【作者】 袁艳丽；

【导师】 王红明； 刘小明；

【作者基本信息】 南昌大学 ， 微纳米材料科学与工程， 2013， 博士

【摘要】 石墨烯化学功能化已经成为改善石墨烯性能的一种重要的方法。在已有的共价功能化方法中,1,3.偶极环加成反应是制备杂原子石墨烯中最便捷、最成功的合成方法。制备所得的功能化石墨烯可以观察到明显的电子结构变化、反应活性变化等,实现了性质可控。然而石墨烯化学功能化的理论模拟研究工作还远滞后于实验的发展,现有理论研究中存在的主要问题是：(1)只局限于简单的结合能计算,不涉及机理研究,其反应机理尚不清楚；(2)仅研究了2种1,3.偶极体,其他常见偶极体的研究在理论上均未知,实验报道亦较少；(3)系统的反应活性研究工作鲜有报道。本文针对上述问题,利用密度泛函方法及完全基组方法系统考察了石墨烯化学功能化的机理,反应活性的影响因素和规律。具体研究工作如下：1、用密度泛函方法研究了叠氮三甲基硅烷对石墨烯的1,3.偶极环加成功能化的机理,考察了在完整石墨烯平面的反应活性差异和影响活性的因素。计算结果显示该反应是一个两步反应过程,第一步是化学吸附过程,即1,3一偶极体叠氮三甲基硅烷化学键合在纳米石墨烯表面,并在纳米石墨烯表面形成五元杂环；第二步是热分解过程,即在外界能量的作用下,第一步形成的五元环会发生分解,最终生成氮气和氮杂环化合物。其中,第二步过程决定了反应的速率。首先,我们比较了两种可能的反应路线[3+2]加成和[3+4]加成,结果证明[3+2]反应路线明显优于[3+4]反应路线。前者是一个对称协同反应,但是后者是一个不对称的协同反应,该结论可通过前线分子轨道理论分析得出；其次,我们详细讨论了纳米石墨烯的反应活性,得出主要影响其反应活性的是反应位点电子云密度的大小,其次是轨道能量因素。因此,纳米石墨烯边界部分的反应活性远比其中间位点的反应活性大得多。2、用密度泛函方法研究了9种1,3.偶极体对石墨烯的化学功能化机理,并对反应活性、电子结构和物质本性导致的差异进行了系统分析和总结。在本文中,我们首先采用密度泛函理论系统地考察了9种1,3.偶极体(重氮化物diazonium betaines,腈叶立德nitrilium betaines和甲亚胺叶立德azomethine betaine)与纳米石墨烯的环加成反应。计算结果显示,绝大多数所考察的1,3.偶极环加成反应都具有较负的吉布斯自由能△G,能在温和条件下自发进行。1,3-偶极体本身的性质对1,3-偶极环加成反应活性的影响最为重要。如按其价电子划分,18价电子甲亚胺叶立德的活性要比其他两个16价电子的甲亚胺叶立德的活性大得多。按其端基原子划分,上述1,3-偶极体的反应活性顺序为是：氧化物<亚胺<叶立德。我们还发现在本体系中,影响反应活化能的变形能实际上取决于1,3.偶极体的形变程度或其两共振键初始键能的大小。此外,和以前报导的其他1,3.偶极环加成反应不一样的是,本体系中的活化能和反应热呈现很好的线形关系,符合哈蒙德假说。3、用密度泛函方法研究了Stone-Wales缺陷石墨烯1,3.偶极环加成反应的机理,计算显示该反应路径仍然是[3+2]加成方式。缺陷存在时,纳米石墨烯的中间缺陷位点的活性显著增加,最活跃的位点是七元环与七元环的交线上。与完整石墨烯平面的1,3-偶极环加成反应比较得出,缺陷的存在也同时增加了边界位点的反应活性,中间位点和边界位点的活性差距远小于完整石墨烯平面边界位点和中间位点的活性差距,并会有一定的重叠。这说明当石墨烯表面存在缺陷时,边界位点和缺陷位点参与1,3-偶极环加成的地位同等重要,会同时参与功能化,与对实验结果负载比的推测是一致的。本体系中,活性差异很好地遵循FMO控制,其次通过变形能作用分析,能很好地解释边界位点和缺陷位点的活性差异。更多还原

【Abstract】 The cheimical functionalization of graphene has become an important method to improve the properties of graphene. Among the1,3-dipolar cycloaddition reaction is the most successful which provides the most convenient synthesis methods to prepare the atoms doping graphene. The functionalized graphene has been observed significant changes in electrons structure, reactivity and tunable energy band, which realizes the control of graphene properties. However, the theoretical simulation research of the chemical functionalization of graphene is still far behind the development of experiments. The main problems exist in the current theoretical researches are:(1) there are only simple binding energy calculations which do not involve the reaction mechanism. So far the reaction mechanism is unclear yet;(2) only two kinds of1,3-dipoles has been studied, but the other common1,3-dipoles are theoretically unknown and also less involved in the experimental reports;(3) systematic reactivity studies are rarely reported.Aiming at these problems, in this paper, we have performed systematic theoretical study into reaction mechanism for the chemical functionalization of graphene and the reactivity study using the density functional method and complete base set of methods. Specific researches are as follows:1. The mechanism of1,3-dipole cycloaddition reaction of azidotrimethylsilane (ATS) onto nanographene(NG) was thoroughly investigated at the B3LYP/6-31G(d,p) level. Calculations reveal that the reaction occurs via a two-step reaction mechanism. The first is the chemical adsorption step, and the second is the decomposition of the thus-formed nitride upon thermal activation, then giving rise to an N-bridged product ultimately. The latter is the rate-determining step. Two possible pathways were compared first, evidencing that the [3+2] channel is favorable over the [3+4] channel. The former is a symmetric synchronous process, but the latter follows an asymmetric concerted way, which can be rationalized by means of FMO theory. The reactivity of NG was then discussed in detail, revealing that it is the electron density at the functionalization site which dominates the reactivity rather than the energetic effect. As a result the edge area is calculated to be much more reactive than the centre. 2. We firstly present a systematic investigation into the reactions of nine1,3-dipoles(diazonium, nitrilium and azomethine) with nanographene(NG) model using the density functional theory(DFT). The calulations shows that the nine1,3-dipole cycloaddtions(DC) studied are almost of largely negative Gibbs free energies(ΔG) values which are spontaneous at the mild conditions. The dipole nature is of the most important influence on reactivity of1,3-DC. The18-valence-electron azomethine ylides shows far more active than the other two16-valence-electron types to NG The order of the reactivity for these dipoles is:oxide<imine<ylide. It was firstly found the distortion energy which determines the activation energy depends on the deformation of1,3-dipole or even the strengths of two resonance bonds in1,3-dipole. Unlike the1,3-dipolar cylcoadditions to some other dipolarophiles, there is also good relationship between the activation energy and the reaction energy, which follows the Hammond postulate. All the reactivities are consistent analyzing in FMO.3. The mechanism of1,3-dipole cycloaddition reaction of azidotrimethylsilane (ATS) onto Stone-Wales nanographene(NG) surface was thoroughly investigated at the B3LYP/6-31G(d,p) level. Calculations reveal that same as the addition to perfect nanographene surface, the [3+2] channel is still the preferential pathway. When the defect exists, the reactivity of the centre defect site increases significantly. The most reactive defect site is on located on the intersection of two seven membered rings. In contrast to the addition results to the perfect graphene, the existing of the defect also improves the reactivity of the boundary sites of nanographene. The difference between the reactivity of the centre and the edge of nanographene is much smaller and there is some overlap. This suggests that when there is the defect on the surface of nanographene, both the roles of the boundary sites and the centre site are equally important in the chemical functionalization by1,3-DC. Both would participate in the reaction. This conclusion is just consistent with the speculation of the functionalization ratio toward the experiment results. In the present system, the reactivity follows the control of FMO well. Moreover, the difference between the reactivity of the defect site and the edge site can be fully explained by the distortion energy.更多还原