Oxidative addition and co insertion of rhodium cupferrate complexes containing arsine ligands

Abstract

Results obtained for the reaction between [Rh(cupf)(CO)(AsPh3)] and CH3I in a range of solvents indicated that the second-order rate constants are comparable with its phosphine analogue. A likely explanation given was that steric factors are probably overshadowed by electronic ones. This suggestion is supported by literature findings specifying the σ-donor order of group 15 ligands as PR3 > AsR3 > SbR3. Rhodium complexes with phosphine ligands should thus have more electron density than those with corresponding arsine ligands. The less steric crowding at the Rh centre of the starting complex was proved by rate enhancement (relative to other similar rhodium complexes) of oxidative addition of the bulky substrate, iodoethane, to the complex. The oxidative addition kinetics showed insignificant intercepts for the plots of kobs vs [CH3I] which implied a reaction thermodynamically favoured towards the Rh(III)-alkyl complex product. Electron density and steric constraints of the rhodium(I) species were manipulated in the study by using different arsine ligands. [Rh(cupf.CH3)(CO)(AsPh3)] and [Rh(cupf)(CO)(AsMePh2)] were thus prepared and analysed. Both complexes enhanced the rate of oxidative addition, which is in agreement with the higher electron density at the metal due to the more efficient electron donating property of the methyl group compared to the phenyl group. The increased reactivity of [Rh(cupf)(CO)(AsMePh2)] could also be ascribed to the smaller cone angle of AsMePh2 compared to AsPh3. The rate of oxidative addition of iodoethane was lower, which indicated that the first step of the reaction was a nucleophilic attack by the metal centre (rhodium) on the α-carbon of the haloalkanes. The first-order rate constant for CO-insertion was determined from IR-multi-scans in acetonitrile and chloroform. Although UV/VIS spectra of the reaction showed biphasic kinetics in acetonitrile and methanol, the second step after the initial oxidative addition reaction was not only CO-insertion but, as revealed from IR-multiscanned spectra, also isomerisation from a Rh(III)-acyl to alkyl species in the same step. In other words, the UV/VIS spectra did not show clearly for which species the specific rate was measured in the second part of the biphasic plots. However, IRmulti- scanned spectra of the reaction at fixed [CH3I] for the oxidative addition step (Rh(I)–CO peak disappearance) was possible since it was proved that the graph of kobs vs [CH3I] was linear with insignificant intercepts. The IR-multi-scanned spectra together with the UV/VIS rate data indicated that the fast equilibrium for the formation of the alkyl complex was followed by a second slower CO-insertion reaction. This was consistent with the proposed mechanism of the title reaction as shown below. See Scheme in full text. The observed rate enhancement due to the increase in electron density of the starting complex together with the activation parameters found (relatively large negative ΔS#) and the type of IR-multi-scans spectra obtained are all indicative of an associative mechanism. Furthermore, the activation of the second-order rate constants from non-polar to highly polar solvents could be taken as an indication of a polar transition state. Although activation parameters and IR multi-scanned spectra were performed in different solvents, i.e. in acetone and chloroform, it would not be expected to affect the proposed mechanism. On the basis of a combined solvent, temperature and pressure dependence study conducted, activation parameters for the reaction between CH3I and [Rh(cupf)(CO)(PPh3)] in acetone and chloroform were very close to those obtained for the title reaction in this investigation. In this study, the effect of organic halide substrates on the rate of the reaction by changing the substrate from CH3I to CH3Br and C2H5I was also investigated. In comparison to literature data, the rate of the reaction with C2H5I was faster, i.e. only a 40-fold deactivation was found. Both CH3Br and C2H5I retarded the reaction. Possible explanations could be due to electronic factors (bond energy, BE, of CH3–Br = 70 vs CH3–I = 56 kcal/mol). Since BE of CH3CH2–I is 53 kcal/mol, decreased reactivity observed could be interpreted in terms of the large size of the ethyl group. With regard to the mode of addition, oxidative addition of CH3I to [Rh(cupf)(CO)(AsPh3)] is expected to proceed by addition of iodide (I-) and CH3 adjacently (cis to one another) like in its phosphine analogue. Another aspect of this study was kinetic runs of the oxidative addition product, [IRh(cupf)(CO)(CH3)(AsPh3)], performed in a range of solvents. Most kinetic studies only employ Rh(I) or Ir(I) complexes as a starting material, but in this study the Rh(III)-alkyl product was also utilized as a starting complex. In all solvents employed, the product was depleted and gave rise to a small peak at ca. 1712 cm-1 corresponding to a Rh(III)-acyl species. The rate of acyl formation was thus determined by the slow appearance of this peak.