锂离子电池具有能量密度高、循环寿命长、安全可靠等优点，广泛应用在便携式电子设备中，并逐渐向电动汽车、新能源储能领域拓展。新能源汽车等行业的快速发展对锂离子电池的能量密度提出了更高的要求。目前商业用的负极石墨具有较低的理论比容量(372 mA h g-1)，已很难满足锂电池对能量与功率密度日益增长的要求。硅的理论储锂容量为4200 mA h g-1，电压平台适中，有望成为新型锂离子电池的负极材料。然而，硅电导性差且充放电过程中伴随着巨大的体积效应，易造成容量很快衰减，这些缺点严重限制了其商业化应用。针对以上问题，本论文开展了等离子体制备高分散纳米硅及其在锂离子电池负极应用的研究。主要研究内容如下：（l）采用热等离子体法制备了粒径分布在50~100 nm、形貌可控、分散性良好的零维纳米硅球（0D-SiNS），其在溶液和树脂中均保持均匀分散。通过热力学和生长动力学分析反应因素对晶体形貌和粒径大小的影响规律，认为等离子体中硅晶体的生长遵循气固成核机理：晶体生成包括过饱和、成核和生长三个阶段。在生长阶段，通入冷却气，提高冷却速率有利于得到分散性高，表面光滑致密的纳米硅球。（2）将等离子体法制备的纳米硅球用作锂离子电池的负极材料，对其电化学性能进行了深入的研究。纳米硅球在电极中能保持良好的分散性和结构稳定性，完全嵌锂时无破裂，体积仅膨胀270%。纳米硅球良好的分散性和稳定性使其循环性能获得了较大的提升，首次比容量高达2388 mA h g-1，首次库仑效率为70%，5次循环后库伦效率能稳定在99%。循环50次后，比容量稳定在500 mA h g-1左右，是石墨碳理论比容量的1.3倍，远远优于体硅的循环性能。（3）对等离子体制备的纳米硅球进一步改进，通过引入多孔碳，得到硅球分散良好的硅碳复合材料（SiNS/PC）。多孔碳包裹在硅球表面，形成封闭的核壳结构，碳层厚度约为10 nm。相比纳米硅球，SiNS/PC作为电池负极有两个主要优点：一是丰富的孔隙结构能够吸收硅嵌锂时的体积膨胀，减少体积效应；二是避免硅颗粒与电解液的直接接触，形成稳定的SEI膜。循环过程中，SiNS/PC颗粒平均粒径由43 nm膨胀至52 nm，体积膨胀率仅为174%。其循环性能也有了很大的提高，首次比容量提高为2510 mA h g-1，100次循环后，比容量是778 mA h g-1，是石墨碳理论比容量（372 mA h g-1）的2.1倍。同时，硅碳复合材料表现出了较优的大电流循环稳定性，在电流密度为4200 mA g-1时，平均比容量为445 mA h g-1，是石墨碳理论比容量的1.2倍。（4）鉴于球磨法得到的复合材料硅/碳结合强度低，循环后期脱落严重。研究中进一步利用喷雾干燥法制备了三维立体球形硅/碳（3D-SiNS/C）复合材料，进一步提高硅/碳在循环过程中的结合强度和结构稳定性。3D-SiNS/C颗粒平均粒径3 μm， Si/C微球（~50 nm）构成二次颗粒的骨架结构，多孔碳作为粘结剂和导电剂，增加二次颗粒的结合强度，并构成三维导电网络。循环过程中，骨架颗粒体积膨胀仅为68%，二次颗粒体积几乎没有明显变化。其首次比容量为2059 mA h g-1、首次库伦效率为88%。整体循环曲线平稳，循环50次后比容量保持在1500 mA h g-1，是石墨碳的4倍，容量保持率为73%。硅基负极材料的循环稳定性得到进一步提高。（5）相比纳米硅球，硅纳米线具有轴向降低体积膨胀和径向锂快离子传输通道的优势。通过强化等离子体高温区保温时间制备了直径30~50 nm、长为几百纳米至微米级的一维纳米硅线。并以其为骨架颗粒得到了三维立体线团结构的硅碳复合材料（3D-SiNW/C）。在循环过程中，三维线团结构通过纳米硅线之间的孔隙变化吸收体积膨胀，整体结构在长时间脱嵌锂条件下（300次）能够保持稳定。虽然3D-SiNW/C首次比容量并不很高，为1206 mA h g-1，但具有较好的首次库伦效率（高达94%）和容量保持率，循环300次容量维持在620 mA h g-1以上。其中，循环100~300次，容量几乎没有明显的衰减，容量保持率在87%以上。

Li-ion batteries (LIBs) have many advantages such as high energy density, long cycle life, safe and reliable. They have been widely applied in the fields of portable electronic devices, gradually developing to electrocars or energy storage. High energy density LIBs become one of the important development directions. Traditional LIBs anode material of graphite (theoretical specific capacity, 372 mA h g-1), is difficult to meet the ever-growing demands for high energy and power density. For anode materials, silicon stands out as the most promising material for the next generation LIBs due to its known highest theoretical capacity (4200 mA h g-1) and appropriate voltage platform. However, silicon as anode material is severely hindered by low electric conductivity and the huge volume changes during lithium insertion/extraction process, which cause the silicon particles dramatically pulverized and eventually lead to the capacity rapidly fading. These disadvantages severely limit the application. To solve the above problems, highly dispersed nano silicons were prepared by plasma and their electrochemical performance for lithium-ion batteries were researched in this dissertation.The main innovative results are listed as follows:(l) Morphology controlled 0D silicon nanospheres (0D-SiNSs) were synthesized by RF plasma with a size distribution of 50~100 nm. They had good dispersion and could be well dispersed in solution and resin. By analysis the influence of reaction parameters on crystal morphology and particle size through thermodynamics and growth kinetics, the Si crystals growth in plasma were thought to the VS nuclear mechanism and including three stages: supersaturated, nucleation and growth. In the growth stage, high cooling rate was helpful to gain well dispersed, smooth, compact SiNSs.(2) Silicon nano-spheres synthesized by RF plasma were used as anode for LIBs and the electrochemical performance was studied in detail. The SiNSs kept good dispersion and stability as electrode, and exhibited no break at fully discharged with a volume change of 270%. Good dispersion and stability made its electrochemical performance improved a lot with high specific capacities and cycle stability. For SiNS electrode, the initial specific capacities was 2388 mA h g-1, the initial coulombic efficiency (ICE) was 71%. After 5 cycles, the CE (coulombic efficiency) stabilized at about 99%. Importantly, after 50 cycles, the capacity of SiNS electrode was still 500 mA h g-1, which is greatly superior than that of bulk silicon.(3) To further improve the electrochemical properties of SiNSs synthesized by RF plasma, porous carbon (PC) was introduced and SiNS/PC composites were synthesized, with Si nanospheres dispersing uniformly in carbon matrix. PC wrapped in the surface of Si and formed a closed core-shell structure. The thickness of carbon layer is ~10 nm. Compared with SiNSs, the obtained composite materials had two main advantages: abundant pore structure supplied room to absorb the volume expansion and relieve the volume effect; Core-shell morphology could avoid the immediate contact of SiNS and electrolyte. During cycles, the particles expanded from 43 nm to 52 nm, with a volume change of only 174%. Therefore, its electrochemical performance exhibited remarkable improvement with an initial specific capacities of 2510 mA h g-1. Importantly, the SiNS/PC electrode still exhibited high capacity of 778 mA h g-1 after 100 cycles, corresponding to 2.1 times that of graphite anode. Also, SiNS/PC composites exhibited super rate capacity and could remain stable at high current density. Even at 4200 mA g-1, the capacity reached to 445 mA h g-1, corresponding to 1.2 times that of graphite anode.(4) 3D porous and spherical Si/C micro-/nano-spheres were further designed and fabricated by spray drying procedure to improve the structure stability. The SiNS/C composite exhibits spherical shape with a diameter of 3 mm. ~50 nm Si/C microsphere constituted the skeleton structure, and the carbon matrix acted as binder or conductor, enhanced the mechanical stability and made a 3D conductive network that improves the overall conductivity. During cycles, ~50 nm Si/C sphere only experienced a small volume variation (~68%) and the secondary particles were almost no obvious change. The obtained composite materials exhibited ultra-high ICE of 88%, improved capacity of 2059 mA h g-1. The overall cycle curve was steady and the capacity kept at 1500 mA h g-1 after 50 cycles, and the retention rate was 73%. The cycle performance of the silicon anode materials were further improved.(5) Compared with SiNSs, silicon nanowires (SiNWs) had the advantage of reduced volume change in axial and improved Li-ion transportation in radial direction. 1D SiNWs were fabricated by strengthening the temperature holding time of plasma, with a diameter of 30~50 nm and length of few hundred nanometers to microns. On this basis, 3D porous wool-ball like SiSW/C spheres were designed. During cycles, the volume changes of Si were obsorbed by the pores between SiNWs, and the microstructure could keep stable after long cycles (300 cycles). As anode for LIBs, the specific capacity was 1206 mA h g-1, with an ICE of 94%, and the capacity kept at 620 mA h g-1 after 300 cycles. The cycling curves tended towards stability and the capacity retention is about 87% from 100th cycle to 300th cycle.