Figure 1. Variation, with pH, of the specific rate of reduction of vitamin B12a (hydroxocobalamin) (2.7 X 10-4 M) with formate at 25 °C and µ = 1.0 M (LiC104). Reactions were carried out under N2 in solutions buffered by phthalate, acetate, phosphate, and borate (see Table III).The solid line represents pseudo-first-order rate constants, calculated from eq 4, with [formate], taken as 0.0810 M, as 6.15 X 10-9 M, and KA as 2.95 x 10-4 M; the circles represent experimental values.Attempts to apply this reaction to other formyl species led to results summarized in Table V, which emphasize the unique effectiveness of formate. The other carboxylates would be expected to coordinate with B12a in the same manner as formate and acetate, but the formyl group is generally not properly positioned for migration to Co111. Modest activity is observed in the case of o-formylbenzoic acid, which may react through a somewhat strained transition state I (featuring a seven-membered ring). The complex of glyoxylic acid, HC(=0)C00H, appears to offer a more favorable orientation for internal hydride migration, but reaction here may be inhibited by conversion of this acid by hydration to its gem-diol form (>0=0 + H20 -* >C(OH)2), a transformation that is known to be very nearly complete under our conditions.21Reduction of B12a by ascorbic acid is found to be rapid, but the biphasic profiles observed for this reaction, in conjunction with the known ability of this reductant to undergo both le and 2e oxidations,22 indicate the operation of a different, and more complex, mechanism for this process.Acknowledgment. We thank Aria White for technical assistance.