Table of Contents:
The results of this work are divided into four parts.
The first part discusses the reactions between CDPs and hexamethyl disilazanide complexes. Both CDPPh and CDPcycl reacted with [Fe(HMDS)2] and the two complexes 1 and 2 (cf. Figure 1) were obtained and characterised.
Compound 1 was synthesized by KNEUSELS and its molecular structure was obtained, but it was not fully characterised. In this work, further investigations were carried out with collaboration partners to clarify the magnetic properties and the bonding situation between the C0 atom and the Fe atom. The calculations show that the bond of C0 and the Fe atom is a double bond. With 2, another complex between a CDP and [Fe(HMDS)2] was obtained and its molecular structure could be obtained. No further investigations on the magnetic properties and nature of the C0-Fe bond have been carried. These would be of interest because they would allow good comparison to 1 in which the CDP ligand is not sterically fixed in its backbone.
In addition, the first literature known CDP complexes with alkali metal cations could be obtained. By reacting group 1 HMDS complexes with a transition metal(II) HMDS complex and two equivalents of CDPs (CDPPh and CDPcycl), the complexes 3-6, as shown in Figure 2, were obtained.
In these complexes, the group 1 cation is complexed by two CDP ligands. A closer look at the atomic distances, as well as the behaviour in solution, suggests that the interactions between the C0 atom and the alkali metal cations are rather weak covalent interactions or a mixture of covalent and electrostatic interactions. This assumption is supported by the large number of atomic distances within the VdW radii between the phenyl groups of the CDP ligands and the alkali metal cations. The resulting dispersion interactions indirectly weaken the bond between C0 and the alkali metal cation. This becomes obvious when comparing the CDPs. CDPPh has a greater alkali metal C0 atomic distance than CDPcycl complexes, but there are more contacts within the sum of the VdW radii in complexes 3 and 5 than in complexes 4 and 6. In order to be able to precisely determine the nature of the interactions between the CDPs and the alkali metal cation, quantum chemical calculations need to be carried out in the future so that the individual proportions of the different interactions can be quantified.
The second part of this thesis dealt with the synthesis of new CDPs based on CDPCl. Here, the aim was to synthesize new CDPs via a metathesis reaction. A large number of substrates were tested for this purpose, but only limited success was achieved. It was possible to reproduce two literature known CDPs CDPNMe2 (8) and CDPNEt2, whereby the latter could only be characterised as a chloride salt in its protonated form (9) (cf. Figure 3).
Although this has demonstrated that the reaction pathway pursued works, it does not provide the easy access to a large number of new CDPs that was hoped for. In consequence of this finding, this synthesis should not be pursued further for new CDPs. Other synthesis should be focused on in order to increase the number of CDPs.
The third part of this work, again founded on the previous work from KNEUSELS, dealt with the reaction of CDPs with Group 15 compounds. KNEUSELS synthesised the compounds 10, 12 and 13 shown in Figure 4 and presented their molecular structure.
For these compounds, the synthesis was optimised and the analysis was completed by extensive NMR studies. Furthermore, it was possible to obtain and completely characterise the CDPPh arsenic adduct 11 shown in Figure 5, which was previously unobtainable, thus completing the series of group 15 compounds of CDPPh.
The reactivities of 10 and 13 were further investigated leading to the discovery of complex 14 shown in Figure 6.
The complex exhibits an interesting dissociation behaviour in solution. First NMR studies suggest that the trifluoromethanesulfonate anions are free in solution.
Based on the findings of KNEUSELS and the intramolecular electrophilic aromatic substitution discovered by KNEUSELS, the complex 16 as an analogue to 10 was presented with CDPcycl (cf. Figure 7). With 16, a similar reaction was attempted, unfortunately in vain, to react with an aromatic substrate in a intermolecular substitution reaction. Nevertheless, chloride substituents of 16 were systematically exchanged for phenyl groups and the compounds 17 and 18 shown in Figure 7 were obtained.
Properties of complexes 16-18 were investigated by NMR analysis as well as DFT methods. The influence on the bond between CDPcycl and the P atom as the central atom could observed. The coupling constant of the 2JPP coupling decreases with decreasing electron density at the central P atom. In DFT calculations, a relative increase in the orbital coefficient at the C0 (HOMO) was observed with increasing electron density at the central phosphorus atom. Further investigations in this area should be aiming to understand the electrophilic aromatic substitution as a pathway to new complexes.
The fourth part of this thesis dealt with the electrosteric characterization of CDPs. For this purpose the CDPcycl could be characterised electrosterically by the synthesis of the complexes shown in Figure 8 and compared to literature known CDPs.
From these two complexes it was possible to determine the %Vbur and from complex 21 from TEP. As expected, the %Vbur of 39.9-41.7 % of CDPcycl is slightly smaller than for the CDPPh (41.9-43.4 %). With a TEP of 2024 cm-1 CDPcycl has the lowest TEP (highest total donor strength) of CDPs known so far. In general, electrosteric characterisation is a method that holds the potential to be used as a standard characterisation technique to classify new ligands. With the respective results, the ligands can be selected exactly for the desired application. In addition, it is possible that surprising results such as the formation of complex 20 (Figure 9), may occur.