Allyl radicals rotation

A comparison of the rotational barriers in allylic radicals A-D provides evidence for the stabilizing effect of the capto-dative combination ... 

Butler and co-workers have taken a unique approach to study the unimolecular dissociation of the vibrationally and rotationally hot allyl radical.150-152 They have examined the secondary C-H dissociation of the allyl radicals that are produced with high internal energies above the allyl dissociation thresholds in the primary photodissociation of allyl chloride and allyl iodide at 193 nm. The production of allene versus propyne (both at mass 40) from the secondary dissociation of the hot allyl radicals are... 

The study of substituted allyl radicals (Sustmann and Brandes, 1976 Sustmann and Trill, 1974 Sustmann et al., 1972, 1977), where pronounced substituent effects were found as compared to the barrier in the parent system (Korth et al., 1981), initiated a study of the rotational barrier in a captodative-substituted allyl radical [32]/[33] (Korth et al., 1984). The concept behind these studies is derived from the stabilization of free radicals by delocalization of the unpaired spin (see, for instance, Walton, 1984). The... 

Stabilized allyl radical will be stabilized further if substituents are introduced. This stabilization occurs to different degrees in the ground state and the transition structure for rotation. In the ground state the substituent acts on a delocalized radical. Its influence on this state should be smaller than in the transition structure, where it acts on a localized radical. In the transition state the double bond and the atom with the unpaired electron are decoupled, i.e. in the simple Hiickel molecular orbital picture, the electron is localized in an orbital perpendicular to the jt(- c bond. 

Table 11 Rotational barriers (kcal mol" ) in allylic radicals."...

The experimental result seems to support this model. Table 11 lists values for rotational barriers in some allyl radicals (Sustmann, 1986). It includes the rotational barrier in the isomeric 1-cyano-l-methoxyallyl radicals [32]/ [33] (Korth et al., 1984). In order to see whether the magnitude of the rotational barriers discloses a special captodative effect it is necessary to compare the monocaptor and donor-substituted radicals with disubstituted analogues. As is expected on the basis of the general influence of substituents on radical centres, both captor and donor substituents lower the rotational barrier, the captor substituent to a greater extent. Disubstitution by the same substituent, i.e. dicaptor- and didonor-substituted systems, do not even show additivity in the reduction of the rotational barrier. This phenomenon appears to be a general one and has led to the conclusion that additivity of substituent effects is already a manifestation of a special behaviour, viz., of a captodative effect. The barrier in the 1-cyano-l-methoxyallyl radicals [32]/... 

An error-propagation analysis allows the conclusion that the captodative substituent effect on the rotational barrier in this allyl radical is at least additive and perhaps slightly greater. 

Delocalization of the odd electron into extended n systems results in considerable radical stabilization. The C—H BDE at C3 of propene is reduced by 13 kcal/mol relative to that of ethane. That the stabilization effect in the allyl radical is due primarily to delocalization in the n system is shown by the fact that the rotational barrier for allyl is 9 kcal/mol greater than that for ethyl. Extending the conjugated system has a nearly additive effect, and the C—H BDE at C3 of 1,4-pentadiene is 10 kcal/mol smaller than that of propene. Delocalization of the odd electron in the benzyl radical results in about one-half of the electron density residing at the benzylic carbon, and the C—H BDE of the methyl group in toluene is the same as that in propene. 

Termination rates ordinarily quoted are 2k, where kt is the bimolecular rate constant for the sum of recombination and disproportionation. The factor of 2 appears because two radicals are removed from the system each time the reaction occurs. See K. U. Ingold, in Free Radicals, J. K. Kochi, Ed., Vol. I, p. 43, for references to particular alkyl radical results. Rotating sector measurements for allylic radicals have been reported by H. J. Hefter, C. S. Wu, and G. S. Hammond, J. Amer. Chem. Soc., 95, 851 (1973). 

The ring-opening reaction of cyclopropyl radical [Equation (7)] was shown to occur at 174°C to give ally radical, but the product stereochemistry was unclear. Ab initio direct dynamics study was carried out to clarify the stereochemical course of the reaction.40 Trajectories were initiated at the ringopening TS obtained at CASSCF(3,3)/6-31G(d), with quasiclassical normal sampling at the experimental temperature of 174°C. ZPE was included, and thermal vibrational energy was sampled from the normal-mode Boltzmann distribution. A rotational energy of RT/2 was added toward the allyl radical product. 

Benzene and other aromatics alike are stable molecules, while cyclobutadiene and other antiaromatic molecules are unstable molecules.27-76 Similarly, allylic species are stable intermediates and possess significant rotational barriers. It may appear as a contradiction that, for example, the tr-component of benzene can be distortive but it still endows the molecule with special stability or that the distortive jr-component of allyl radical can lead to a rotational barrier. We would like to show in this section that these stability patterns derive from the vertical resonance energy which is expressed as a special stability because for most experimental probes (in eluding substitution reactions) the o-frame restricts the molecule to small distortion167 in which the vertical resonance energy is still appreciable, as shown schematically in Figure 5. 

Shaik and Bar102 demonstrated that allyl anion has a distortive jr-component but at the same time exhibits a rotational barrier. This analysis was reaffirmed later for allyl radical.5 Subsequently, Gobbi and Frenking93 pointed out that the total distortion energy of allylic species is very small because it reflects the balance of jr-distortivity opposed by the a-symmetrizing propensity. They further argued that along with this jr-distortivity, the allylic species enjoys resonance stabilization which is the source of the rotational barrier. A detailed VB analysis by Mo et al.149 established the same tendency. 

Figure 9. Hess cycle used to relate the rotational barrier of allyl radical to its resonance energy. All energies are given in kcal/mol (data for the cycle is taken from ref 93 B values from ref 111).

The most likely multistep mechanism of this type is shown in the lower part of Figure 15.17. It is a two-step mechanism where the diastereomeric diradicals F and G are the two intermediates that allow for rotation about the configuration-determining C—C bond. Each of the two radical centers is part of a well-stabilized allyl radical (cf. Section 1.2.1). Biradicals F and G cyclize without diastereocontrol to deliver the [4+2]-cycloadducts biradical F forms a mixture of l 2trans,cis-[D]2-C and 12trans,trans [D]2-C, since a rotation about the C2—C3 bond is possible but not necessary. For the same reason, biradical G forms a mixture of 1 2cis,cis-[D]2-C and 12cis,trans [D]2-C. 

The same group has further shown that the alkenyl double bond geometry of geraniol- as well as nerol-derived ketones 41 and 43 remains unaffected during the photocyclizations. This suggests that the rate of cyclization is much higher than the rotation about the double bond in the allylic radical center of the biradical (Scheme 8.12). 

The energetic effects of conjugation are largest when empty or half-empty p-orbitals interact with a 7r-system. Typical examples include allyl cations or allyl radicals, respectively. In these cases, the allylic stabilization was estimated to be 20 kcal/mol.26 In comparison, the effect on neutral, closed-shell molecules is relatively small. The conjugative effect on the rotation of 1,3-butadiene 2 is, for example, with 3 kcal/mol much smaller. 

To illustrate the technique we will consider a few examples of free radicals which have been prepared in the rotating cryostat. In particular phenyl and acetyl radicals and methyl-substituted allyl radicals are of interest as they have not been trapped previously or identified with certainty. Since electron spin resonance has been used extensively to detect and identify the free radicals, account of the results will inevitably involve some description and analysis of their spectra, but we wish to focus the main discussion on the conclusions that can be drawn about structure and reactivity of the radicals. For information about the principles of e.s.r. and the interpretation of the spectra of free radicals the reader is referred to review articles and books on the subject (Symons, 1963 Norman and Gilbert, 1967 Maki, 1967 Horsfield, 1967 Carrington and McLachlan, 1967 Ayscough, 1967 Carrington and Luckhurst, 1968). 

The hyperfine splittings in the e.s.r. spectra of radicals of the allylic type are considerably less than those of alkyl radicals, and for radicals trapped in their parent compounds the resolution is insufficient to determine all the hyperfine coupling constants. However, by use of the rotating cryostat, the unsubstituted radical and three methyl-substituted allyl radicals have been prepared in a matrix of adamantane and it has been possible to resolve all the hyperfine couplings. 

---