Spectroscopic and Computational Investigations of Stable Radical Anions of Triosmium Benzoheterocycle Clusters

Abstract

The variable temperature (1)H and (13)C NMR and EPR spectra of the stable radical anions [Os(3)(CO)(9)(micro(3)-eta(2)-L)(micro-H)] (LH=phenanthridine, 1; 5,6-benzoquinoline, 2), and [Os(3)(CO)(10)(micro(3)-eta(2)-L)(micro-H)] (LH=quinoxaline, 3) are reported. The radical anions 1(-), 2(-), and 3(-) can be prepared by both exhaustive electrolysis and partially by chemical reduction with cobaltocene and with sodium dispersion (only with sodium dispersion in the case of 3(-)). DFT calculations on 1-3 reveal that the LUMO for the electron-deficient compounds 1 and 2 involves significant contributions from both the heterocyclic ligand and the two metal atoms bridged by the ligand and the micro-hydride. The character of this orbital rationalizes the previously observed regioselective reactions of these complexes with nucleophiles. In contrast, the LUMO for the electron precise 3 involves only ligand-based orbitals. Partial chemical reduction of 1 and 2 requires an excess of either cobaltocene or sodium, and their (1)H and (13)C NMR spectra reveal selective line broadening of those proton resonances that are predicted by DFT calculations to bear the greatest amount of free spin density. The variable temperature behavior of the partially chemically reduced species of 1 and 2 indicates that electron transfer between the reduced/unreduced cluster pair and between the cobaltocene/cobaltocenium pair occurs on the NMR timescale. The radical anions of 1 and 2 prepared by exhaustive electrolysis show an EPR signal at room temperature, while the NMR signals are uniformly broadened. Compound 3 appears to be partially reduced by sodium at room temperature and shows uniformly broadened (1)H NMR resonances at room temperature that sharpen significantly at -80 degrees C. The temperature dependence of the spectra are discussed in terms of the effects of relative electron nuclear relaxation processes, chemical exchange, and the results of the DFT calculations.

Experimental and Computational Studies of Nucleophilic Attack, Tautomerization, and Hydride Migration in Benzoheterocycle Triosmium Clusters

“…Natural Population Analysis for Evaluating the Site of Nucleophilic Attack. We previously reported that the reason the site of nucleophilic attack in complexes such as 1 was changed from the 5-position (in 1 ) to the 2-position (in 2 ) upon adding H - and H + was related to the disposition of the LUMO in the complex relative to the free ligand . This approach gave good agreement with experiment for several systems and explained both the chemical and spectroscopic properties of a range of heterocyclic complexes. , However, in molecules as complex as those discussed here, where there are many similar LUMO's, it can be dangerous to assign chemistry on the basis of the properties of one of many virtual orbitals.…”

Reactivity of Silole within a Core‐Modified Porphyrin Environment: Synthesis of 21‐Silaphlorin and its Conversion to Carbacorrole

“…Oxidation was followed systematically by mass spectrometry and H NMR resonances, which suggests merely the formation of a radical species that, even if present in a minute amount, may lead to significant line broadening. [43][44][45][46] The ESI spectrum measured immediately after addition of DDQ to the solution of the sample in chloroform yielded an intense peak at m/z = 866.4, which corresponds to the molecular formula of 1-SiMe 2 or its isomer iso-1-SiMe 2 . Separate peaks in the same spectrum were readily assigned to oxygen or dioxygen adducts of 1-SiMe 2 or to its desilylation products.…”

Investigations of Pyridine-2-thiol as a Ligand: Synthesis and X-ray Structures of the Mixed Mo−Mn Dinuclear Complex CpMoMn(CO)3(μ-CO)(μ-η2-pyS)(μ-η1-pyS), the Electron-Deficient Trimolybdenum Cluster Cp3Mo3(μ-CO)2(μ-S)(μ3-S)(μ-η2-NC5H4), and the Mononuclear CpMo(CO)2(μ-η2-pyS)

“…Both of these compounds as well as 5 contain relatively bulky cyclopentadienyl ligands as well as face-capping and/or multiple-edge-bridging ligands. It is well-known from molecular orbital calculations that the site and rate of nucleophilic attack on electron-deficient clusters depends primarily on the accessibility of the LUMO . It may be that the presence of the bulky cyclopentadienyl ligand and the multiple bridges in 5 make such an orbital kinetically inaccessible.…”