Electron paramagnetic resonance spectroscopic study of radicals derived from glycine, dl-α-alanine, and β-alanine in aqueous solution

Abstract

The substrates glycine, dl-α-alanine, and β-alanine have been examined in turn within an aqueous continuous-flow system using the titanous chloride – hydrogen peroxide radical-generating method. The temperature was 25 ± 2 °C and the pH of the reaction mixture was varied within the range of values ca. 2 to ca. 5 by addition of sulfuric acid. The electron paramagnetic resonance spectra of the following transient radicals were observed: from glycine, •CH(NH2)COOH; from dl-α-alanine, CH3Ċ(NH2)COOH and •CH2CH(NH3+)COOH; and from β-alanine, CH2(NH3+)ĊHCOOH. The structure of the radical from glycine has been established as •CH(NH2)COOH, not •CH(NH3+)COO−; in clarification of conflicting analyses by earlier workers, who employed reaction conditions similar to those of this present study but who did not obtain spectra as well resolved as those reported here. One group of these earlier workers claimed that glycine gave the radicals •CH(NH3+)COO− and •CH2COO− and the other group postulated correctly that the radical formed was •CH(NH2)COOH. The structural assignment •CH(NH2)COOH is confirmed by the observation in the present investigation of a similarly-structured radical, CH3Ċ(NH2)COOH, from dl-α-alanine. This latter radical was not found by the earlier workers. An additional, small, previously unreported proton hyperfine coupling has been observed in the spectrum from •CH(NH2)COOH when no sulfuric acid is added to the reaction mixture, i.e. pH ca. 4. It has been assigned to the carboxyl-group proton in the above radical. This appears to be the first time such an assignment has been made, and hence, arguments have been cited for its justification. The coupling-constant data obtained for the •CH(NH2)COOH radical are markedly different from those for the radicals •CH(NH3+)COC− and •CH(NH2)COO− reported by other investigators to be formed from glycine under reaction conditions clearly unlike those employed in this present study, even though all three formulas differ only with respect to protonation. These differences are discussed.