Enantioselective synthesis of (3R)- and (3S)-piperazic acids. The comparative unimportance of DMPU mediated retro-hydrazination

In response to a recent literature report by Decicco and Leathers (Ref. 13), the work of Hale, Delisser, and Manaviazar (1992) on the asymmetric synthesis of (3R)- and (3S)-piperazic acids has been reinvestigated, and the originally claimed product yields fully substantiated. The claims made in reference 13 about the proportions of cyclised product 6 and starting bromide 20 isolated from the low temperature electrophilic hydrazination-nucleophilic cyclisation of 20 with di-t-butylazodicarboxylate (DEAD) and DMPU as an additive are inaccurate. The retro-hydrazination reaction that they claim is problematic when DMPU is added to the hydrazinated reaction mixture has been demonstrated not to have a seriously detrimental effect on cyclisation product yield and to be unimportant. The other main assertion of reference 13, that the electrophilic hydrazination and nucleophilic cyclisation of 20 gives 6 in 91% isolated yield when, n-Bu(4)NI is employed as an additive (instead of DMPU) has also been shown to be in error. We have carefully repeated a scaled-down version of the n-Bu(4)NI catalysed procedure (Ref. 13) and have found that 6 is generally isolated in yields of 50-56% after flash chromatography. We have concluded that n-Bu(4)NI does not significantly increase the yields of cyclisation products 6 or 17 when it is employed as a cyclisation additive. Herein, we report details of our two preferred ''crude'' experimental procedures for preparing the enantiomers of piperazic acid in high optical purity, neither of which requires chromatographic purification of the reaction intermediates en route. Both these preferred ''crude'' methods for preparing 11 and 19 have been consistently reproduced many times in these laboratories over the past few years. In our view, they remain the most expedient and highest yielding methods currently available for obtaining 11 and 19 in high optical purity.