锂离子电池合金负极材料的理论设计和合成

【作者】 张敬君；

【导师】 夏永姚；

【作者基本信息】 复旦大学 ， 物理化学， 2008， 博士

【摘要】 石墨材料作为锂离子电池负极材料,其利用率已经达到了它的极限（372mAh/g）,开发新一代具有更高容量的锂离子电池已经成为当前锂离子电池领域研究的热点。金属和合金被认为是具有潜力的锂离子电池负极材料,因为它们一般具有较高的容量和合适的充放电电位。如金属Sn,它具有高达990 mAh/g（7200 mAh/L）的理论容量。其质量比容量是石墨的2.7倍,而体积比容量竟是石墨的8.8倍。然而,金属的单质却不能直接用来做锂离子电池的负极材料,主要的原因是它们在与锂离子发生反应生成锂一金属合金的过程中存在较大的体积膨胀和收缩,这将导致材料受内部应力的作用而龟裂,从而失去活性,最终导致材料的循环性能下降。提高金属和合金材料的循环性能成为开发合金锂离子电池材料的关键问题。近年来,对于如何抑制金属材料的体积膨胀和收缩以提高合金材料作为锂离子电池负极材料的循环性能的研究倍受关注。其中一种比较有效的方法是使用金属间化合物（通常称为合金）MM’,其中包含M为“活性金属”,也即在电池环境下,可以与锂发生反应的金属;而M’为“非活性金属”,即在电池环境下不能和锂发生反应的金属。其中Cu₆Sn₅,FeSn₂,Cu₂Sb等就是属于此类合金。该体系之所以能够抑制体积膨胀是因为金属M’起到了一种体积“缓冲剂”的作用,它可以从一定程度上抑制活性金属M和锂离子发生反应产生的体积膨胀,减少材料内部因为体积膨胀而产生的龟裂,防止活性材料和集电体脱离。然而,在开发此类MM’型金属间化合物作为锂离子电池负极材料的研究中遇到许多困难,其中最主要的是:如何选择组成合金的金属M和M’,以及如何调整它们之间的比例。众所周知,在元素周期表中有近十种能与锂合金化的金属（M）,如Al,Si,Zn,Ag,In,Sn,Sb等,还有多种不能与锂合金化的金属（M’）,如Ti,Mn,Fe,Co,Ni,Cu等。它们相互间的结合,根据元素种类不同和组成成分比例不用,将是无限多的化合物。至今为止,合金材料的研究和开发,主要建立在大量的实验数据和个人经验的基础上,适合锂离子用的高性能的合金材料的开发,将花费大量的研究经费和人力。基于密度泛函的第一性原理量子化学计算方法提供了一种只从化合物的元素种类,晶体结构出发,预测其物理、化学性能的有效手段。通过量子化学计算,可以更快,更方便的得出一种材料是否适用于电池材料,许多不能通过常规实验合成或者难于合成的材料,可以用计算来检测其是否具有电化学电池性能。从而指导实验开发,这将给材料的开发带来极大的便利。本课题借助量子化学计算的方法,进行合金材料的设计,选择合适的制备方法,结合电化学和多种物理化学现场表征方法。研究开发结构稳定,膨胀率小,循环性能好的锂离子电池用新型高能量合金材料,阐明锂离子嵌入/脱嵌过程中晶体结构,粉体结构与膨胀率,循环性能之间的关系,在量子计算的基础上开发高性能的锂离子电池用高容量合金材料,将开创一个新的研究领域。本论文第一章将简单介绍锂离子电池以及常见负极材料。第二章介绍了基于第一性原理的密度泛函理论计算方法以及本论文中采用的VASP计算软件包。接着第三章首先介绍量子化学计算在锂离子电池材料研究中的应用。着重讨论如何利用量子化学计算结果来预测材料的电化学性能。然后对元素周期表中常见的合金材料,设计结构和理论模型,进行吉布斯自由能计算。其目的是看它们是否能和锂发生电化学反应,以判断哪些合金适合用于锂离子电池负极材料。通过系统的计算发现,其中Co-Sn合金包含Co₃Sn₂,CoSn和CoSn₂三种化合物,而CoSn₂是一个非常具有开发潜力的合金材料。所以第四章将对它们进行系统的合成和研究。电化学性能测试表明这三种合金,随着Sn含量的增加,脱嵌锂反应变得更加容易,容量也逐渐增大。另外,在利用高能球磨法合成样品时,在原料中添加少量的石墨,可以有效地提高材料的循环性能。通过以上的理论、实验开发和研究,我们发现锂离子电池合金负极材料的电化学性能与组成合金的元素种类及成分,材料的晶体结构,电子结构,以及电荷分布等性质有着密切的关系。在第五章中,为了寻找合金嵌锂容量与晶体结构和元素组成的关系,特地选取了两种晶体结构非常相似的合金材料Cu₆Sn₅和Ni₃Sn₂进行研究。本论文设计实验合成具有单相Ni₂In型六方结构的合金系列Ni_xCu_6-xSn₅（x=0.0,0.5,1.0,2.0,4.0）,利用量子化学计算和实验相结合的方法研究,结果发现随着Ni掺杂含量的增加,合金容量逐渐下降,但循环性能逐渐变好。并且,在锂离子与合金发生锂脱嵌反应过程中,锂离子首先在合金晶包内部的空位进行嵌入,这将导致晶包能量上升,在一定量锂离子的嵌入之后才发生锂和Cu或Ni的取代。在这个过程中,能量的上升相当于反应要越过一个能垒,如果能垒过高,如Ni₃Sn₂,反应将无法进行下去,这就是导致许多合金材料不能发生脱嵌锂反应,容量很低的原因。在第六章中,设计实验合成了合金系列Co_xCu_6-xSn₅（0≤x≤2）,并测试了它们的电化学性能。研究发现适量Co掺杂时,及CoCu₅Sn₅合金,具有最好的电化学性能。从理论计算的角度,本论文从合金材料的晶体结构、热稳定性,以及与锂离子发生脱嵌反应生成的中间产物的电子结构、电荷态密度分布等角度进行了深入探讨,很好的解释了材料电化学性能和其结构的关系。结果表明,合金Co_xCu_6-xSn₅在脱嵌锂反应过程中生成中间相化合物Li₂Co_yCu_1-ySn,并且此中间相化合物随着Co含量的增加,逐渐变得不稳定,从而导致电化学性能发生变化。论文第七章,为了深入研究锂离子在合金内部扩散的动力学性质,采用经典的电化学测试方法（PITT和EIS）研究了Cu₆Sn₅合金中锂离子迁移的扩散系数。结果发现锂离子在合金材料中的扩散并没有想象中的慢,在充放电的大部分电位区间内,扩散系数在10^-11～10^-10cm²/s之间。但是,在充放电平台电位,锂离子扩散非常地慢,扩散系数低于10^-11cm²/s。而且在扩散系数对放电电位作图地曲线中,明显地存在两个扩散系数的极小值,与Cu₆Sn₅放电曲线地两个平台吻合。PITT和EIS两种方法得到的实验数据也非常吻合。这很可能是由于合金中的相变过程阻碍了锂离子的扩散。更多还原

【Abstract】 It has been demonstrated that metals and alloys present an attractive altemative to graphite as anode materials for lithium-ion batteries due in particular to the high capacity, an acceptable rate capability,and operating potentials well above the potential of metallic lithium.For example,Sn yields a maximum theoretical capacity of 990 mAh/g or 7200 Ah/L.However,one major problem preventing them from the commercial application is that they undergo large volume changes during cycling,which result in disintegration of the electrodes and subsequent rapid capacity fading.An interesting approach for overcoming these problems is the use of the intermetallic compounds（or called alloys）,M’M,which consisting of an "inactive phase M’”,referring to the metals that do not react with lithium,and an "active phase M", referring to the metals that react with lithium.The role of"inactive phase M’”is mainly to provide a matrix that will absorb the massive volume change occurred within the electrode upon the lithiation（expansion）/delithiation（contraction） process,thus maintains the mechanical integrity between the particles and also with the current collector.However,many problems occurred during the research of the M’M intermetallic compounds,and one of them is how to choose of the metals of M’ and M,as well as how to optimize the composition since there are dozens of active metals M,like Al,Zn,Ag, In,Sn,Sb,and so on,and even more inactive metals M’,such as Ti,Mn,Fe,Co,Ni,Cu, etc.There will be a great number of such intermetallic compound of M’M,and it will takes hundreds of man hours of experimental work for the tradition research methods.Since first principles quantum chemistry can now predict physical and chemical properties of molecules and solids,attempts should be made to use this tool for predicting favorable new electrode materials,thereby avoiding needless experimentation and focusing work solely on materials that promise success.The important physicochemical performance parameters of a rechargeable battery are its energy density, specific energy,power density,and cycling stability.The challenge is now to predict some of these properties by first principles quantum mechanics.In this thesis,we want to show that it is possible to predict parameters such as the average voltage,energy density,and specific energy of lithium-ion batteries.After the method of first-principles density-functional theory with pseudopotentials and plane wave basis and the software package of VASP we used were introduced,, some cases of first-principles study in the materials of lithium ion batteries were introduced in Chapter 3.Then,a series of simulations for the most common alloys were carried and their Gibbs energies of the reaction with lithium were calculated.According to this Gibbs energy,the estimation of the alloys’ possibility of electrochemical reaction with lithium,hence their potential use for the anode of lithium ion batteries become possible.Base on the above calculation,the CoSn₂ alloy shows great reactivity with lithium ion in a battery environment and was expected to have a high capacity.So,in Chapter 4,the Co-Sn system,including Co₃Sn₂,CoSn and CoSn₂ alloys were systematically studied.A series of Co-Sn intermetallic compounds were prepared by mechanical ball-milling followed by annealing at appropriate temperature.The charge/discharge profile and cycling performance was compared as negative electrodes for lithium ion battery.Moreover,the insertion/extraction mechanism of these alloys were investigated through ex-situ XRD measurement.Further more,the cycle performance of Co-Sn alloy was improved by properly controlling the composition and addition of the graphite in the raw materials during ball-milling.It shows that the electrochemical performance of alloy anodes are closely related with their compositions, crystal structures,and the electronic structures.Firstly,In Chapter 5,a series of Ni doped Ni_xCu_6-xSn₅（χ= 0,0.5,1,2,4） alloys were synthesized.The effects of doped-Ni amount on the structure and lithium intercalation were examined.The relationship between the electrochemical performance and the crystal structure was also studied using the first-principles calculation.The electrochemical tests show that the cycle performance improves while the reversible capacity decreases with the increasing amount of Ni content.A proper amount of Ni doped alloy,NiCu5Sn5 and Ni₂Cu₄Sn₅ show an improved cycling stability with a reversible capacity of 200 mAh/g（1680 mAh/ml）.The first-principles calculations suggest that the lithiation depth of Cu₆Sn₅ can reach to Li_4.4Sn,however,the reversible capacity is very small for Ni₃Sn₂ and it may attribute to the high energy barrier required for the structure transformation from the hexagonal structure to the cubic structure.Secondly,in Chapter 6,a series of Co doped Co_xCu_6-xSn5（0≤x≤2） alloys were prepared experimentally.The electrochemical performance,including the average intercalation/alloying voltage,the sharp of charge/discharge curve and the cycling performance,were studied using both the experiment and computational simulation.The results show that the meta-stable intermediate phases of Li₂Co_yCU_1-ySn form during the Li-ion insertion process of Co_xCu_6-xSn₅ and become unstable and even undetectable with increasing amount of Co substituted.A proper amount of Co doped alloy,CoCu₅Sn₅ showed improved cycling stability at the expense of litter capacity,whereas a heavy Co-doped alloy,Co₂Cu₄Sn₅,resulted in poor cycling ability.The crystal and electronic structure,thermodynamic stability of Co_xCu_6-xSn5 and half-lithiated alloy,Li₂Co_yCu_1-ySn, as well as the average voltage of alloying reaction in terms of different discharge depths were investigated using first-principles calculation.In the Chapter 7,the kinetics of Li-ion intercalation into Cu₆Sn₅ electrode was determined by galvanostatic intermittent titration technique（GITT） and electrochemical impedance spectroscopy（EIS） method.Although the Li-ion diffusion coefficient（DLi） values are relatively high(between 10^-11 to10^-10 cm²/s) in the most voltage range,two minimums(below 10^-11 cm²/s) in the D_Li VS.voltage curve were observed at～0.4 V and 0.1 V,coinciding with the voltage plateau in the charge/discharge curves,indicating reversible structural phase transition or order/disorder transition in the compound.The D_Li derived from GITT and EIS were confirmed well both in the magnitudes and in the variation with Li composition（voltage） range.更多还原