【作者】 王丽丽；

【导师】 刘成卜；

【作者基本信息】 山东大学 ， 理论与计算化学， 2011， 博士

【摘要】 分子磁性材料处于化学、物理、材料和生物等多学科交叉点,是当今化学学科的前沿领域之一。多核过渡金属配合物是分子磁性材料的重要组成部分,这类化合物兼具无机和有机材料的特点,可以通过使用不同的配体和顺磁中心,来调控磁体的结构和性质,是分子磁性材料研究的重点。多核过渡金属配合物不仅在材料科学领域发挥着重要作用,而且对于生命科学也具有重要意义。在分子磁性材料的理论研究中,对多核过渡金属的的研究主要集中在磁耦合机制和磁-结构关联两个方面。本论文采用密度泛函理论结合对称破损方法（DFT-BS）系统深入地研究了多核过渡金属配合物体系的磁耦合性质、磁-结构关联,从理论上给出具体的设计规则。具体的研究成果如下：一、总结了分子磁性材料的理论研究方法和研究模型,研究方法主要包括分子轨道理论Hoffmann方法、价键理论Kahn方法和Noodleman对称性破损方法。研究模型主要是Heisenberg模型和Anderson超交换理论模型。二、系统地研究了异桥联双核过渡金属配合物的磁-结构关联和磁耦合机制。对三种异桥联双核铜体系和一系列异桥联双核锰体系中的轨道补偿和反补偿效应及磁耦合机制给出了合理的理论解释。在异桥联双核体系中,如果两种桥配体的最高占据轨道与磁轨道的对称性或反对称性组合相匹配,它们就将以互补的形式发生作用,称为轨道补偿效应；如果一个桥联配体的最高占据轨道与磁轨道的对称性组合相匹配,而另一个桥联配体的最高占据轨道和磁轨道的反对称性组合相匹配,它们就会以反互补的形式发生作用,称为轨道反补偿效应。在[Cu2（L-F）（μ-zaindole）] （H3L-F=1,3-bis（3-fluorosalicylidene-amino）-2-propanol）和{Cu（mepirizole）-Br}2（μ-OH）（μ-pyrazole）的研究分析中发现,这两种异桥联双核铜化合物虽然具有相似的结构但磁耦合性质却截然相反。分子轨道研究发现,[Cu2（L-F）（μ-azaindole）（H3L-F=1,3-bis（3-fluorosalicylidene-amino）-2-propanol）中桥联配体羟基和氮杂吲哚的p轨道与磁中心Cu（Ⅱ）的d轨道相互作用方式不同,分别为σ方式和π方式。而在Cu（mepirizole）-Br}2（μ-OH）（μ-pyrazole）中,两个桥联配体的p轨道与磁中心Cu（Ⅱ）的d轨道以相同的方式（π方式）相互作用。由以上研究结果,我们给出了一个合理的并可以普遍解释异桥联金属体系中磁耦合迥异现象的方法：当两个不同的桥联配体与磁中心之间的轨道相互作用相同时,则发生轨道补偿效应,反之则发生轨道反补偿作用。这种方法在异桥联双核铜体系[Cu2Cl2（μ-Cl）（μ-OCH3）（C10H9N3）2]的研究中得到了很好的印证。通过对一系列双核Mn（Ⅱ）化合物[（R-Bpmp）Mn2（μ-OAc）2]的研究发现,桥联配体的电负性是影响该类体系磁耦合作用的重要因素,体系的反铁磁耦合作用随R-Bpmp电负性的减弱而增强。由自旋密度和轨道分析得出,[（R-Bpmp）Mn2（μ-OAc）2]的磁耦合性质对R-Bpmp上R的电负性非常敏感,通过变化不同的R取代基,可以有效地控制[（R-Bpmp）Mn2（μ-OAc）2]类化合物的磁耦合性质,为实验研究提供了理论指导。三、研究了环状三核铜体系Cu3（μ3-X）2（μ-pz）3中的磁耦合机制和磁-结构关联。通过构造动态模型研究了桥联角度的变化对磁耦合常数的影响。与双核体系不同的是,三核体系中存在多种自旋态的偶然简并,三个金属离子上的电子自旋不可能达到两两相反排列,至少有一对离子上的电子自旋平行排列,这样即使两两离子对间磁耦合作用为反铁磁性的,自旋多重度较高的能量状态也有可能作为基态出现,从而发生自旋阻挫现象。因此,我们运用自旋态与磁耦合常数的关系曲线以及分子轨道图解释了体系中的自旋阻挫现象。我们通过改变初始模型的桥联角建立了动态模型。研究发现,当X=Cl或Br时,随着桥联角从760增大到100°,体系的J值有小幅度的增加,从桥联角为1000开始,J值随着桥联角的增大而减小。值得注意的是,在桥联角为1080时,体系的J值变为负值。也就是说,在桥联角为1080时,体系的磁相互作用由铁磁性变为反铁磁性。而当X=O时,体系的J值变化趋势与以上两种体系有很大的不同,在桥联角从760到1200变化的整个过程中,J值都是呈减小趋势,在桥联角为112°时,模型E的J值由正值转变为负值。在桥联角在760-108°范围时,桥联角每变化4°,氧桥联体系的J值变化AJ为180cm-1远大于卤素桥联体系,这表明,氧桥联三核铜体系的磁性质对桥联角的变化比卤素桥联三核铜体系更为敏感,这一点与双核铜体系相似。这表明,通过改变桥联角可以更有效的控制三核铜体系的磁耦合性质。自旋态的能量E与J′/J曲线图可以用来判断体系是否存在自旋阻挫,体系的基态取决于J/J′比值。对于铁磁性体系来说,基态为高自旋态E（2/3,1）,无论J/J′如何变化都不会有阻挫发生。对于反铁磁性体系,基态随J/J′的变化而变化,当0<J/J′<1时,基态为E（2/1,0）E（2/1,0）；J/J′>1时基态为E（2/1,1）；当J/J′趋于1时,体系基态为简并态,单电子自旋无法确定到底向上或向下,从而发生自旋阻挫。通过研究发现,在我们的计算体系中,J与J′值相当,因此,在这些体系中存在自旋阻挫现象。通过分子轨道分析发现,Cu（Ⅱ）离子上的磁轨道主要成分为dx2-y2。在HOMO中,其中两个磁中心铜离子上的dx2-y2轨道与p轨道之间以。方式重叠,轨道间的相互作用达到最大,而另一个铜离子与配位原子的相互作用趋于零。从而进一步证明了在三核铜体系中存在自旋阻挫效应。四、研究了四核锰体系[Mn4（μ-pzbg）2（Hpzbg）2（CH3O）4（CH3OH）（H2O）]磁耦合机制和磁-结构关联效应。一般来说,在多核体系中只会存在一种磁耦合作用,而在我们研究的体系中既有铁磁耦合也有反铁磁耦合,并存在三种磁耦合常数。为了解释这一现象,我们采用动态模型对该体系的磁耦合作用进行了系统研究。该体系中的三种磁耦合常数分别对应三个双核锰体系。磁-结构关联效应的计算表明,之所以产生三种不同的磁耦合作用主要归因于不同的桥联角度。当桥联角从91°变化到119°时,在桥联角为99°时体系从铁磁性变为反铁磁性。这与原始模型中的核心桥联角度相同。当桥联角为99°时,体系都为弱的反铁磁性,与实验结果一致。当桥联角为91°时,体系为铁磁性,而桥联角为111°时,体系为反铁磁性,当两种桥联角同时存在时,体系呈现反铁磁性。这说明,桥联过渡金属化合物的磁性性质可以通过变化桥联角来很好地控制。另外,桥联配体μ-OCH3中O-C键偏离Mn-（μ-O）-Mn平面的二面角α也是影响Mn（Ⅲ）之间磁耦合相互作用的重要因素。通过对二面角与J值之间关系的研究发现,当二面角从0°变化到45°的过程中,J值逐渐变小,在α=400时,J值从正值变为负值。这表明,当二面角增大,体系的铁磁性减弱并最终转变为反铁磁性。与桥联角相比,二面角（扭转角）α的变化对体系的磁耦合相互作用的影响要小得多：当桥联角变化28°（从91°到119°）,J值变化了31cm-1（从-0.87到30.3cm-1）,而当二面角α变化45°,J值只变化了5cm-1。因此,把桥联角作为影响体系磁耦合性质的重要因素是非常必要的。通过轨道分析发现,在异桥联体系中存在轨道互补效应。d轨道的成分对磁中心锰离子之间的磁耦合作用有重要影响,d₂的贡献大小是发生铁磁耦合的重要因素,d₂的贡献越大,体系越倾向于发生铁磁性耦合。金属离子的磁轨道之间的相互作用可分为三类。同平面内的磁轨道相互作用通常导致较强的反铁磁耦合,平行面对面的相互作用会产生最弱的耦合作用,而面内和面外相互作用的共同影响则导致中等强度的磁耦合作用。五、首次研究了一维链混合价CuⅡCuⅠ化合物[Cu12CuⅡBr2（Hm-dtc）2（CH3-CN）2]n[Hm-dtc-=hexamethylene dithiocarbamate]和[Cu12CuⅡI2（Hm-dtc）2（CH3-CN）2]n的磁耦合机制。与一般过渡金属配合物不同的是,这两种体系的桥联配体为一个双核CuⅠ基团,并且两个相邻磁中心CuⅡ之间的距离较长,为10.5（?）,因此在这两种体系中应存在较弱的磁耦合作用。我们通过磁耦合常数及分子轨道和自旋密度的分析,对以上体系的磁耦合机制进行了系统的理论研究。研究结果表明以上两种体系中都存在非常弱的反铁磁耦合作用（J值分别为-3.4cm-1和-1.1cm-1）。分子轨道和自旋密度分析表明,体系中弱反铁磁耦合作用的发生归因于体系中的桥联配体双核CuⅠ基团,该桥联配体的轨道贡献非常小,从而导致Cu（Ⅱ）之间的相互作用较弱。更多还原

【Abstract】 Now, the molecule-based magnetism is one of the frontiers of chemistry, which involves the chemistry, physics and biology researches. As a significant part of the molecule-based magnetism, the polynuclear transition metal systems have the advantages of both the organic and inorganic materials. The properties of the polynuclear transition metal systems could be controlled by changing the ligands and the magnetic metal centers. The polynuclear transition metal systems are not only the significant component in the field of functional material, but also play a dominant role in the life science. The active centers in the proteases in the organisms are always polynuclear transition metal systems, which could control the complex living processes and catalyse the various living reactions. Now, the theoretical researches of the polynuclear transition metal systems mainly focus on the magnetic mechanism and magneto-structural correlation. In this thesis, I investigate the magnetic properties of the transition metal systems including dinuclear and polynuclear systems by the density functional theory combined with the broken-symmetry method (DFT-BS).The main content of this thesis is listed as follow:1. Generally introduce the theoretical calculation method of the molecular magnetic materials. Based on Heisenberg model, there are three methods used to evaluate the magnetic coupling constant of molecular magnets:Hoffmann method of molecular orbital theory, Kahn method of valence bond theory and Noodleman broken symmetry (BS) method.2. We have systematically investigated the magnetic mechanisms and magneto-structural in the hetero-bridged dinuclear transition metal systems, including three dinuclear copper systems and a series of dinuclear Mn systems. In the hetero-bridged dinuclear systems, if the HOMO of the bridge ligands both match the symmetric or antisymmetric magnetic orbitals, there will be orbital complementarity, or else there will be orbital anticomplementarity.In the investigation of the hetero-bridged dinuclear copper systems [Cu2(L-F)(μ-azaindole)(H3L-F=1,3-bis(3-fluorosalicylidene-amino) -2-propanol) and{Cu(mepirizole)-Br}2(μ-OH)(μ-pyrazole), we proposed a new way to explain the orbital complementarity and anticomplementarity in the hetero-bridged systems:there would be orbital complementarity when the orbitals of the bridge ligands interact with the magnetic orbitals of the metal centers in the same way. According to the calculation results, it is found that there exist; on the contrary, there would be orbital anticomplementarity. This method also well explained the magnetic properties in the hetero-bridged dinuclear CuⅡsystems [Cu2Cl2(μ-Cl)(μ-OCH3)(CioH9N3)2].According to our researches on a series of [(R-Bpmp)Mn2(μ-OAc)2] systems, it can be concluded that the electronegativity bridging ligands is a significant factor that influences the magnetic properties of the systems. According to the spin density and orbital analyses, it can be concluded that the magnetic properties of the [(R-Bpmp)Mn2(μ-OAc)2] are insensitive to the electronegativity of the R radicals in the R-Bpmp bridging ligands. So the magnetic properties of the [(R-Bpmp)Mn2(μ-OAc)2] systems could be controlled by changing the R radicals.3. We have investigated the magnetic coupling mechanism and the magneto-structural correlation in the triangle trinuclear copper(Ⅱ) complexes. Different from the dinuclear systems, the different spin states of the trinuclear systems always degenerate and the spin on one pair of the three metal ions must be a parallel. Hence, even if the magnetic coupling interactions between two metal ions are antiferromagnetic, the state with high spin multiplicity may be the ground state of the system, which results in the spin frustration.In order to explore the mechanism of the magnetic coupling interaction for the trinuclear Cu(Ⅱ) complexes, we build three dynamic models [Cu3(μ3-X)(μ-pz)3Cl3] (X = Cl, Br and O, respectively) by pulling theμ3-X bridge away from the center of the Cu3 triangle plane. As X= Cl or Br, with the Cu-(μ3-X)-Cu angles increasing from 76°to 100°, the J values of the models have a slight increase and the J values begin to decrease with the Cu-(μ3-X)-Cu angles increasing from 100°to 120°. Significantly, at the point of Cu-(μ3-X)-Cu=108°, the J values change from positive to negative. In the other word, the magnetic interactions of both the models are changed from ferromagnetic coupling to antiferromagnetic coupling at this point. However, When X=O, the variation trend of the J values for the model is much different. The J values of the model X=O decrease in the whole range of the Cu-(μ3-O)-Cu angle (from 76°to 120°). Moreover, in the range of 76°to 108°, the variation value△J as the Cu-(μ3-O)-Cu angle changing 4°is about 180 cm-1, which is much larger than the△J (the maximum value is 70 cm-1) for models X=C1 and Br. This indicates that the magnetic properties of the trigonal trinuclear Cu(Ⅱ) systems are much more sensitive to the oxygen bridging ligand than the halogen bridging ligand, which is in agreement with the results of the dinuclear Cu(Ⅱ) complexes. This suggests that the magnetic interaction of theμ-O bridged systems can be effectively controlled with changing the Cu-(μ3-O)-Cu angle.The plot of the E/J/J’ versus is a criterion to judge the spin frustration in the trinuclear transition metal systems. E denotes the energy of different spin state and J and J is the two different coupling constants in the trinuclear systems. For the ferromagnetic coupling system, the ground state is high spin state E(3/2,1). Hence, in despite of variation of the ratio J/J’, there is no spin frustration in these systems. In the antiferromagnetic coupling systems J and J’ are negative, and the ground state varies with the changing ratio J/J’ (Figure 5, bottom). For 0<J/J’ <1, the ground state is E(1/2,0) and for J/J’>1, the ground state is E(1/2,1) When J/J’ is equal to 1, the ground state is accidentally degenerate and the spins are unable to decide which state to stand in. Hence, the system is shown to be frustrated. For the models we investigated, the J values are equal to the J’values. Hence, there exists spin frustration phenomena when the coupling interaction is antiferromagnetic.By the molecular orbital analysis, it is found that the HOMO of the systems are mainly composed of the dx2-y2. The p orbitals of theμ3-X bridging ligands interact with the dx2-y2 orbitals of two of the Cu centers by the a pathway which is the most effective pathway to get the largest overlap of the orbitals. However, there is almost no interaction between the p orbitals of theμ3-X and the dx2-y2 orbitals of the other Cu center. This further proves that there is spin frustration in this trinuclear Cu system.4. The magnetic coupling mechanism and the magneto-structural correlations of the tetranuclear Mn system [Mn4(μ-pzbg)2(Hpzbg)2(CH3O)4(CH3OH)(H2O)]Cl2 has been investigated. Generally, in the multinuclear Mn magnetic systems, there always exists only one kind of the magnetic coupling interaction:either ferromagnetic or antiferromagnetic. But in the complex that we investigated, there exist both antiferromagnetic and ferromagnetic coupling between the Mn atoms correlated, and there are three different magnetic coupling constants.The three different magnetic coupling constants correspond to three dinuclear Mn systems. According to the magneto-structural correlation calculations, the three different magnetic coupling interactions in the system are due to the different bridge angles. As the bridge angle∠Mn-(μ-OCH3)-Mn changing from 91°to 119°, the magnetic coupling in the system change from ferromagnetic to antiferromagnetic at the point of Mn-(μ-OCH3)-Mn=99°. This is consistent with the experimental results:the system is weakly ferromagnetic with Mn-(μ-OCH3)-Mn=99°. When∠Mn-(μ-OCH3)-Mn=91°, the magnetic coupling is ferromagnetic and is antiferromagnetic as∠Mn-(μ-OCH3)-Mn=111°. When these two different bridge angle both exist in the system, the magnetic coupling is antiferromagnetic. It is concluded that the magnetic properties of the bridged transition metal systems could be well controlled by using different bridge angle.In addition, the torsion angle (α) of theμ-O-CH3 bond with the Mn-(μ-O)-Mn plane is also an important parameter that influences on the magnetic coupling between the Mn(Ⅲ) centers. Hence, influence of the torsion angle on the J values of the system is analyzed. The results show that asαchanging from 0°to 45°, the J values are decreasing and changed from positive to negative atα=40°. This indicates that with the torsion angleαincreasing, the ferromagnetic coupling is weakened and changed to be antiferromagnetic coupling. Moreover, compared with the bridging angle, the influence for the change of the torsion angle (α) of theμ-O-CH3 bond on the magnetic coupling is much weaker:the magnetic coupling constant is changed about 31 cm-1 (from -0.87 to 30.3cm-1) with the bridging angles changing by 28°(from 91°to 119°), while the magnetic coupling constant is changed only 5cm-1 with the torsion angleαchanging by 45°. So it is essential to make the bridge angle as the significant factor that affects the magnetic coupling interactions in the transition metal compounds.In the tetranuclear Mn(Ⅲ) system, magnetic coupling interactions between the Mn centers are significantly influenced by the the component of the magnetic d orbitals. The lager the contribution of d2 is corresponding to the stronger ferromagnetic coupling. There exist three kinds of strategies for the magnetic coupling interactions between the metal centers. The co-planar interaction of the magnetic orbitals always results in the strong antiferromagnetic coupling, the parallel interaction leads to the weakest coupling, while the intermediate coupling is always due to the mixing of the in-plane and out-plane interaction.5. We have investigated the magnetic coupling mechanism of two mixed-valence CuⅠCuⅡcomplexes with one dimensional (1D) structures, namely, [CuⅠ2CuⅡX2(Hm-dtc)2(CH3-CN)2]n [Hm-dtc-=hexamethylene dithiocarbamate; X=Br or I] for the first time. The distance between two adjacent CuⅡions is 10.5A, which is much longer than the normal transition metal magnetic systems, so there should be weak magnetic coupling interactions in these systems. According to our calculated results, it is found that there are very weak magnetic coupling interactions in the two systems (the J values of them are -3.4cm-1 and -1.1cm-1, respectively). The analyses of the molecular orbitals and spin density distribution reveal that the weak magnetic coupling interactions in the systems are due to the small orbital contribution of the bridging dinuclerar CuⅠradical.更多还原