Electron-transfer reactions with neutral metal compounds

The reactions are applied as photosensitizers of polymerization. If electron transfer occurs with neutral vinyl compounds, an ion-radical instead of a radical should be generated. In this case, the highly selective initiation depending upon the metal complex-monomer pair and the possibility of ionic propagation must be envisaged. 

Dissolving metals initially convert aldehydes, ketones, and esters into radical anions. Subsequently, proton donors may react with the latter, which leads to neutral radicals. This mode of reaction is used, for example, in the drying of THF or ether with potassium in the presence of the indicator benzophenone. Potassium and benzophenone react to give the deep-blue potassium ketyl radical anion A (Figure 14.45). Water then protonates ketyl A to the hydroxylated radical B as long as traces of water remain. Further potassium reduces B via another electron transfer to the hydroxysubstituted organopotassium compound C. C immediately tautomerizes to the potassium alkox-ide D. Once all the water has been consumed, no newly formed ketyl A can be pro-tonated so that its blue color indicates that drying is complete. 

Upon reaction with a metal, the NMR spectrum of a neutral hydrocarbon first disappears, due to one-electron transfer from the metal. This step can be studied by EPR spectroscopy [la, 5j. The reduction of carbon-rich compounds usually occurs stepwise by one-electron transfers as depicted in Scheme 13.1 [lb]. The same set of events occurs in electrochemical processes and in alkali metal reductions. This reaction has become the main synthetic route to i-conjugated anions. 

Probably the most familiar radical reactions leading to 1,2-D systems are the so called acyloin condensation and the different variants of the "pinacol condensation". Both types of condensation involve an electron-transfer from a metal atom to a carbonyl compound (whether an ester or an aldehyde or a ketone) to give a radical anion which either dimerises directly, if the concentration of the species is very high, or more generally it reacts with the starting neutral carbonyl compound and then a second electron is transferred from the metal to the radical dimer species (for an alternative mechanism of the acyloin condensation, see Bloomfield, 1975 [29]). 

It is clear from the examples reported that carbon monoxide, when coordinated to a metal in a neutral complex, is not sufficiently activated to react with organic nitro compounds under mild conditions. More precisely, the first act of this reaction is the electron transfer from the metal to the nitro group to give a radical couple and this requires a very basic metal. This explains why basic ligands usually activate transition metal carbonyls in these catalytic reactions. Moreover, basic ligands such as Bipy favor the in-situ formation of the [Rh(CO)4] species from rhodium clusters. The effect of co-catalysts such as halide anions is more subtle, but even the action of these might, at least in part, be directed toward an increase of the electron density of the metal. 

The reductive coupling of carbonyl compounds with active metals (Na, Mg, Al) yields pinacols. An electron transfer from the metal surface to the carbonyl oxygen (ketyl formation), a soft-soft interaction, is undoubtedly involved. The conversion of esters to acyloins (22, 23) on the surface of metallic sodium is well known. Here the enediolate products can be trapped in situ by Me3SiCl (24). The chlorosilane does not interfere with the coupling, yet it effectively removes the alkoxide ions and neutralizes the enediolate ions immediately on formation. The elimination of RO is imperative, for otherwise Claisen or Dieckmann condensations would compete with the normal course of reaction. These complicating processes require a hard base (e.g. RO ) to abstract a proton from the starting esters, whereas the desired coupling is accomplished by a soft base which is the electrons on the metal surface. 

---