Through the Alkylation of Hydrogenphosphonates and Hydrogenphosphinates

Within this area, the most recent developments in the synthesis of esters of phosphonic acids have been the direct alkylation of hydrogenphosphonates using diazoalkanes in the presence of copper-containing catalysts in benzene as the solvent Of those catalysts examined, the most effective seem to be [Cu(acac)2] and [Cu(OTf)2], with Cu, Pd and Rh acetates and [Ni(acac)2] being less effective. The overall reaction is that represented in equation 18, in which R and R may be H, Ph or a simple alkyl group, but they may also consist of a functionalized alkyl group in reactions catalysed by trifluoromethanesul-phonic acid A similar procedure has been applied to the hydrogenphosphinate Ph(MeO)P(0)H  

The classical procedure, and the one still extensively employed, consists in the alkylation of compounds containing the P(0)H moiety, as an appropriate metal salt, with an alkyl halide or similar type of compound such a procedure can sometimes be a successful alternative when the classical Michaelis-Arbuzov reaction fails, one such example being illustrated in equation 19. No reaction takes place between triethyl phosphite and 3-chloro-cyclopentadiene at below 120 °C, above which the main reaction is then dehydrochlorination the use of sodium dialkyl phosphites leads, however, to the desired dialkyl cyclopent-2-enylphosphonates.  

Examples of the high reactivity of benzylic halides and of allylic halides have been reported. In the latter case, the well established S V rearrangement occurs when a secondary or tertiary allyl halide is used, and this leads to the same products as are obtained from the isomeric primary halide (equation 20), Surprisingly, 3-phenylprop-2-enyl halides afford only low yields in sluggish reactions .  

Although it might be expected that reactions which employed triarylmethyl halides would occur very readily, such reactions are rendered potentially more complex by the known nature of the halides and their propensity for involvement in free radical reactions. Whereas normal alkylation proceeds between sodium diethyl phosphite and diphenyl-methyl halides, success, or otherwise, in the use of the triphenylmethyl halides depends to some extent on the individual halide and on the metal in the phosphite salt. Thus, in an early study (in 1939), Arbuzov found that in reactions between silver dialkyl phosphites and triphenylmethyl bromide, dialkyl triphenylmethylphosphonates were indeed formed, but the use of the corresponding alkyl chloride provided the phosphite triester instead (metal dialkyl phosphites possess ambident anions ). A later study confirmed the behaviour of the silver salts towards the chloride, but also showed that, whereas dialkyl phophites with primary alkyl groups yielded phosphonic diesters (as had already been found), those with secondary alkyl groups afforded phosphite triesters moreover, the presence and nature of aromatic substituents were also able to control the course of the reaction. Reactions which involve triarylmethyl halides and sodium dialkyl phosphites may well be of a free radical nature since repeated studies have demonstrated the forma- 

Those Michaelis-Becker reactions between even relatively simple primary or benzylic-type halides and sodium dialkyl phosphites are not without their unwanted side-reactions. Halomethylfurylcarboxylic esters, for example, undergo concomitant dehalogenation or Michaelis-Becker phosphonation (the two processes may also occur side by side) depending on the relative positions in the furan nucleus of both carboxylic ester and halomethyl groups and on the halogen. Chlorides react normally, bromides do not . 

---