| Chemical systems consist of chemical elements and chemical bonds. Generally, chemical bonds exist between open-shell species (strong covalent bonds); or between closed-shell species (strong ionic bonds), if they carry opposite electric charges. Between closed-shell species of zero charge or, the same nominal charge, no strong attractions are expected. Indeed, two closed-shell metal cations would normally be expected to repel each other. In the case of inorganic or organometallic bonded Au(I), Ag(I), Cu(I) cluster compounds, evidence has, however, recently been accumulated for an entire family of closed-shell interaction between d10-d10 or s2-s2 systems. It is well established that attractive intra- and intermolecular closed-shell interactions between d10 cations lead to the formation of dimers, oligomers, chains, and sheets. Since 1980's, many closed-shell interaction systems have been prepared. It has been confirmed experimentally that such closed-shell interaction's distance is longer than covalent or ionic bonds but shorter than the sum of the metallic van der Waals radius; and the closed-shell interaction energy is weaker than most covalent or ionic bonds but stronger than other van der Waals bonds, and roughly comparable in strength with typical hydrogen bonds.'Aurophilic interaction'has been introduced to name such strong closed-shell interaction. These interactions appear to be similar in strength to hydrogen bonds, and chemists have started to use them to design new structures with unusual physical properties. The construction of striking'rotator phases'of gold complexes based solely on their metal-metal interactions. The behaviour of gold complexes in these rotator phases could potentially be modified to produce interesting and useful structures, such as metal-based liquid crystals or photoluminescent materials. For compounds involving heavier elements, these interactions are particularly strong. Such closed-shell interactions usually exist among heavier elements and go with the strong Coulomb interaction. Since the relativistic effects and the correlation effects can not be ignored, it is a new challenge for the current quantum chemistry on such topic. It is a popular topic about the mechanism of aurophilic interaction now.In this thesis, we want to investigate the mechanisms of aurophilic interaction. The most popular and advanced program packages ADF 2004 and GAUSSIAN 98 were used. We choosed high level basis sets, and applied various theoretical methods for studying closed-shell interaction systems. We used current constructive program to show the electron density difference plots. Many different systems which have not been researched or have been researched a little were chosen as objects, and the geometric structures (bond lengths, bond angles) were calculated and compared to the experimental values. The electronic structures (orbital, HOMO-LUMO gap, bond energies,…) of the chosen systems were analyzed to study the mechanism of closed-shell interaction. The more meaningful things are not the calculational results such as orbital energies, bond lengths, bond energies,…but the physical pictures or chemical rules obtained from the investigations to explain the mechanisms of such closed-shell interactions and to direct the experiments.The experimental and theoretical researching actualities on such topic and the general theory of the implemented methods are given in Chapter I. In chapter II, we have studied the dependence of the aurophilic attraction in perpendicular model systems of the type (Cl-Au-PH3)2 on the ab initio and DFT methods. The size of the basis set and the relativistic effect were considered. The effects of varying the "halogen"(X=H, F, Cl, Br, I, CN, CH3, SCH3) the"phosphine"(L=PH3, PMe3, PPh3) and the metal (M=Au, Ag, Cu) on the M(I)…M(I) interaction of the (X-M-L)2 dimer are also studied. For such systems, MP2 method can not fit for them, the structural parameters obtained by Xαare the closest to the experimental values. In order to understand the aurophilic interaction mechanisms more deeply, we have investigated the distribution of the electron density.In succession, the geometric and electronic structures of the centered systems of gold(I) complexes X(AuPH3)n q+(X=C,N,O,S,Se, P,n=2-6, q=0-2) have been investigated by means of relativistic DFT methods. It is found here for the investigated compounds that simple local density functionals (LDF) of the Xαtype gave the consistent results with the experimental values. Mulliken populations show that there are 0.4～0.7e transferring from Au5d to Au6sp, there are 0.4 to 0.7 electron holes in the Au5d shell. Since the Au5d shell is not completely closed and that it thus may contribute to the closed-shell interaction. For n=2, the electronegativity of central atom X correlates with the Au-X-Au angle, the HOMO-LUMO gap and the electronic structure. For n=3, the change of RX-M, RM-P, RM-M bonds with different metals (M=Au, Ag, Cu) are compared, and the metallophilic interaction energy are analyzed and decomposed. For n=4, it has been found that the classical tetrahedral structure in these four-coordinate compounds is abandoned in favour of a square-pyramidal geometry once the radius of the central element is too large to allow for metal…metal bonding in a tetrahedral geometry. P(AuPH3)4+ is predicted favored on Td structure which confirmed by CC2 method too.Some central-atom gold(I) complexes are grouped in pairs through inter-molecular Au…Au contacts. In chapter IV and V, we reported a theoretical study of the geometric structures and electronic structures of several gold complexes [X(AuPH3)2+]2 (X= Cl, Br, I) and [HS(AuPH3)2]22+. Gold complexes [F(AuPH3)2+]2 and [S(AuPH3)2]2 are predicted. Experimental structure parameters of the title compounds were reproduced at Xαlevel. At the same time, the intermolecular Au…Au interactions are analyzed.In chapter VI, we gave a short discussion on the geometric and electric structures in eight-membered rings of type (AuS2CH)2 and (AuPH2CH2PH2)22+ at ab initio HF, MP2 and DFT levels. The relativistic effects are considered. For such eight-membered rings with different ligands, the HOMO and LUMO orbital components are affected by ligands. In chapter VII, we studied the geometric structures and chemical deformation electron densities of s2-s2 closed-shell interaction systems.For closed-shell interaction systems, Pyykk? and other researchers have carried out theoretical investigations by using ab initio HF and MP2 methods. Pyykk?'s calculations were extremely revealing that aurophilicity is due to electron correlation (or dispersion forces) and also that the attraction is strengthened by relativistic effects. The problems of common DF methods with long range dispersion are well known. On the other hand the DF approach is assumed to cover nondynamical correlations of connected electronic systems. For aurophilic interaction systems, it was thought that DFT method can not be fit for it due to its dispersion interaction previously. Many closed-shell interaction systems have been studied by using DFT method, and we found that geometric structures obtained by Xαmethod are consistent to the experimental values. For such systems, Xαis superior to MP2 method. When MP4 and CCSD(T) have been applied on closed-shell systems, there were still deviations between theoretical and experimental data. Our Xαmethod gives the same good results or even better results than CCSD(T). Why DFT Xαmethod can be suitable for closed-shell interaction systems? Our explanations are given in Chapter VIII. We have investigated the distribution of the electron densities between Au atoms in the foregoing systems. We found that there is about 0.17-0.35 e/(A|°)3 electron density between the two Au(I) atoms. The amount of the electron density is correspond to the electron density between two H atoms at twice equilibrium bond lengths (there still exists attractions between two H atoms). Since the closed-shell interaction has partly covalent components, DFT methods can be fit for closed-shell interaction systems. Our general conclusions are presented in section Summary (chapter IX). |
PDF Full Text Request