of Ni[superscript 3+] to get reduced to Ni[superscript 2+] and the consequent volatilization of lithium during synthesis. Moreover, high Mn4+ content causes a decrease in oxygen mobility and loss. In addition, the chemically delithiated samples were found to adopt either the parent O3 type structure or the new P3 or O1 type structures depending upon the composition and synthesis temperature of the parent samples and the proton content inserted into the delithiated sample. In essence, the chemical and structural stabilities and the electrochemical performance factors of the layered (1-z) Li[Li[subscript 1/3]Mn[subscript 2/3]]O₂ · (z) Li[Co[subscript 1-y]Ni[subscript y]]O₂ solid solution cathodes are found to be maximized by optimizing the contents of the various ions.