In inorganic chemistry, there are numerous compounds with dimolybdenum units, and in these molecules the Mo-Mo bonds are short and strong. In EMACs molecule family, those with dimolybdenum units have the similar characteristics. In those EMACs with dimolybdenum, the Mo-Mo bonds are as short as about 2.10 Å, and the interaction between molybdenum and 3d transition metals are weak, these bond lengths are extremely long.
Some examples form the experimental crystal data form the tri-nuclear EMACs containing dimolybdenum unit are list in Table 1.1.
Table 1.1: Tri-nuclear Mo2ML4X2with dimolybdenum units. M represents the third metal, while L and X are surrounding ligands and axial ligands separately.
M L X Mo-Mo (Å) Mo-M (Å)
However, while displaying similar features, dimolybdenum EMACs shows a good variety of surrounding ligand conformations. Besides the traditional helical conforma-tion, both remarkable non-helical and hemihelical ligand conformations are found in this dimolybdeundum EMACs family.
To begin with, for EMACs listed in Table 1.1, those containing npo ligands are wroth to take a closer look. While the common EMACs are surrounded with helical organic ligands such as dpa ligands, the recently synthesized Mo2M(npo)4(NCS)2(M= Mn, Fe, Co, Ni, Zn) molecules, illustrated in Fig. 1.7 with the npo ligand EMACs, are the first EMACs family ever synthesized that shows nearly planar surrounding ligands. Furthermore, compared to Mo2M(dpa)4(NCS)2, which contains the same central metallic atoms but heve more
commonly used helical dpa ligands, Mo2M(npo)4(NCS)2consistently gives a larger single-molecule conductance based on the break-junction measurement.
What make these npo ligand Mo2M(npo)4(NCS)2EMACs become non-helical ligand conformation will be a straight forward question to ask. In this work, we will try to explain the geometric factor that leads Mo2M(npo)4(NCS)2into a non-helical conformation, and what makes a difference between npo and other ligands, for example dpa ligand, on their ligand conformations. And then we further investigate the influences of ligand helicity on there electronic properties.
Figure 1.7: Tri-nuclear Mo2ML4Cl2(M=Mn, Fe, Co, Ni, L=npo, dpa)
Following the non-helical EMACs, another interesting EMAC with dimolybdenum units is the penta-nuclear Mo2NiMo2(tpda)4(X)2, (X=Cl, NCS) family (shown in Fig. 1.2 with NCS axial ligands), which have hemihelical surrounding ligands mentioned in the first section. Unlike a classical EMAC, Mo2NiMo2(tpda)4(X)2 has ligands which have different chirality at each ends, and have a helical perverse point at the middle of each li-gands. In a organic chemistry point of view, we may call it a meso conformation EMACs.
In fact, when this kind of penta-nuclear EMACs is crystalized, two kinds of crystals with different outlooks are synthesized at the same time. With Single-crystal X-ray ex-periments, both them have been identified. For Mo2NiMo2(tpda)4(Cl)2, two crystals are perfectly the crystal of two conformations separately. However, for the NCS axial ligands one, one of the crystal have the meso conformation and the other one is identified as the cocrystal of helical and meso conformation Mo2NiMo2(tpda)4(NCS)2. Metal-metal bond lengths and related angles for these crystals and the related tri-nuclear EMAC are listed in Table 1.3. For both conformations, the X-ray experiments show similar results on Mo-Mo distances, which is about 2.1Å. And as for the Mo-Ni distances, the meso conformations have a about 0.1 Å longer than the helical ones. In meso conformation the Mo-Ni bond length is about 2.5 Å and for the helical conformation the distance is about 2.4 Å. All the
∠NMM are nearly 90 degrees, which shown almost no pyramidalities. Furthermore, since in meso conformations helical perversion occurs in the middle of the chain, we can see that the N-M-M-N dihedral angles are smaller than in helical ones and be nearly planar at the helical perverse point.
Table 1.2: Bond lengths and angles form the crystal structure of Mo2NiMo2(tpda)4X2and its related molecule
Mo-Mo Mo-Ni ∠NMoMo ∠NMoNi ∠NMoMoN ∠NMoNiN
(Å) (Å) (degree) (degree) (degree) (degree)
meso(NCS)a 2.102 2.516 90.75 85.15 9.15 3.69
2.098 2.509 90.34 85.19 8.40 3.52
meso(Cl) 2.102 2.516 90.09 85.39 7.30 4.63
helical(NCS) 2.086 2.412 89.80 84.57 13.62 19.39
helical(Cl) 2.092 2.412 90.31 86.39 14.58 21.53
Mo2Ni(dpa)4(NCS)2b 2.104 2.546 90.70 86.05 10.52 20.75
aFirst row is from the crystal only containing meso conformation and the second row is from cocrystal
bData from crystal structure in Ref.[13]
As shown in Fig. 1.2 and Table 1.2, in this molecule there are two short dimolybdenum units on each end and one nickel ion at the center of the chain with extremely long Mo-Ni distances. Since the length of dimolybdenum units are similar in all case, the critical reason why these two conformation form separately lays highly possibly on the Mo-Ni bond.
Consisting with the geometrical explanation, based on Eq. 1.1, for the question why EMACs curve into helices, in this work, we try to find out the condition of forming these unique meso conformation EMACs, and try to demonstrate that while the form-ation of ligand helicity is highly correlated to the unbalance of ligand and central metal chain length, the formation of these unique helical perverse ligands in meso EMACs, Mo2NiMo2(tpda)4X2, can also be explain in a geometric manner explained in Ch.1.2.1.
While ligands on this molecule cannot be totally planar due to the steric effect, and the short dimolybdenum length, we focus on the two rather long Mo-Ni bonds at the center, which are far beyond the theoretical length where dihedral angle vanishes in Fig. 1.6. By Eq. 1.1, these long metal-metal bonds favor a planer conformation. We will try to clarify how this relation between ligands and central metal chain give rise to the hemihelical sur-rounding ligands in Mo2NiMo2(tpda)4X2. By demonstrating how meso conformations, which is partially planar at the helical perverse point, are stabilized with extremely long metal-metal bonds at the center, we further predict other penta-nuclear EMACs that are highly possible being meso conformations.
Last but not least, in addition to existing finite EMAC systems, we wish to elaborate the correlation between the central metal atom chain and the ligand conformation in a more general manner. we attempt to built up a simple Hamiltonian to describe and investigate the relation between central atom chain and the ligand conformation.