Electron paramagnetic resonance (EPR) spectroscopy using
sterically
hindered amine is extensively applied to detect singlet oxygen (1O2) possibly generated in advanced oxidation processes.
However, EPR-detectable 1O2 signals were observed
in not only the 1O2-dominated hydrogen peroxide
(H2O2)/hypochlorite (NaClO) reaction but surprisingly
also the 1O2-absent Fe(II)/H2O2, UV/H2O2, and ferrate [Fe(VI)] process
with even stronger intensities. By taking advantage of the characteristic
reaction between 1O2 and 9,10-diphenyl-anthracene
and near-infrared phosphorescent emission of 1O2, 1O2 was excluded in the Fe(II)/H2O2, UV/H2O2, and Fe(VI) process.
The false detection of 1O2 was ascribed to the
direct oxidation of hindered amine to piperidyl radical by reactive
species [e.g., •OH and Fe(VI)/Fe(V)/Fe(IV)] via
hydrogen transfer, followed by molecular oxygen addition (forming
a piperidylperoxyl radical) and back reaction with piperidyl radical
to generate a nitroxide radical, as evidenced by the successful identification
of a piperidyl radical intermediate at 100 K and theoretical calculations.
Moreover, compared to the highly oxidative species (e.g., •OH and high-valence Fe), the much lower reactivity of 1O2 and the profound nonradiative relaxation of 1O2 in H2O resulted it too selective and inefficient
in organic contaminant destruction. This study demonstrated that EPR-based 1O2 detection could be remarkably misled by common
oxidative species and thereby jeopardize the understandings on 1O2.