Mechanistic and synthetic studies with alkylcobaloximes : a new method for preparing oxygen-labelled aldehydes and ketones

• Main Author: Wong, Ah Kee
• Published: University of Warwick 1982
• Subjects: 572; QD Chemistry
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.330757

1. There is a lack of suitable methods for enriching aldehydes and ketones at the oxygen position. We have shown that imine (-C-N-) derivatives (iminium salts, N-l-haloalkylamides and enamides) are readily hydrolysed with 1 mol. equiv. of [170]/180]water to yield the parent [170)/(1801-labelled carbonyl compound. We prepared as examples 31 atom-% [170]BuCH0, 23 atom-% [180]cyclopentanone and [180)PhCH0. A sterically hindered ketone (camphor) or a branching ketone (e.g. pentan-2-one) may also be prepared by our methods. Enamides, as we have found, are generally applicable for labelling many aldehydes and ketones and hence provide a convenient route to many oxygen-enriched alcohols. 2. Our methods were extended to prepare 7 atom-% [170] BrCI2CHO and [170Ac]-2acetoxyethyl-2 (pyridine)cobaloxime. A highly enriched 31 atom-% [170-formyl)formylmethyl(pyridine)cobaloxime was also prepared. 3. Solvolysis experiments on these highly enriched compounds indicated that: (i), in phosphate buffer (pH 6.8), the hydration of [170]pentanal at 0 °C has a half-life of 1 min, whereas the cobalt complex [170-formyl)formylmethyl(pyridine)cobaloxime has a half-life of 2 h at 50 °C (ii), in aqueous dioxan, the hydrolysis of (170Ac]-2-acetoxyethyl(pyridine)cobaloxime at 50 °C has a half life of 3 h and this hydrolysis proceeds by an unusual B^lI route. That the hydration of formylmethylcobaloxime is slow relative to aliphatic or even aromatic aldehydes or ketones is due to a interaction between the cobalt atom and the neighbouring -*9"° bond (o ton hyperconjugation). The preference for an AL route over an AC route is consistent with the intervention of a n-complex intermediate. 4. Although 25 atom-% [180)BuCH0 lost its 180-label rapidly in citrate-phosphate buffers (pH 4.2) compared to 23 atom-% [180]PhCH0, there is evidence to suggest that the former can be rapidly reduced to [180]BuCHg0H in the presence of an excess of NADH and HLADH enzyme at this pH. These results implied that had a cobaloxime model such as H0CHRCH(18OH)(CHg)gCo(dmgH)8Py (R ■ H or Ph), been prepared, then photolysis in aqueous acetic acid to give the *80-labelled aldehyde, followed by trapping and detection of the extent of 180-content in the alcohol product, would have measured the amount of 1,2-OH shift- 5. 4,4-Diethoxycarbonylpentyl(pyridine)cobaloxime was photolysed in aqueous acetic acid (pH 3) to induce radical species. However, we d'd not detect the radical rearrangement product diethyl propylsucclnate by GLC analysis with an authentic sample. Thus, preliminary findings show that this cobaloxime is not a suitable model to parallel the 1,2-carbon skeletal shift catalysed by methylmalonyl-CoA mutase. 6. We have shown that the photolysis of octyl(pyridine)cobaloxime in ethanol gives > 98 % oct-l-ene by GLC analysis. Similarly, photolysis of 2-hydroxyhexyl(pyridine)cobaloxime in water gave a quantitative yield of hexan-2-one, whereas the acid-catalysed fragmentation in CDCI3 occurs readily to form hex-l-ene. Alkylcobaloximea are readily prepared from an alkylhalide or tosylate and thus provides a simple route to an alk-l-ene, an efficient process, for which, as a literature survey shows, is generally lacking.