【作者】 阚二军；

【导师】 杨金龙；

【作者基本信息】 中国科学技术大学 ， 化学物理， 2008， 博士

【摘要】 磁性材料是古老而用途十分广泛的功能材料。从古代出现的指南针，到近现代的电话机等，都可以看到磁性材料的身影。随着现代信息技术的发展，磁性材料也广泛的被应用于信息存储。我们知道，在信息的存储过程中，我们只利用了电子的自旋特性，而信息的处理过程，则是利用了电子的电荷特性，这两个过程是分开实现的。因此，如果同时利用电子的自旋和电荷特性，不仅会大大加快信息处理过程，也会提高信息存储的密度和持久性。对这一课题的研究也产生了新的领域—自旋电子学（spintronics）。但是对于这一新的领域，很多有趣的性质，比如稀磁半导体中的室温铁磁性，都没有完全被理解。因此相应的理论研究十分必要，在此基础上做出的理论预言也会进一步加快新磁性材料的研究和应用。得益于计算机科学和理论物理化学的发展，基于密度泛函理论的第一性原理计算方法使理论研究磁性材料成为了现实。本论文共五章，分为三个部分：第一部分包括前面三章，主要是研究了目前非常受关注的石墨烯材料的磁性和电子结构。第二部分是研究了基于ZnO的稀磁半导体的磁性质和电子结构，包括电子掺杂的效应和表面的效应。第三部分研究了具有类铜酸盐结构的Cs₂AgF₄材料的铁磁性来源问题。第一章简要介绍了量子化学的历史及发展现状、密度泛函理论的理论框架和近年来的发展。Kohn建立了密度泛函理论基础，即：以基态电子密度为基础，假定一个多粒子体系的任何性质都是基态电子密度的函数。通过将多体相互作用包含在交换相关能量中，使得多体问题化为有效单粒子问题。因此密度泛函理论发展的主线就是寻找合适的交换相关能量泛函。本章最后还简要介绍了本论文所使用的密度泛函计算软件包。在第二章和第三章，我们把目光转向引发当前研究热潮的纳米材料石墨烯。第二章中，我们对它们的研究进展给了一个简单的介绍。石墨烯材料因为其独特的性质，例如反常量子霍耳效应，无质量的狄拉克粒子，吸引了众多理论和实验研究。从第三章开始，我们从理论上，预言了几种基于石墨烯的材料。首先，通过在armchair石墨烯纳米带上吸附过渡金属原子链的方法，我们发现可以实现体系的电子结构的调制，同时产生我们需要的磁性质。并且发现当纳米带的宽度小于2．1nm的时候，这样的杂化结构可能会产生半金属。而zigzag的石墨烯纳米带因为其边界态的存在，导致自身自旋极化的产生，更是我们关注的重点。基于密度泛函的理论研究，我们证实可以通过施加高的横向电场来实现把zigzag石墨烯纳米带转化成半金属。但是因为其电场强度太高，实现成本高。因此我们设计了两种基于化学合成的方法来实现此类材料的半金属性。第一种方法是通过边界修饰的方法，我们通过在zigzag石墨烯纳米带两端修饰不同的化学基团，NO₂和CH₃，可以得到我们需要的半金属性。同时，通过降低化学基团的浓度，可以让材料变得非常稳定，使化学合成变成可能。其次，我们预言了在石墨烯纳米带里面掺杂一条BN链，也可以实现半金属材料的产生。分子动力学的模拟显示此类材料在室温下是稳定的。第四章分析了两种稀磁半导体材料的电子结构。我们首先讨论了Co原子掺杂的ZnO的电子结构和磁性质。研究显示Co原子对倾向于近邻分布，磁性原子之间的磁耦合是反铁磁的。通过掺杂电子，可以显著改变稀磁半导体的电子结构和磁性质。当掺杂的电子浓度为每个Co原子1个电子的时候，我们发现，材料会从半导体转变成为金属，同时Co原子之间的磁性耦合会变成铁磁性。为了研究ZnO这类半导体材料表面对磁性掺杂的影响，我们研究了Cu掺杂的ZnO表面。理论研究的结果表明，不同的表面对于磁性的调制和掺杂原子的布居有非常大的影响。实验上发现的不同磁性的部分根源很有可能是不同表面调制的结果。第五章探讨了实验上合成的，具有类似高温超导材料La₂CuO₄结构的Cs₂AgF₄。实验显示Cs₂AgF₄具有二维AgF层。一个有趣的实验现象就是Cs₂AgF₄表现出与La₂CuO₄不同的磁性质，其二维AgF层是铁磁性的。建立在中子衍射探测的结构基础上，我们进行了理论研究探讨非常规的铁磁性的来源。研究显示，由于二维平面内的F原子偏离了原来的中心位置，触发了平面内Ag原子的z²-y²和z²-x²交错轨道序，通过中间的连接F原子的超交换作用稳定了这种平面的铁磁性。更多还原

【Abstract】 Magnetic materials, as a kind of functional materials, are widely used in many apparatus, ranging from the ancient compass to the modern telephone. With the development of information technology, magnetic materials are also utilized for information storage. As we know, for traditional information devices, electron charge is utilized in the information processing, while electron spin is used in the information storage. To fabricate smaller devices, combining electron charge with spin has attracted many interests.This new area in magnetic materials has been marked as spintronics. However, many novel properties of spintronics have not been fully understood. Thus, theoretical studies are required to provide the physical insights. Due to the progress in computationalmethods and enhancement of computational ability, density functional theory （DFT） has become one of the most important methods in such magnetic materials.The dissertation contains five chapters, which are divided into three parts. The first part containing three chapters focuses on the magnetic and electronic properties of graphene. The second part is to study the magnetic and electronic properties of diluted magnetic semiconductors （DMS）, and the third part are devoted to discuss the ferromagnetism of Cs₂AgF₄.In chapter 1, we introduce the basic idea of DFT and review its recent progress. DFT, which describes that any properties of a many body system can be determined by the charge density at ground state. With the help of Kohn-Sham equation, many body interaction is included in the exchange-correlation energy and the many body problem becomes an effective single particle problem. Finding good approximation of exchange-correlation functional is one of the main targets in DFT. At the end of this chapter, we briefly introduce some DFT based simulation packages used in the dissertation.In chapter 2 and chapter 3, we begin to focus on the graphene. First of all, we give a simple introduction of graphene. Due to its novel properties, such as anomalous quantum hall effect and massless dirac fermions, graphene has been widely stud- ied. In chapter 3, by theoretical treatments, we give several materials designs based on graphene nanoribbons （GNRs）. Firstly, we predict that the strong hybridization between the titanium chain and the GNRs gives rise to ferromagnetism and metallicity for one-dimensional titanium chains adsorbed on semiconducting armchair GNRs. Meanwhile,the adsorption system may offer half-metallicity when the width of GNRs is less than 2.1 nm. Next, for zigzag GNRs, they are semiconductors with two localized electronic edge states, which are ferromagnetically ordered, and antiferromagnetically coupling each other. The calculated results show that such zigzag GNRs can be converted to half metal under enough transverse electric field. However, the high electric field makes the applications difficult. As an alternative way, we theoretical predict that half metal can also be gained by chemical approaches. The first method is edge-modifications. We show that by modified the zigzag GNRs with different chemical groups, the ribbons can become half metal. With the reduction of chemical groups, the ribbons become more stable. We also predict that the ribbons can be changed to half metal by implanting a BN row into zigzag GNRs. Our molecular-dynamics simulations confirm such hybrid structures are stable under room temperature.In chapter 4, we study two kinds of DMS. For Co-doped ZnO, we find that Co atoms tend to distribute on the nearest neighbor sites, and antiferromagnetic （AFM） ordering between the Co ions is favored over ferromagnetic （FM） ordering. But FM ordering can be induced by electron doping. To investigate the surface effect of ZnO on the magnetic properties, we study the Cu-doped ZnO surfaces. Theoretical calculations suggest that surfaces have great influence on the distribution of magnetic atoms and the magnetic coupling.In chapter 5, we investigate one kind of new materials: Cs₂AgF₄, which is an ideal isostructural analog of layered cuprates La₂CuO₄. Magnetic measurements reveal that Cs₂AgF₄ is FM. Based on the structures resolved by neutron diffraction, we find that the ground state has a staggered order of z²-y² and z²-x² orbitals. The superexchange interaction through the Ag（z²-y²） - F - Ag（z²-x²） bridge stabilizes the ferromagnetism.更多还原