The EPR spectrum of the novel radical Mes*(CH₃)P—PMes* (Mes*=2,4,6-(tBu)₃C₆H₂) was measured in the temperature range 100-300 K, and was found to be drastically temperature dependent as a result of the large anisotropy of the ³¹P hyperfine tensors. Below 180 K, a spectrum of the liquid solution is accurately simulated by calculating the spectral modifications due to slow tumbling of the radical. To achieve this simulation, an algorithm was developed by extending the well-known nitroxide slow-motion simulation technique for the coupling of one electron spin to two nuclear spins. An additional dynamic process responsible for the observed line broadening was found to occur between 180 K and room temperature; this broadening is consistent with an exchange between two conformations. The differences between the isotropic ³¹P couplings associated with the two conformers are shown to be probably due to an internal rotation about the P—P bond.

EPR spectra show that one-electron reduction of bis(3-phenyl-6,6-(trimethylsilyl)phosphinine-2-yl)dimethylsilane (1) on an alkali mirror leads to a radical anion that is localized on a single phosphinine ring, whereas the radical anion formed from the same reaction in the presence of cryptand or from an electron transfer with sodium naphthalenide is delocalized on the two phosphinine rings. Density functional theory (DFT) calculations show that in the last species the unpaired electron is mainly confined in a loose P — P bond (3.479 Å), which results from the overlap of two phosphorus p orbitals. In contrast, as attested by X-ray spectroscopy, the P — P distance in neutral 1 is large (5.8 Å). As shown by crystal structure analysis, addition of a second electron leads to the formation of a classical P — P single bond (P — P 2.389 Å). Spectral modifications induced by the presence of cryptand or by a change in the reaction temperature are consistent with the formation of a tight ion pair that stabilizes the radical structure localized on a single phosphinine ring. It is suggested that the structure of this pair hinders internal rotation around the C — Si bonds and prevents 1 from adopting a conformation that shortens the intramolecular P — P distance. The ability of the phosphinine radical anion to reversibly form weak P — P bonds with neutral phosphinines in the absence of steric hindrance is confirmed by EPR spectra obtained for 2,6-bis(trimethylsilyl)-3-phenylphosphinine (2). Moreover, as shown by NMR spectroscopy, in this system, which contains only one phosphinine ring, further reduction leads to an intermolecular reaction with the formation of a classical P — P bond.

Two new, “user-friendlyâ€� derivatives of triptycene containing AsH₂ and SiH₃ fragments were synthesized. Both solids are crystalline, air-stable compounds characterized by elevated melting points and resistance toward moisture. The highly reactive As−H and Si−H bonds are protected by the presence of the surrounding phenylene hydrogen atoms, which ensure a remarkable kinetic stabilization of these primary hydrides. After X-ray irradiation of a single crystal of triptycenesilane, a persistent silyl radical was trapped and characterized.

The reduction products of two diphosphaalkenes (1 and 2) and a bis(diphosphene) (3) containing sterically encumbered ligands and corresponding to the general formulas Ar−X==Y−Ar‘−Y==X−Ar, have been investigated by EPR spectroscopy. Due to steric constraints in these molecules, at least one of the dihedral angles between the CXYC plane and either the Ar plane or the Ar‘ plane is largely nonzero and, hence, discourages conformations that are optimal for maximal conjugation of P==X (or P==Y) and aromatic π systems. Comparison of the experimental hyperfine couplings with those calculated by DFT on model systems containing no cumbersome substituents bound to the aromatic rings shows that addition of an electron to the nonplanar neutral systems causes the X==Y−Ar‘−Y==X moiety to become planar. In contrast to 1 and 2, 3 can be reduced to relatively stable dianion. Surprisingly the two-electron reduction product of 3 is paramagnetic. Interpretation of its EPR spectra, in the light of DFT calculations on model dianions, shows that in [3]^2- the plane of the Ar‘ ring is perpendicular to the CXYC planes. Due to interplay between steric and electronic preferences, the Ar−X==Y−Ar‘−Y==X−Ar array for 3 is therefore dependent upon its redox state and acts as a “molecular switch�.

Reduction of a solution of octamethylcyclo-di(m-silylphenylenedisiloxane) 4 in THF on a potassium mirror leads to EPR/ENDOR spectra characterized by a large coupling (~20 MHz) with two protons, similar to the spectra obtained after reduction of the m-disilylbenzene derivative 5, consistent with a localization of the extra electron on a single ring of 4. The spectra recorded after reduction of 4 at low temperature in the presence of an equimolar amount of 18-crown-6 exhibit couplings of ~10 MHz with four protons and indicate that embedding the counterion in crown-ether provokes the delocalization of the unpaired electron on the two phenyl rings of 4. The measured hyperfine interactions agree with those calculated by DFT for the optimized structure of 4^•-. Direct information on the structure of this anion is obtained from the X-ray diffraction of crystals grown at -18 °C in reduced solutions containing 4, potassium, and crown ether in a THF/hexane mixture. Both DFT and crystal structures clearly indicate the geometry changes caused by the addition of an electron to 4: the interphenyl distance drastically decreases, leading to a partial overlap of the two rings. The structure of 4^•- is a model for an electron transfer (ET) transition state between the two aromatic rings. The principal reason for the adoption of this structure lies in the bonding interaction between the LUMO (π* orbitals) of these two fragments; moreover, the constraints of the macrocycle probably contribute to the stabilization of this structure.

X-irradiation of single crystals of Tp–GeH₃ (Tp: triptycene) led to the trapping of the radical Tp–^√GeH₂. The angular variations of the resulting EPR spectra were recorded at 300 and 77 K. The drastic temperature dependence of the spectra was caused by both a strong anisotropy of the g-tensor and a rotation of the ^√GeH₂ moiety around the C–Ge bond. The determination of the EPR tensors as well as the analysis of this motion required to take the presence of disorder in the crystal into account. In accordance with DFT calculations, Tp–^√GeH₂ is shown to be pyramidal and to adopt, in its lowest energy structure, a staggered conformation. Rotation around the C–GeH₂ bond is blocked at 90 K and is almost free above 110 K. The experimental barrier, obtained after simulation of the EPR spectra as a function of the rotational correlation time, is equal to 1.3 kcal mol^−1, which is slightly inferior to the barrier calculated by DFT (3.6 kcal mol^−1). Calculations performed on Tp–CH₃, Tp–GeH₃ and Tp–^√GeH₂ show that the rotation barrier ΔE_rot around the C–Ge bond drastically decreases by passing from the germane precursor to the germanyl radical and that ΔE_rot increases by passing from the germane to its carbon analogous. Structural parameters involved in these barrier differences are examined.

The paper shows the possibilities of the complementary use of the density matrix formalism for the simulation of the anisotropic EPR spectra and the DFT potential energy surface calculations to obtain a detailed picture of the motions of radical molecules. The combined approach is illustrated by a comparative EPR study of three phosphorus derivatives of barrelene. Three compounds were chosen as the model molecules for the observation of different temperature dependent dynamics of radical fragment. Each molecule based on the same barrelene skeleton has a different set of substituents which by influencing the local chemical environment are likely to modify the internal dynamics. The temperature dependent EPR spectra are simulated by means of the density matrix formalism and the geometry of radicals are calculated with DFT. The motion is described in terms of rotational barriers, DFT calculated energy profiles and hypothetical intramolecular distortions. These two approaches lead to a similar microscopic picture of the intramolecular radical motion.