Extensive Studies On Materials For Lithium-ion Batteries

Our ceaseless demands for energy sources have greatly promoted the development of energy storage devices. As the most successful energy storage device invented in the past two decades, Li-ion batteries have dominated the portable electronic market since first commercialized by Sony in 1990. Meanwhile, with the development of high performance of central processing unit in laptops and the use of 3D techniques in cellular phones, people never stop their steps to seek higher-power and longer-life batteries, which lead to the study of Li-ion batteries to be a hotspot in materials science. This Ph.D thesis includes the synthesis of cathode materials (LiCoO₂,LiMn₂O₄,LiFePO₄ and LiNi_0.5Mn_1.5O₄), the improvement and optimization of electrode materials (nominal "Li_1±xCoO₂" and Si), the exploration of new vanadium-based electrode materials and the study of storage mechanism of lithium in transition metal oxides. Besides, the effect ofγ-ray radiation on Li-ion batteries is also investigated.In Chapter 1, a general introduction is given on following aspects: the development and status of the Li-ion batteries, their working principle, three important cathode materials (LiCoO₂, LiMn₂O₄ and LiFePO₄) and three common anode materials (graphite, Li₄Ti₅O₁₂ and Si), the methods of electrode synthesis and approaches to improve their performance (mainly by coating or doping).In Chapter 2, we briefly introduce the experimental processes and equipments used in the project of this thesis. A detailed description on the process of making a coin cell is presented. The electrochemical and structural analyses methods are also included.In Chapter 3, we invent a new synthesis method named "Radiated Polymer Gel" method to prepare LiCoO₂ and LiMn₂O₄ powders. This method is developed from my experience in "Undergraduate Research Program, USTC", where I have investigated the polymerization process of acrylic acid. Herein, I expand it to the synthesis of inorganic materials. In comparison with the traditional "Sol-gel" method, our method has some advantages including: fast (time-consuming, usually less than 1 h), easily controllable (not necessary to control the temperature and pH value during gel process), potential of massive production and easy extension to other materials.With regard to the recent research hotspot of LiFePO₄, in Chapter 4 we invent a new route for LiFePO₄ synthesis. In this method, we use Fe as the starting material and formic acid as a solvent. Fe is dissolved in formic acid and H₂ gas is released, which can protect Fe²⁺ ions against further oxidation. In comparison with the usual synthesis methods, we directly use a Fe²⁺ salt as the precursor with any possible impurities can be removed by calcination. Thus during the synthesis, a washing step is not needed.As the working voltage of LiCoO₂ and LiMn₂O₄ is only about 4 V, the 5 V cathode material "LiNi_0.5Mn_1.5O₄" can deliver a energy density of 30% higher than that of LiCoO₂ or LiMn₂O₄,and can be considered as a potential cathode material for EV/HEV applications. In Chapter 5, we improve the traditional co-precipitation method, using chloride as starting materials and ammonia as precipitator, then the chloric impurities are removed by the thermal decomposition of NH4Cl so that we finally obtain LiNi_0.5Mn_1.5O₄ with spinel phase (with octahedral crystalites) and a narrow size-distribution (about 2μm). In comparison with the traditional co-precipitation method, we don't need to wash the product, thus the final composition can be exactly stoichimetric.In Chapter 6, we investigate the properties of nominal Li_x0CoO₂ (X₀=0.8, 0.9, 1.0, 1.1 and 1.2), and find a Li₂CO₃ layer on the over-stoichimetric Li_x0CoO₂ (X₀=1.1, 1.2), which can affect the cyclability of Li_1.1CoO₂ and Li_1.2CoO₂ under the low voltage range, yet improve the cyclability under the high voltage range by avoiding the decomposition of electrolyte on Li_x0CoO₂.The conductance study of over-stoichimetric Li_x0CoO₂ (X₀=1.1 and 1.2) reveals that the conductance change doesn't follow with a semiconducting behavior with the increase in temperature, whereas the stoichimetric and under-stoichimetric Li_x0CoO₂ (X₀=0.8, 0.9 and 1.0) behave as semiconductors, and have the maximum conductance at around 100℃. Besides, we also derive the diffusion coefficient of Li-ion (D_Li⁺) in Li_xCoO₂ by A.C. impedance method, and find that the D_Li⁺ values vary in the range from 10^-13 to 10^-8cm²s^-1.In Chapter 7 and Chapter 8, we extensively investigate Si as anode mterial for Li-ion battery. Silicon working as an anode for Li-ion batteries has attracted much attention thanks to its very high capacity (-4200 mAh g^-1). However, due to the large volume expansion during lithiation, the capacity of silicon fades very fast. In Chapter 7, we focus on the issue to fight the capacity fading. Results show that Si with sodium carboxymethyl cellulose (Na-CMC) as a polymer binder exhibits a better cyclability than that with poly(vinylidene fluoride) (PVDF). Yet different from the system using PVDF, the addition of vinylene carbonate (VC) does not improve or even worsens the performance of the system using Na-CMC. In addition, the small particle size of Si, a large amount of carbon black, the good choice of electrolyte/conducting salt and charge-discharge window also play important roles to enhance the cyclability of Si. It is found that electrode consisting of 40 wt.% nano-Si, 40 wt.% carbon black and 20 wt.% Na-CMC (pH=3.5) displays the best cyclability, and in the voltage range from 0 to 0.8 V, after 200 cycles, its capacity can still keep 738 mAh g^-1 (C/2, in 1M LiPF₆ ethylene carbonate/diethyl carbonate electrolyte, with VC free), almost twice as that of graphite. In Chapter 8, the chemical diffusion coefficients of Li⁺ ions (Du⁺) in nano-Si are determined by cyclic voltammetry (CV), electrochemical impedance spectroscopy (EIS) and galvanostatic intermittent titration technique (GITT). D_Li⁺ values are estimated to be -10^-12 cm² s^-1 and exhibit a "W" type variation with the Li⁺ ion concentration in silicon. Two minimum regions of D_Li⁺ (at Li_2.1±0.2Si and Li_3.2±0.2Si) are found, which probably result from two amorphous compositions (a-Li₇Si₃ and a-Li₁₃Si₄).In addition to the two minimum regions, one maximum D_Li⁺ is observed at Li₁₅Si⁴,corresponding to the crystallization of highly lithiated amorphouse Li_xSi.We also observe that the volume expansion of Li_xSi particles during lithiation is not linearly related to x and the particles are even degraded into srnaller particles at Li_3.75Si during the crystallization. In Chapter 9, we study the effects of y-radiation on lithium-ion cells in following two aspects: electrode and electrolyte. The radiation can cause oxygen vacancies in crystal lattice and affect the Li-ion transportation. The radiation can also lead to the decomposition of LiPF₆ in the electrolyte into LiF and PF₅.PF₅ is a strong Lewis acid and can further induce the polymerization of ethylene carbonate (EC) forming a PEO-like polymer, which can be clearly observed from the color change of electrolyte (from colorless into brown). To avoid such effect, we suggest that the conducting salt in the electrolyte should be LiBOB instead of LiPF₆.Besides, carboxyl is also observed in the radiated electrolyte confirmed by IR and NMR spectra. As a result, there is an extra voltage plateau measured for the full cell at 1.75 V, which comes from the reaction between carboxyl and Li and is confirmed by thermodynamic calculation. In the full cell test, we find the cell shell can not shield the radiation effects and after radiation, the capacity of a half cell (using Li as anode) fades for about 20% and, for a full cell, the capacity fading even reaches 50%.In Chapter 10, we explore the new vanadium-based electrode materials. By simply changing the composition of starting materials, we can obtain vanadium oxides with different morphologies: nanowires, nanosheets, nanoneedles and nanoparticles. Their structures are characterized by a series of techniques in TEM, such as EDX, EELS and ED patterns. Results show that the sheet-like VO₂·0.43H₂O synthesized at 180℃exhibits a significant superiority, with high capacity (about 160 mAh g^-1), high energy efficiency (95%) and excellent cyclability (keeps 142.5 mAh g^-1 after 500 cycles). Electrochemical kinetics study reveals that the activation energy of VO₂·0.43H₂O is lower than other electrode materials. The low activation energy leads to that VO₂·0.43H₂O can keep quite good C rate retention under the high current density. We believe that VO₂·0.43H₂O can be considered as a potential candidate cathode material to design the safe and high energy density batteries for EV/HEV.In Chapter 11,we systematically investigate the electrochemical reactions of lithium with metal oxides. A series of remarkable phenomena are observed, which can not be explained by the conversion reaction mechanism. For example, the voltage profile of Ga₂O₃ does not show the characteristic Li/Ga alloying process. Moreover, by a simple, well-designed experiment, we found that lithium storage in CoO is not through the formation of Li₂O as proposed by the conversion mechanism, but instead it proceeds through the formation of some sort of intermediate lithiated MO state, which is an active lithium source that can react with N₂.In addition, we have also studied MnO₂, which shows a distinctive reduction process. An intermediate state formed by charge transfer between metal oxide nanoparticles and lithium, which can reversibly store lithium in metal oxides, has been proposed. We anticipate that such a new understanding of lithium storage mechanism in metal oxides may guide the further exploration of the lithium-related rechargeable batteries and super-capacitors.In Chapter 12, we use transmission electron microscope to investigate the oxidation process of Fe/CoO nanocrystals and the final product of V₂O₅ after reduction by ethanol.Finally, the author gives an overview on the achievements and the deficiency in this thesis. Some prospects and suggestions of the possible future research directions are pointed out.