【作者】 蔡永茂；

【导师】 陈岗；

【作者基本信息】 吉林大学 ， 材料物理与化学， 2011， 博士

【摘要】 过渡金属氧化物属于强关联电子体系,在这些体系中,晶格点阵、自旋、电荷、与轨道自由度之间存在着强的耦合作用,使得过渡金属氧化物展现了大量奇特的物理、化学性质,如在掺杂的莫特绝缘体中发现的高温超导与庞磁阻效应,在YMnO₃、TbMnO₃、TbMn₂O₅、LuFe₂O₄等氧化物中发现的多铁性。这些过渡金属氧化物由于存在多自由度的耦合,可以通过电场、外应力等控制材料的磁化,也可以通过磁场、外应力等控制材料的极化,因此具有重要的应用潜力与价值。目前基于密度泛函理论,形成了许许多多流行的量子模拟软件包,如CASTEP、VASP、PWscf、CPMD、GAUSSIAN等,构成了第一性原理方法研究的重要工具,被广泛的应用于凝聚态物理、材料科学、半导体、化学领域,并取得的了巨大成功。本论文使用CASTEP软件包来研究过渡金属氧化物的电子结构与性质。选择高压调制的结构相变：以4d4低自旋电子体系碱土钉氧化物SrRuC₃以及BaRuO₃为例,研究结构相变以及对材料磁性的影响；温度调制的结构相变：以混价氧化物尖晶石A1V₂O₄菱方相为例,研究低温混价过渡金属氧化物的电荷歧化以及磁性；以及锂离子调制下的材料电化学性质：选择锂离子电池正极材料,以新型锂离子正极材料LiMSO₄F （M=Fe、Co、Ni）为例,研究锂离子的脱出对材料晶体结构与电子结构的影响,以及它们的电化学性质。高压调制的结构相变：SrRuO₃, BaRuO₃与CaRuO₃是化学组成和结构相关的体系,由于占据ABO₃中A位置的Ca²⁺与Sr²⁺的离子半径比较小（容忍因子j小于1）,通常常压下CaRuO₃和SrRuO₃均为正交钙钛矿相。2007年,在高压下（压强为21-25GPa,温度为1173-1473K）,Akaogi M.研究组实验上观测到CaRuO₃转变为后钙钛矿相,而对SrRuO₃体系,实验上当压强增加到34GPa,Hamlin等人仍没有发现后钙钛矿结构相变。BaRuO₃由于A位置的Ba²⁺离子半径比较大（容忍因子略大于1）,在常压下通常为六方结构,在不同的压强条件下,合成出了菱方结构（9R）,四层六方（4H）,六层六方（6H）,立方钙钛矿（3C）结构。本论文首先通过采用第一性原理的方法,研究了ARuO₃（A=Sr, Ba）高压下结构的相变以及此过程中的材料磁性的变化。通过计算研究分析得出以下主要结论：（1）SrRuO₃正交钙钛矿相的结构畸变程度随着压强的增大而变大,并且在40GPa的静水压下,发生后钙钛矿结构的相变。SrRuO₃结构相变伴随着系统体积的突变以及磁性的转变。SrRuO₃在钙钛矿结构中为巡游铁磁,而在后钙钛矿中变为了非磁金属。（2）在BaRuO₃体系中,证实了BaRuO₃在高压下顺磁性的六方相到铁磁性的立方相的相变,随着压强增加体系先后经历了六方9R、4H、6H,最后到立方3C的结构相的转变,并且温度在此结构相变中起到了非常重要的作用,并预测了BaRuO₃在6H与3C相的相变边界的克拉伯龙斜率为负值,由此解释了6H相为顺磁态,而3C为铁磁态。温度调制的结构相变：尖晶石结构AlV₂O₄在温度为700K时发生导电率与磁化率异常,伴随着结构从正方尖晶石相变成了菱方相。这个结构相变被认为电荷序的转变。然而对于低温这个电荷序,两组科学家分别提出不同模型（Matsuno等提出了three-one型电荷序模型,以及Horibe等提出了V的七聚体自旋佩尔斯态模型）。对AlV₂O₄奇特的低温结构（V七聚体）相的电子结构,仍让人有些难以理解。因此论文第二部分通过第一性原理详细计算了菱方相尖晶石AlV₂O₄的电子结构,定量分析了V离子上的电子布居,研究了体系中V的电荷歧化。计算优化AlV₂O₄菱方相,得到了实验上发现的V的七聚体结构。定量分析AlV₂O₄菱方相的电子结构发现,在菱方相中V原子沿c轴分别形成V1,V2以及V3面,V原子形成了V1（2.5-δ1）+-V3（2.5+（δ1-δ2/6））+-V2（2.5+δ2）+-V3（2.5+（δ1-δ2/6））+-V1（2.5-δ1）+层状的电荷序。然而,关于AlV₂O₄菱方相绝缘体的本质,仍然需要进一步的深入研究。锂离子调制的电子结构变化（材料的电化学性质）：LiFePO₄由于价格低廉、高安全性以及无毒性,曾被认为锂离子动力电池理想的正极材料。然而其本身的缺陷（低导电率,制备非常复杂以及低温性能不够理想）也很难解决。最近Tarascon J-M研究组通过引进氟原子以及采用[SO₄]2-代替[PO₄]3-成功制备出了氟代聚阴离子型正极材料LiFeSO₄F。这个新材料显示的电压平台为3.6V（vs.Li/Li+）略高于LiFePO₄的,其比容量为130 mAhg-1,而且它的离子电导率是LiFePO₄的103倍。LiFeSO₄F的发现不仅是LiFePO₄强有力的竞争者,而且也暗示了一类新的氟代聚阴离子型正极材料。随后Tarascon J-M研究组又制备出了Li（Fe1-xMx）SO₄F （M= Co, Ni, Mn）,然而LiMSO₄F （M=Co,Ni, Mn）体系在2.5V到4.2V电压区间没有电化学活性。锂离子电池电极材料的电化学以及物理性质与材料的结构与电子结构紧密关联。因此,为了深入理解这类新型的氟代聚阴离子型正极材料的电化学性质,论文采用第一性原理方法计算了材料的晶体结构与电子结构,并定量分析了L1MSO₄F体系随着Li离子的脱出电子的转移情况,研究分析得到如下结论：（1）预测了LiNiSO₄F的晶体学数据。（2）计算得到了LiMSO₄F （M=Fe, Co, Ni）的平均插入电压——Fe²⁺/Fe3+：3.54V、Co²⁺/Co3+：4.73V、Ni²⁺/Ni3+：5.16V,成功解释了LiCoSO₄F与LiNiSO₄F体系在2.5 V到4.2V电压区间没有电化学活性。（3）对LiCoSO₄F与LiNiSO₄F体系来说,更多的电子转移（56%与57%）是来源于O的2p能带,比LiFeSO₄F的（28%）多出很多。这种从O的2p能带上转移如此多的电荷,会产生过氧根离子O-,过氧根离子的形成往往在表面产生氧气的释放,导致晶格的塌陷,引起安全问题。（4）随着锂离子的脱出,体系由LiFeSO₄F莫特-哈伯德（MH）绝缘体转变为FeSO₄F电荷转移（CT）绝缘体,而在LiCoSO₄F与LiNiSO₄F体系中并没有发现此类转变。然而这种MH到CT的转变的物理机制还不是很清楚,需要进一步深入地研究。更多还原

【Abstract】 Transition metal oxides belong to the strongly correlated electronic systems. In these sytems, the lattice, spin, charge and orbital degrees of freedom are still active and strongly coupled each other, which make the transition metal oxides exhibit abundant peculiar physical and chemical properties, such as, the high transition temperature superconductivity in the cuprate, the colossal magnetoresi stance in the manganite, the multiferroics BiFeO₃, YMnO₃, TbMnO₃, TbMn₂O₅, LuFe₂O₄ etc.Due to the strongly coupling among the degrees of freedom, the magnetization can be controlled by the electric field or stress, and in turn the polarization could also be controlled by the magnetic field or stress. So, these materials have valuable potential applications.At present, there are many quantum simulation software packages, based on the density functional theory, such as CASTEP, VASP, PWscf, CPMD, GAUSSIAN etc. These packages consist of very important tools of the first-principles methods, which have been largely applied in the condensed matter physics, materials science, semiconductor, chemistry and made a great sucess. The dissertation mainly discusses the electronic structure and properties of transition metal oxides via first-principles calculation with CASTEP code. The dissertation selects the pressure induced phase transition:taking the 4d4 low spin state systems SrRuO₃ and BaRu3 for example to study the structure phase transition and the effects on the magnetic properties; temperature induced phase transition, taking the spinel AIV₂O₄ for example to study the charge disproportionation and the magnetic properties of the mixed-valence transition metal oxides at low temperature; finally select lithium ions induced the changes of electronic structure （electrochemical properties of cathode materials）: taking the new cathode materials fluorosulfate LiMSO₄F （M=Fe, Co, Ni） for example to study their electrochemical properties and the effects of lithium ion extraction on the crystal structure and electronic structure of cathode materials.Pressure induced phase transition:SrRuO₃, BaRuO₃ are chemical composition and crystal structure related to CaRuO₃. For the ionic radii of Ca²⁺and Sr2÷are small （the tolerance factor t<1）, CaRuO₃ and SrRuO₃ usually display orthomibic perovskite phase. In 2007, Akaogi M. research group discovered that the orthomibic perovskite structure of CaRuO₃ becomes into post-perovskite structure under high pressure （Pressure:21-25GPa, Temperature:1173-1473K）. As for SrRuO₃, Hamlin J J et al. did not found the post-perovskite phase transition up to 34 GPa experimentally. Due to the large ionic radius of Ba+（the tolerance factor t>1）, BaRuO₃ crystallizes as hexagonal polytypes. Sintering under high pressure,9R,4H, 6H and 3R polytypes can be gained in turn. The dissertation first studied the phase transition of ARuO₃（A=Sr, Ba） under high pressure by first-principles calculations. The calculations results mainly show as follows:（1） The structural distortion of orthorhombic SrRuO₃ perovskite is enhanced with increasing pressure. And it undergoes phase transition to post-perovskite structure at 40GPa. The SrRuO₃ post-perovskite phase transition companies with the discontinuous volume contraction and collapse of the magnetism. （2） In BaRuO₃, it was confirmed that BaRuO₃ undergoes phase transition from paramagnetic hexagonal phase to ferromagnetic cubic phase. With increasing pressure, BaRuO₃ undergoes 9R,4H,6H hexagonal phase and 3R cubic phase in sequence. The temperature plays an important role during these phase transitions. It was showned that the negative Clapeyron slope appears boundary between 6H-BaRuO₃ and 3C-BaRuO₃, which explains paramagnetic behavior of 6H-BaRuO₃, and while ferromagnetic behavior in 3C-BaRuO₃.Temperature induced phase transition:The AIV₂O₄ shows a phase transition at about 700K with anomalies of transport and magnetic properties, which was considered as a charge ordering transition. However, two research groups proposed different models （Matsuno et al.:three-one type charge odering and Horibe et al.: spin singlet of V heptamer）. But it is still hard to understand the electronic structure of the V heptamer. So, in the second part of the dissertation, I studied the charge disproportionation of V ions in the rhombohedral spinel AIV₂O₄ by analysizing the electron population of V ions via first-principles calcution. The V heptamer structure was also gained by optimizing the experimental crystal structure of AIV₂O₄. The calculated results indicate that charge disproportionations take place in these V ions with valence states of+（2.5-δ1）,+（2.5+δ2） and+（2.5+（δ1-δ2）/6） （δ1>δ2>0）, respectively, and form a charge ordering along the c-axis direction layer by layer with a sequence as V1-V3-V2-V3-V1. However, the nature of the insulating state of AIV₂O₄ is still mysterious and needs further study.Lithium ions induced the changes of electronic structure （Electrochemical properties of cathode materials）:LiFePO₄ was once considered as the promising cathode material for lithium-ion batteries for vehicles, due to the low price, high safety and non-toxic properties. However, the drawbacks of LiFePO₄ （low conductivity, complexity of the material synthesis, bad low-temperature performance） are also very hard to resovle. Recently, Tarascon J-M group synthesized a new fluorosulfate cathode material LiFeSO₄ F by introducing a fluorine atom and replacing the [PO₄]3+ group of LiFePO₄ with [SO₄]2-. The material shows a voltage plateau at 3.6 V vs. Li/Li+, with a reversible specific capacity of 130 mAhg-1. It was found that the ionic conductivity of LiFeSO₄F is about 103 times higher than that of LiFePO₄. The discovery of LiFeSO₄F not only provides a strong competitor of LiFePO₄, but also suggests a new class of fluoro-oxyanion cathode material for lithium ion batteries. Subsequently, Tarascon J-M group synthesized Li（Fe1-xMx）SO₄F （M= Co, Ni, Mn）, they found that LiMSO₄F （M=Co, Ni, Mn） system did not show any electrochemical activity, cycling between 2.5 V and 4.2 V. At microscopic scale the electrochemical and physical properties of electrode materials are strongly correlated to their electronic structures. Therefore, we purpose to calculate the crystal and electronic structures of LiMSO₄F （M= Fe, Co, and Ni） as well as their delithiated forms using first-principles calculations, in order to present a deep understanding on the electrochemical and physical properties of the LiMSO₄F/MSO₄F systems. The main results are as follows （1） The crystallographic data of LiNiSO₄F was given. （2） The theoretical intercalation voltages of LiMSO₄F are 3.54 V （Fe）,4.73 V （Co） and 5.16 V （Ni）, respectively, which are close to corresponding LiMPO₄ phosphates, which explains none electrochemical activity of LiMSO₄F （M=Co, Ni）, cycling between 2.5 V and 4.2 V. （3） The values of electron-charge transfer taking place on the oxygen anions are 56 % and 57 % for LiCoSO₄F/CoSO₄F and LiNiSO₄F/NiSO₄F respectively, which are much larger than that of LiFeSO₄/FeSO₄F（28%）. The removal of such a lot of electrons from the O-2p band will result in a large amount of O- anions. This process is accompanied with the following peroxide formation, leading to ultimate loss of oxygen from the lattice surface, which may cause lattice collapse and safety problems. （4） LiFeSO₄F transforms from Mott-Hubbard insulator to charge-transfer insulator with Li+extraction. However, this transformation does not happen in L1CoSO₄F and LiNiSO₄F systems. The physical mechanism of this transformation is still unclear, which is worth of study in future.更多还原