利用探针在金属表面进行单原子操纵的理论研究

【作者】 谢逸群；

【导师】 干福熹； 庄军；

【作者基本信息】 复旦大学 ， 光学， 2008， 博士

【摘要】 自从Eigler等人使用扫描隧道显微镜(Scanning Tunnelling Microscope,STM)在极低温下(4K)对吸附在Ni(110)表面的单个Xe原子进行了成功的操纵以来,单原子操纵技术在过去十几年的时间里一直是表面科学的研究热点。这种技术对纳米科学是一个重要贡献,具有迷人前景。比如,它使得人们有可能以自下而上的方式(Bottom-up)构造各种理想的纳米结构和器件。目前,STM和原子力显微镜(Atomic Force Microscope,AFM)是原子操纵的两种常用工具。利用显微镜的探针人们可以对吸附在材料表面的单个原子进行横向和纵向操纵。操纵的可靠性无疑是原子操纵的一个重要问题,这是原子操纵技术能否得到实际应用的关键.然而,目前的研究很少涉及到操纵的可靠性。本论文一方面着重于横向操纵可靠性的理论研究,旨在获得提高操纵可靠性的方法。另一方面,提出了分别适用于在平整表面和台阶表面进行单原子纵向操纵的新方法。本论文的第一章简要回顾了近年来原子操纵技术的研究背景和研究进展,并提出了一些有待进一步研究的问题。第二章概述了本论文使用的理论方法和相关的理论基础,包括原子嵌入势,分子力学方法和第一性原理方法的基本原理。第三章,利用分子力学方法结合半经验势模拟了在Cu(111)表面利用Cu探针对单个Cu原子的横向操纵,研究了操纵的可靠性,并对主要结果使用第一性原理计算做了验证。主要研究了探针结构,探针高度和探针方向对操纵可靠性的影响,并揭示了其中的物理机制。我们考虑了三种探针结构,即单原子,双原子和三原子探针。研究结果表明,单原子探针的操纵可靠性随探针的高度下降而逐渐提高;相对于单原子探针,双原子和三原子探针都能够提高操纵的可靠性,原因在于两种机制的作用。分别是:增强的相互作用机制和原子悬浮机制。对于双原子探针只有前者起作用,而对于三原子探针两者都起作用,因此和另外两种探针相比,三原子探针能够在宽广的探针高度范围内获得可靠的操纵。此外,基于三原子探针良好的横向操纵能力以及对吸附原子较强的纵向吸引能力,我们提出了一种在没有电场或电流的条件下对Cu(111)表面单个Cu原子进行可逆纵向操纵的新方法。第四章,用分子力学方法结合半经验势研究了Pt,Al,Pd等五种fcc(111)金属表面的横向操纵的可靠性和单原子纵向操纵。对于这五种体系我们得到了相似的结果,主要以Pt/Pt(111)体系为例进行讨论和分析。我们发现,在较低探针高度下(接近或低于吸附原子的高度)单原子探针的横向操纵可靠性对探针的方向比较敏感;选取合适的探针方向,能够获得较好的操纵可靠性。我们还提出了一种在有台阶的表面实现可逆纵向操纵的方法。使用三原子探针我们能够把吸附在台阶边沿的甚至是台阶本身的单个原子提取出来,而以一定的方式我们又可以把吸附原子从探针释放到台阶表面。这种方法同样不依靠电场或者电流的辅助,只需要对探针的位置进行适当的调节,因此简单而且易于控制。第五章,使用第一性原理方法研究了Al(111)台阶表面的单原子纵向操纵。先用分子力学方法模拟了提取和释放台阶本身的单个原子的过程。更有趣的是,利用这种技术我们首次展示了对表面吸附团簇进行可逆的单原子级别修饰的能力。作为一个具体的例子,我们把一个包含十个原子的六角形铝团簇修饰成三角形状。此外,为了检验热效应对操纵过程的影响,采用了第一性原理分子动力学方法模拟了修饰团簇的过程。结果发现我们能够在一定的温度下(≤100K)实现这种可逆操纵,表明我们提出的操纵方法能够经受实际应用中一定程度的热效应的干扰,为进一步的实验提供了可靠的理论参考。第六章,总结了本论文的主要结果和贡献并对后继的研究工作做了简要介绍。更多还原

【Abstract】 Single-atom manipulation has been the subject of intense research in scientific society during the past years since the pioneer work by Eigler and co-workers, who used a scanning tunnelling microscope(STM) to move and reposition single Xe atoms adsorbed on Ni(110) surface at a cryogenic temperature (4K). This technique is a new contribution to nanoscience since, for example, it makes possible for people to fabricate designed nanostructures in a bottom-up way. STM and atomic force microscope(AFM) are the two main tools for single-atom manipulation. With the tip of a STM or AFM, single atoms can be manipulated laterally or vertically on surfaces. The reliability of the manipulation is naturally a key issue for single-atom manipulation, which is important for the real application of the technique, yet few studies have dealt with this problem so far to our knowledge. In this thesis we focus on the theoretical investigation of the reliability of the lateral single-atom manipulation, and aim to find the method to improve the reliability. Moreover, we present new methods for the reversible vertical manipulation of single atoms on the flat and stepped metal surfaces, respectively.Chapter 1 provides historical background and new developments of single-atom manipulation and raises some important problems that deserve further studies. Chapter 2 describes briefly the theoretical foundations and methods used in our study, including embedded atom method (EAM) potentials, molecular static (MS) method and density functional theory based first-principles methods.In chapter 3, we study the reliability of the lateral manipulation of a single Cu adatom on a Cu(111) surface with single-atom, dimer and trimer apex tips using both semiempirical and first-principles simulations. The influences of the tip-apex geometry, tip height and tip orientation on the manipulation reliability are examined. For the single-atom apex tip the manipulation reliability increases monotonically with decreasing tip height. For the dimer and trimer apex tips the manipulation reliability is greatly improved compared to that for the single-atom apex tip over a certain tip-height range. Two kinds of mechanisms are found responsible for this improvement. One is the so-called enhanced interaction mechanism and the other is the suspended atom mechanism due to the strong vertical attraction of the timer apex tip on the adatom. Both mechanisms occur in the manipulations with the trimer apex tip, while in those with the dimer apex tip only the former is effective. Moreover, we present a method to realize reversible vertical manipulation of a single atom on a Cu(111) surface with the trimer apex tip, based on its strong vertical and lateral attraction on the adatom, without any electric fields.Chapter 4 deals with the reliability of the lateral manipulation of single adatom on five fcc(lll) metal (Pt, Al, Pd, Ni, Au) surfaces, using MS method with EAM potentials. We obtain similar results for these systems and those for Pt/Pt(lll) system are presented. It is found that for the single-atom apex tip the manipulation reliability is sensitive to the tip orientation in a lower tip-height range, and choosing an appropriate tip orientation one can obtain a better manipulation reliability. Moreover, we propose a novel method for the reversible vertical manipulation of single atoms on the stepped surfaces. With the trimer apex tip, we can pick up the adatom adsorbed near the step edge and even extract individual atoms in the step, and in reverse we can release the atom from the tip to the surface in a certain way. This reversible process is simple and very controlled as it only requires to adjust the tip position properly while electric fields are not necessary.Chapter 5 focuses on the single-atom extraction on a stepped Al(111) surface using both MS and molecular dynamics (MD) simulations based on first-principles calculations. We demonstrate a very controlled process about how to extract and reposition single atoms on the stepped Al(111) surface. More interestingly, based on this result we present for the first time a method to modify nanoclusters on surfaces in a reversible way on an atom-by-atom basis. As an illustration, we modify a ten-atom hexagonal cluster to a triangular one. The MD simulations show that the modification process can be completed at a temperature less than or equal to 100K, which indicates that the proposed method can be reliable against the thermal disturbance and provides a theoretical guidance for the real experiments.The last chapter presents a summary of this thesis and some prospects for the ongoing investigations.更多还原