Benzyl halides, reactions with nickel carbonyl

The development of the Grignard-type addition to carbonyl compounds mediated by transition metals would be of interest as the compatibility with a variety of functionality would be expected under the reaction conditions employed. One example has been reported on the addition of allyl halides to aldehydes in the presence of cobalt or nickel metal however, yields were low (up to 22%). Benzylic nickel halides prepared in situ by the oxidative addition of benzyl halides to metallic nickel were found to add to benzil and give the corresponding 3-hydroxyketones in high yields(46). The reaction appears to be quite general and will tolerate a wide range of functionality. 

Aromatic halides are reported to give only carbonylated products with nickel tetracarbonyl. In contrast, pentafluorophenyl iodide in DMF gives decafluorobiphenyl in 70% yield (27). From the other products obtained (pentafluorobenzene, decafluorobenzophenone) it has been suggested that a radical mechanism is involved. The reactions of benzyl halides with nickel carbonyl in various solvents have been reported (28). The main reaction involves carbonylation, as discussed in Section III. Using benzene as solvent, a 33% yield of bibenzyl may be obtained. Here again a mechanism involving a 7r-allylnickel derivative should perhaps be considered, particularly since such a system is known to exist in (XVI) (29). 

Benzyl halides have been reported to react with nickel carbonyl to give both coupling and carbonylation (59). Carbonylation is the principal reaction in polar nonaromatic solvents, giving ethyl phenylacetate in ethanol, and bibenzyl ketone in DMF. The reaction course is probably similar to that of allylic halides. Pentafluorophenyl iodide gives a mixture of coupled product and decafluorobenzophenone. A radical mechanism has been proposed (60). Aromatic iodides are readily carbonylated by nickel carbonyl to give esters in alcoholic solvents or diketones in ethereal solvent (57). Mixtures of carbon monoxide and acetylene react less readily with iodobenzene, and it is only at 320° C and 30 atm pressure that a high yield of benzoyl propionate can be obtained (61). Under the reaction conditions used, the... 

In 1970 Hashimoto published a report on the reaction of potassium hexacy-anodinickelate with organic halides in aqueous solutions [295]. Benzyl bromides were transformed into dibenzyl ketones in the presence of CO in a water-acetone solution. If the reaction was carried out in a water-methanol solution, trans-fi-bromostyrene was transformed into methyl trani-cinnamat. Surprisingly, cinna-maldehyde was also formed in a 10 % yield (Scheme 2.47). The reaction of nickel carbonyl [Ni(CO)4l with organic halides was studied by Bauld in 1963 [296]. Aryl iodides were reacted with Ni(CO)4 in methanol and produced the corresponding methyl benzoate in good yields. If the reaction was carried out in THF, arils were formed. The reaction of allyl halides with Ni(CO)4 in the presence of MgO will produce but-3-enylsuccinic acid [297]. 

In spite of the usefulness of these complexes, it is generally not possible to cause the satisfactory reaction with transition metals in the metallic state [86] under mild conditions due to their poor reactivity. We have reported that activated metallic nickel, prepared by the reduction of nickel halide with lithium, underwent oxidative addition of benzylic halides to give homocoupled products [45]. We reported that carbonylation of the oxidative adducts of benzylic halides to the nickel proceeded smoothly to afford symmetrical 1,3-diarylpro-pan-2-ones in moderate yields, in which the carbonyl groups of alkyl oxalyl chlorides served as a source of carbon monoxide [43] see Equation 7.5. 

Palladium complexes also catalyze the carbonylation of halides. Aryl (see 13-13), vinylic, benzylic, and allylic halides (especially iodides) can be converted to carboxylic esters with CO, an alcohol or alkoxide, and a palladium complex. Similar reactivity was reported with vinyl triflates. Use of an amine instead of the alcohol or alkoxide leads to an amide. Reaction with an amine, AJBN, CO, and a tetraalkyltin catalyst also leads to an amide. Similar reaction with an alcohol, under Xe irradiation, leads to the ester. Benzylic and allylic halides were converted to carboxylic acids electrocatalytically, with CO and a cobalt imine complex. Vinylic halides were similarly converted with CO and nickel cyanide, under phase-transfer conditions. ... 

The hydroxycarbonylations (carboxylations) of alkyl, aryl, benzyl and allyl halides are from a retrosynthetic and mechanistic standpoint closely related. This type of reaction is widely used in organic synthesis [6], although a stoichiometric amount of salt by-product makes these methods less attractive on a large scale. The use of water-soluble catalysts for carbonylation of organic halides was scarcely studied in the past. Up to now palladium, cobalt, and nickel compounds in combination with water-soluble ligands have been used as catalysts for various carboxylations. 

In contrast with the results obtained with simple allqfl halides, benzyl bromide leads to the formation of 77 and the ketone 78 in variable ratios (Scheme 26). A similar result has been reported in the reactions between the oxidative addition product of Ni(COD)bpy or Ni(COD)TMEDA with cw-4-cyclohexen-l,2-dicarboxylic anhydride and alkyl iodidesWith allyl bromide as the electrophile, ketone 79 is the only product isolated. However, when the reaction is performed with isolated nickelacycle 66 in the absence of Ni(CO)2Me2Phen, allylated alanine 80 is formed exclusively (60% yield) (Scheme 26). These results show that the carbonyl nickel complex is not inert because with certain reagents it transfers CO to the nickeMactone 66. Alternatively, the formation of ketones in these reactions could be explained by alkylation of the primary oxidative addition product or by carbonylation of allyl or benzyl bromide to give acyl bromides which react with 66 to give the observed products. However, this last reaction pathway seems unlikely because acetyl or benzoyl chloride do not react with in situ generated nickelacycle 66. 

---