Heterogeneous
catalysts are often “black boxes” due
to the insufficient understanding of the detailed mechanisms at the
catalytic sites. An atomic-level elucidation of the processes taking
place in those regions is, thus, mandatory to produce robust and selective
heterogeneous catalysts. We have improved the description of the whole
reactive scenario for polymeric carbon nitrides (PCN) by combining
atomic-level characterizations with magic-angle spinning (MAS) solid-state
nuclear magnetic resonance (NMR) spectroscopy, classical reactive
molecular dynamics (RMD) simulations, and quantum chemistry (QC) calculations.
We disclose the structure–property relationships of an ad hoc
modified PCN by inserting an aryl amino group that turned out to be
very efficient for the production of H2O2. The
main advancement of this work is the development of a difluoromethylene-substituted
aryl amino PCN to generate H2O2 at a rate of
2.0 mM·h–1 under the irradiation of household
blue LEDs and the identification of possible active catalytic sites
with the aid of 15N and 19F MAS solid-state
NMR without using any expensive labeling reagent. RMD simulations
and QC calculations confirm and further extend the experimental descriptions
by revealing the role and locations of the identified functionalities,
namely, NH linkers, −NH2 terminal groups, and difluoromethylene
units, reactants, and products.