Metal-free Photocatalytic Aerobic Oxidative Cleavage of C?C Bonds in 1,2-Diols

Article abstract of DOI:10.1002/cssc.202001466

The preparation of carbonyl compounds by the aerobic oxidative cleavage of C?C bonds in 1,2-diols under mild reaction conditions is a very significant reaction and is widely employed in various scenarios. Avoiding the use of harmful stoichiometric oxidants and adopting a greener chemical process remain a challenge for this reaction to date. In this manuscript, a heterogeneous metal-free photocatalytic strategy without any additive was developed for aerobic oxidative cleavage of C?C bonds in 1,2-diols at ambient conditions with visible light. The reaction mechanism was further studied through a series of control experiments and density functional theory (DFT) calculations. In addition, the catalytic system showed a broad substrates scope, including aliphatic (linear or cyclic) 1,2-diols, benzylic, alkenyl 1,2-diols, and α-hydroxy acids (such as lactic acid). Thus, this strategy could serve as a method for the transformation of 1,2-diols to corresponding carbonyl compounds by the aerobic oxidative cleavage of C?C bonds.

Full text of DOI:10.1002/cssc.202001466

Chemistry–Sustainability–Energy–Materials
Accepted Article
Title: Metal-free Photocatalytic Aerobic Oxidative Cleavage of C-C
Bonds in 1,2-Diols
Authors: Rui Zhu, Gongyu Zhou, Jia-nan Teng, Xinglong Li, and Yao
Fu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.202001466
Link to VoR: https://doi.org/10.1002/cssc.202001466
0
1/2020

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
Metal-free Photocatalytic Aerobic Oxidative Cleavage of C-C
Bonds in 1,2-Diols
[
a]
[a]
[a]
[a]
[a]
Rui Zhu , Gongyu Zhou , Jia-nan Teng , Xinglong Li , Yao Fu*
[a]
R. Zhu, G. Zhou, J. Teng, X. Li, Prof. Dr. Y. Fu
Hefei National Laboratory for Physical Sciences at the Microscale, iChEM CAS Key Laboratory of Urban Pollutant Conversion
Anhui Province Key Laboratory of Biomass Clean Energy, University of Science and Technology of China
Hefei, Anhui, 230026 (P. R. China)
E-mail: fuyao@ustc.edu.cn
Supporting information for this article is given via a link at the end of the document.
workers in 2017[ , and a homogeneous vanadium catalyst
was developed by Licini and co-workers in 2018[ . These
two catalytic systems are capable of efficiently cleaving C-C
bonds in 1,2-diols under 80–100 °C using ambient pressure
18]
Abstract: The preparation of carbonyl compounds by the aerobic
oxidative cleavage of C–C bonds in 1,2-diols under mild reaction
conditions is a very significant reaction and it is widely employed in
various scenarios. Avoiding the use of harmful stoichiometric
oxidants and adopting a greener chemical process are still the
19]
2
of O . However, these two noble metal-free catalytic
challenge for this reaction to date. In this manuscript,
a
systems were still limited to reaction temperatures and
posttreatments for product collection. Another recent work
was reported by Gao and co-workers[ , in which cobalt-
based heterogeneous catalyst exhibited excellent
performance on the oxidative cleavage of some 1,2-diols
into corresponding aldehydes or carboxylates (Scheme 1b).
heterogeneous metal-free photocatalytic strategy without any
additive was developed for aerobic oxidative cleavage of C–C bonds
in 1,2-diols at ambient conditions with visible light. The reaction
20]
mechanism was further studied through
experiments and DFT (Density Functional Theory) calculations. In
addition, the catalytic system has broad substrates scope,
a series of control
It should be noted that the work of Anastas[ and Gao
18]
[20]
a
including aliphatic (linear or cyclic) 1,2-diols, benzylic, alkenyl 1,2-
diols and α-hydroxy acids (such as lactic acid). Thus, this strategy
could serve as a method for the transformation of 1,2-diols to
corresponding carbonyl compounds by the aerobic oxidative
cleavage of C–C bonds.
was not applicable to a wide scope of substrates, which
limited to benzyl diols. In addition, Li and co-workers
[
21]
reported that the silver (I)-based catalyst could efficiently
cleave C-C bonds in 1,2-diols under mild conditions
(Scheme 1c). However, the critical drawbacks persisted as
the usage of excess amount of base and a complicated
post-processing. Although many catalytic strategies[
have also been developed to achieve this catalytic reaction,
there are still many limitations. Hence, a catalytic strategy
of high atomic economy with broad applicability, mild
reaction conditions and simple process needs further
development.
22-26]
Introduction
The aerobic oxidative cleavage of C–C bonds in 1,2-diols
to prepare corresponding carbonyl compounds (carboxylic
acid, aldehyde, ketone) is a very significant fundamental
reaction in synthetic organic chemistry. This reaction is
widely employed in various scenarios, such as organic
synthesis , biomass valorization[2-4], analytical studies in
[
1]
[
5]
biochemistry and so on. In the early 20th century, the
Malaprade reaction[ and the Criegee oxidation[7] were the
classical methods to transform 1,2-diols to aldehydes or
ketones. Traditional methods for effective cleavage of C-C
bonds in 1,2-diols require the consumption of stoichiometric
6]
[
8-
or excess amount of oxidants, such as high-valent iodine
10]
and lead[11]. These toxic reagents not only damage
human health, reduce atom economy, but also generate
equimolar quantities of hazardous waste, which is harmful
to the environment. Hence, a catalytic system with high
atom economy and green reaction process is urgent to be
developed. In this context, many improved catalytic
systems have been proposed using the widely accepted
clean oxidant molecular oxygen[12-17]. Some of these
protocols were able to selectively cleave the C-C bond in
specific 1,2-diols, but most of them required the use of toxic
metal reagents or harsh reaction conditions. Driven by the
need for low-cost conversion and high atomic economy, a
heterogeneous catalyst, sodium-manganese layered mixed
oxide (Scheme 1b), was developed by Anastas and co-
Scheme 1. Methods for the oxidative cleavage of 1,2-diols to
corresponding carbonyl compounds.
1
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
In recent years, semiconductor materials have been
extensively studied[27]. These photosensitive materials were
attempted to conduct photocatalytic reactions in various
organic synthesis driven by the need of clean and
renewable organic synthesis strategy. In these
semiconductor materials, graphitic carbon nitride has
attracted much more attention in the field of hydrogen
evolution reaction[28], biomass valorization[29] and organic
[
30-31]
synthesis
. Properties of large surface area, low cost,
easy functionalization, visible light adsorption, plenty of
nitrogen atoms or vacancies and high photocatalytic activity
make C
3 4
N one of the most promising catalysts for visible
light photoredox catalysis, and the flexible organo two-
[
42]
dimension character of C
N
3 4
may provide an ideal
surface for many substrates[29]. In this work, a base-free
and metal-free strategy (Scheme 1d) was proposed for the
aerobic oxidative cleavage of C-C bonds in 1,2-diols to
corresponding carbonyl compounds catalyzed by
mesoporous carbon nitride (mpg-C
conditions (visible light, room temperature, O
3
N
4
)
under mild
balloon).
2
Reaction conditions were optimized and the reaction
mechanism was further investigated through a series of
control experiments and DFT (Density Functional Theory)
calculations, exhibiting the interaction between the catalyst
and the substrate. The catalytic system also has a broad
substrate scope, including aliphatic 1,2-diols, benzylic 1,2-
diols, alkenyl 1,2-diols and α-hydroxy acids, such as lactic
acid.
Figure 1. Catalytic oxidative cleavage of 1a to 2a and 3a with different
catalysts. Reaction conditions: 0.2 mmol 1a, 10 mg catalyst, 1 mL MeCN,
2
390 nm LED, O balloon, room temperature, 12 h. The conversion and
yield of this reaction were detected by gas chromatography (GC).
A series of solvents were adopted to investigate the
solvent effect on the photocatalytic oxidative cleavage of 1a.
As shown in Figure 2, DCM (CH Cl ) was the optimal
2 2
Results and Discussion
Initially,
adopted for photocatalytic oxidative cleavage of a model
substrate 1,2-diphenylethane-1,2-diol 1a at room
temperature with a 390 nm light source. As shown in Figure
a, using CH CN as the solvent and O as the oxidant,
mpg-C showed the best activity on the conversion of 1a
95%) and yield of 2a (28%) and 3a (34%). Under identical
conditions, 1a conversion decreased to about 40-60% over
WO and Fe catalysts. In addition, SnO , CdS, and Mo
showed about 10% conversion of 1a. When Al or TiO
were used as the catalyst, 1a could hardly react and the
conversion was lower than 5%. It is speculated that the
photocatalytic activity of a catalyst is closely related to its
absorbance of light, as electrons will be emitted under the
interaction with photons. Therefore, UV-visible (UV-Vis)
absorption spectra were obtained for all the tested
semiconductor materials, as shown in Figure S2d. As
a
series of semiconductor materials were
solvent obtaining 99% conversion of 1a and 81% yield of 3a.
Although a conversion of higher than 90% was obtained
when using CH CN and dimethyl sulfoxide (DMSO), the
3
(
)
yield of 3a was lower, and a considerable amount of by-
product 2a was detected. As DMSO itself could be oxidized
1
3
2
[
32]
in some photocatalytic reaction system , it was not
suitable to be employed as the solvent in this reaction
3 4
N
(
system. In the alcohol solvents (such as MeOH), a few of
carboxylates were detected, leading to a low yield of target
product 3a. In addition, the alcohol solvent, MeOH, could be
activated to alkoxy radicals in some photocatalytic
systems[ . The alcohol solvents have the potential to
compete with 1,2-diols under light conditions. Therefore,
alcohols are also not suitable as the solvents in this
reaction system. In addition, we speculated that the
interaction between the substrate, catalyst, and solvent is
also one of the factors that affect the experimental results.
In the following, through theoretical calculations, we find
that the active sites on the surface of the catalyst can
transfer electrons or holes through the interaction with the
hydroxyl group of the substrate, thereby catalyzing the
reaction. When the solvent used contains groups that are
likely to interact with hydroxyl groups, the interaction
between the substrate and the active site of the catalyst
may be affected, thereby reducing the catalytic activity of
the catalyst. In addition, when the solvent used is likely to
interact with the active site of the catalyst, the same effect
would be produced. As we all know, the hydroxyl group of
the substrate easily forms hydrogen bonds with atoms
3
2
O
3
2
2
S
2
O
3
2
33]
2 3 2
anticipated, Al O and TiO had poor absorbance at
wavelength of 390 nm. Therefore, it is reasonable that
these two catalysts showed rare activity in the
photocatalytic oxidative cleavage. WO
mpg-C had excellent absorbance at 390 nm light source.
While compared with mpg-C , their activity on 1a
3 2 3
, CdS, Fe O and
3 4
N
3 4
N
conversion was significantly lower. The UV-Vis results
indicated that for a photocatalyst, a good absorbance for
specific light source was necessary for a high photocatalytic
activity. However, the catalytic performance of a strong
absorbent material might be poor, which suggested the
nature of the catalyst, like activity sites, was another
determining factor on its photocatalytic activity.
(
such as O, N, S) with large electronegativity, thereby
affecting the interaction between the substrate and the
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
Figure 3. The aerobic oxidative cleavage of 1a to 2a and 3a with different
light sources.
Figure 2. The aerobic oxidative cleavage of 1a to 2a and 3a with different
solvents. DCM= CH Cl , MT= methanol, EA= ethyl acetate, THF=
2 2
tetrahydrofuran, DMF= N,N-dimethylformamide, DMSO= dimethyl
sulfoxide, GVL= γ-Valerolactone.
catalyst. Therefore, when we used MeOH, ethyl acetate
2
(EA), DMSO, N,N-dimethylformamide (DMF), H O, etc. as
the reaction solvent, we did not get the ideal experimental
results. In addition, when different solvents were used, the
product distribution of benzaldehyde (2a) and benzoic acid
(
3a) was different. From the following kinetic studies, we
knew that substrate 1a was first converted to benzaldehyde
2a), and then benzaldehyde (2a) was oxidized to benzoic
(
acid (3a). Both of these reactions were radical processes,
and both required the participation of catalysts. Therefore,
the above solvation effects seem to be applicable to both
processes.
In addition, the effect of light source irradiation
wavelength on the photocatalytic activity was explored
(
Figure 3). A variety of light radiation sources were
screened including 370 to 467 nm LEDs and sunlight, with
black condition as comparison. As shown in Figure 3, with
LED wavelengths ranging from 390 to 440 nm, a high
conversion of 1a with a good selectivity towards 3a was
achieved. As the wavelength increased to 456 and 467 nm,
the conversion and yield decreased continuously. These
results were aligned with the UV-Vis results, which
Figure 4. The aerobic oxidative cleavage of 1a to 2a and 3a with different
reaction time. Reaction conditions: 0.2 mmol 1a, 10 mg catalyst, 1 mL
DCM, 390 nm LED, O balloon, room temperature. The conversion and
2
yield of this reaction were detected by gas chromatography (GC).
catalyst has the best quantum efficiency under a light
source with a wavelength of 390 nm (Table S1).
3 4
suggested that mpg-C N had high absorbance for
The effect of reaction time on the generation of 3a from
substrate 1a was also investigated for further understanding
the reaction mechanism, as shown in Figure 4. With the
progress of reaction, conversion of 1a increased
monotonously in the first 6 hours, accompanied by an
increasing amount of 2a generated. At the full conversion of
wavelengths between 259 nm and 442 nm (Figure S2c). It
should be noted that under 370 nm light source, the
conversion and yield were much lower than those from 390-
440 nm, which was not consistent with the UV-Vis results.
This might be because that the glass tube used for this
reaction had inhibiting effect on 370 nm light source,
leading to a low transmittance of radiation source. Under
black condition, the conversion and yield were determined
as lower than 5%, indicating that this catalytic process was
indeed triggered by photons. In addition, when sunlight was
used as the radiation source, 22% 2a and 7% 3a were
obtained with a conversion of 63%. At the same time, we
also tested the quantum efficiency of the catalyst under
different light sources. The results also showed that the
1a, 2a was gradually converted into 3a under catalytic
conditions. Therefore, the yield of 2a decreased while the
yield of 3a increased. For the total yield of 2a and 3a, a
distinct drop was detected between the sixth hour and
seventh hour. This might be due to the conversion of 2a
into an undetected intermediate. In order to analyze the
possible intermediates, we performed ESI-MS (Figure S2f)
detection on the in-situ reaction solution. We could see from
3
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
3 4
Table 1. Photocatalytic cleavage of 1,2-diols over mpg-C N
. [a]
Yield [b]
Entry
Substrate
Product
Yield [b]
Entry
Substrate
Product
1
81%
15
16
17
18
19
20
72%
2
3
78%
70%
53%[d]
49%[d]
OH
OH
83%
88%
4
5
6
78%
88%
85%
Ph
1r
O
O
Ph
Cy
3
s
43%[d]
[
a] Reaction conditions: 0.2 mmol substrate, 10 mg mpg-C
nm LED, O balloon, room temperature, 12 h. [b] The yield was isolated yields.
c] The yield was determined by 1H NMR spectroscopy with the biphenyl as
3 4
N , DCM 1 mL, 390
2
[
7
8
75%
96%
the internal standard in D -DMSO. [d] The yield was determined by GC.
6
the figure that there were obviously characteristic peaks of
benzaldehyde (m/z = 105) and benzoic acid (m/z = 121). At
the same time, we also detected characteristic peaks of
other substances (such as m/z = 227, 301 etc.). From this
ESI-MS, we speculated that the acetal (m/z = 301) formed
by the reaction of produced aldehyde and substrate diol
could be the intermediate. And the ester compound (m/z =
9
2%
4%
9
227) formed by the produced radical II (Scheme 3) and
8
activated benzoic acid during the reaction may also be an
intermediate of the reaction. Overall, the kinetics study
indicated that 3a was converted from 2a under this catalytic
condition, and 2a was an intermediate for the entire
reaction. In addition, the recyclability test of this catalyst
was performed under five consecutive runs utilizing
identical conditions (Figure S2h). The slightly reduced
product yield was mainly caused by the loss of catalyst
during the catalyst transfer process. Nevertheless, the high
product yields observed herein indicated that the catalysts
could be recycled and reused several times.
O
1
0
1
91%
69%
Ph
Ph
3j
OH
1
Ph
OH
1k
With the optimal reaction conditions on hand (O
2
balloon,
1
2
3
48%[c]
69%[d]
room temperature, mpg-C as the photocatalyst, 390 nm
3 4
N
light, DCM as the solvent), the substrate scope of this
photocatalytic system for a variety of 1,2-diols was
examined, with results tabulated in Table 1. The mpg-C N
3 4
1
photocatalytic system was effective for a wide range of
substitute and non-substitute 1,2-diols. Generally, 1,2-diols
with electron withdrawing groups could be converted into
corresponding acids in higher yields compared with those
14
65%[d]
bearing electron donating groups (1b
–1f). The electron-
donating aromatic 1,2-diols (1b 1c) gave the target
–
4
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
Figure 5. Optimized interaction patterns (top view and side view) and adsorption energies of substrate 1a and 1y adsorbed on catalyst surface
Pattern A: a and d for H atom on hydroxyl group of substrate 1a; Pattern B: b and e for O atom on hydroxyl group of substrate 1a; Pattern C: c
and f for O atom on carbonyl group of substrate 1y).
(
products (3b–3c) in good yields, giving 4-methylbenzoic
acid (3b) in 78% yield and 4-methoxybenzoic acid (3c) in
7
1
0% yield respectively. The electron-withdrawing aromatic
,2-diols (1d 1f) afforded the target products (3d 3f) in
–
–
excellent yields, giving 4-trifluoromethylbenzoic acid (3d) in
8% yield, 4-fluorobenzoic acid (3e) in 88% yield, 4-
7
chlorobenzoic acid (3f) in 85% yield. Due to the steric
hindrance effect, yield of 3g was lower than 3f. The
polysubstituted 1h
,
1i and 1j was converted into
corresponding ketone (3i) in 96%, 92% and 91% yield
respectively. In addition, it should be emphasized that this
mpg-C
linear and cyclic) 1,2-diols
Particularly, a novel strategy for the synthesis of adipic acid
3l) was verified with substrate 1l, with a yield of 48% being
obtained. It should be noted that when mono-substituted o-
diols (substrate 1o 1q were used as the reaction
N
3 4
catalyzed system was also effective for aliphatic
(
(
1k 1q in good yields.
–
)
(
-
)
substrates, another product was formic acid (Figure S2g).
In addition, this catalytic system was also applicable to the
structure of α-hydroxy acids (1r
and lactic acid, giving
cyclohexyl(phenyl)methanone (3s) and acetic acid (3t) in
-1t), such as mandelic acid
benzoic acid 3r),
(
83%, 88% and 43% yield respectively.
The reaction mechanism was further explored through a
series of control experiments (Scheme 2). Under optimal
reaction conditions, almost full conversion of the model
substrate 1a was achieved. The main product was 3a in
81% yield with a tiny amount of 2a (entry a). When using α-
hydroxy ketone structure 1x as the model substrate, under
identical conditions, the conversion was only 64%, with the
yield of main product 3a being 56% (entry b). Further, with
o-dione structure 1y as the model substrate, the reaction
almost did not occur, resulting in a conversion lower than
5
% and trace amount of products being detected (entry c).
Scheme 2. Control experiments.
For heterogeneous catalysis, the effective interaction
between catalyst surface and substrate is the most key
5
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
point. In the skeleton of mesoporous carbon nitrides, a
variety of nitrogen functional groups would be the sites for
adsorption of substrate molecules[34]. To illustrate the
interactions between substrate molecules (1a and 1y) and
catalyst surface, DFT (Density Functional Theory)
calculations using a three-layer model were conducted, as
shown in Figure S3. A 4×4 C N sheet was used as the
3 4
model of catalyst. The geometry optimization result
suggests that a corrugated conformation is the most stable
3 4
structure of C N sheet. With optimized catalyst model,
surface interactions were further investigated as shown in
Figure 5 (other adsorption conformations can be found in
the Supporting Information, Figure S4). Interaction pattern
A demonstrates the H atom on hydroxyl group of 1a
approaching N atom on the edge of catalyst surface, with
the two benzene rings of 1a being nearly perpendicular.
Pattern B shows the O atom on hydroxyl group of 1a
approaching the central N atom on C N surface, while the
3 4
two benzene rings are nearly parallel. The adsorption
energy is -0.55 eV and -0.47 eV respectively, indicating that
3 4
substrate 1a has stable chemical adsorption states on C N
by H atom or O atom on the hydroxyl group. While for
substrate 1y, the adsorption energy of O atom on catalyst
surface is -0.28 eV, which might be attributed to the lack of
hydroxyl H atom on 1y structure compared with 1a. As the
adsorption energy is lower than 40 kJ/mol, an instable
physical adsorption of 1y on C N structure can be inferred.
3 4
The weak interaction between 1y and catalyst is not
beneficial for the electron transfer from catalyst to substrate,
which might be the reason for the trace conversion of 1y
.
The yield of acids from these three substrates are in
sequence of o-diol, α-hydroxy ketone and o-dione (Scheme
2
). Therefore, it can be concluded that the photocatalytic
reaction pathway was not the conventional process of first
oxidation into ketone[30] followed by C-C bond cleavage.
Scheme 3. a) Detection of Superoxide Radical (O
mpg-C using the spin-trap method. b) Proposed mechanism for the
photocatalytic cleavage of C-C bonds in 1,2-diols with mpg-C
•−) formed by irradiated
2
3
N
4
Combining with the kinetics study (Figure 4), it is
speculated that the direct radicalization of o-diol compound
first occurred, followed by the β-scission, which resulted in
the C-C bond cleavage to generate the corresponding
aldehyde compound. The aldehyde compound then
underwent further oxidation under identical conditions
forming corresponding carboxylic acid. At the same time,
we also found that the oxygen was critical to the cleavage
of C-C bonds in 1,2-diols. When using argon instead of
oxygen (entry d in Scheme 2), the substrate 1a could hardly
react, and the product benzoic acid 3a could rarely be
detected in the reaction system. In addition, to verify this
conjecture of radical reaction process, excessive amount of
radical scavenger TEMPO was added for 1a conversion
under optimal conditions (entry e). As TEMPO captured the
radicals that generated in the photocatalytic process, trace
amount of products 2a and 3a was detected although 1a
was almost fully converted. This result confirmed that the
photocatalytic oxidation of 1a followed the radical
mechanism. In addition, photocatalytic conversion of 2a
was also performed for mechanism study (entry f). With a
full conversion of 2a, the yield of corresponding carboxylic
acid product 3a was 92%, proving that the hypothesis of
first oxidation to aldehyde then forming acid was
reasonable, which was consistent with the speculation from
kinetics study that 2a was the intermediate of 1a conversion
3
4
N .
In order to confirm the active species of oxygen involved in
the present reaction, electron paramagnetic resonance
(EPR) experiments were performed with 5,5-dimethyl-1-
pyrroline N-oxide (DMPO) as the spin trapper. The spectra
•− [43]
(
Scheme 3a) indicated the formation of O
2
. No EPR
signal was detected in the dark, indicating that light was
•−
necessary for the formation of O
2
. Those data indicated
•−
that O
2
was the reactive oxygen species. Therefore, based
on the above results, a possible reaction mechanism was
proposed in Scheme 3b. This reaction was an oxidation
process driven by light radiation. Firstly, the photocatalyst
was excited by light radiation source, generating holes and
electrons separately in the valence and conduction band.
Under the effect of photocatalyst, O
2
was catalytically
-•. Simultaneously, the hydroxyl group was
converted into O
2
radicalized with releasing a hydrogen for substrate 1a
interacting with the photocatalyst, which then combined with
the oxygen radical forming HOO•. The formed radical
then cleaved through β-scission
I
was
in this reaction
conditions, generating a molecule of 2a and the radical II
[35]
.
Hydrogen atom transferred from radical II to the solvent or
combined with HOO• to be converted into HOOH and the
second benzaldehyde would be generated simultaneously.
Subsequently, the generated 2a was further oxidized into
to 3a
.
6
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
3
a
. The conversion of benzaldehyde to benzoic acid was a
atmosphere. Subsequently, hydrofluoric acid was used to
radical-induced oxidation process, which was a classic
fundamental reaction, thus was omitted in this manuscript.
remove SiO through stirring with the above material for 24
2
h. Then, after the following filtration, water and ethanol
washing for several times, and drying at 80 °C for about 12
h, the pale-yellow solid (about 4 g) was finally obtained and
Conclusion
3 4
denoted as mpg-C N .
Other semiconductor materials (shown in Figure 1a) are
purchased and used directly without any further treatment
in this work.
In summary, we have developed a heterogeneous metal-
free photocatalytic strategy to achieve aerobic oxidative
cleavage of C–C bonds in 1,2-diols at ambient conditions
with visible light using O
2
as the oxidant. Reaction
General procedure for Photocatalytic Reactions:
conditions were optimized, achieving the excellent yields of
corresponding products without any additive. Significantly,
this metal-free catalytic strategy has a broad substrate
scope, including aliphatic (linear or cyclic) 1,2-diols,
benzylic, alkenyl 1,2-diols and α-hydroxy acids, such as
lactic acid. In addition, the reaction mechanism was further
investigated through a series of control experiments and
DFT calculations. Mechanistic investigations revealed that
H or O atom on the hydroxyl group could be adsorbed on
To a 10 mL reaction tube with a stir bar was added the
substrate lactone (0.2 mmol) and then 10 mg catalyst mpg-
C N
3 4
was added to the reaction tube. Substrately, 1 mL
Cl was added as the solvent. The reaction mixture was
stirred for 12 h at room temperature with a 390 nm light
source at 1 atm O pressure (O Balloon). After completion
CH
2
2
2
2
of the reaction, the mixture was subjected to purification.
Evaporation of the solvent and short column
chromatography of the resulting crude material over silica
gel afforded the pure products. The conversion of
substrates and the products yield were calculated according
3 4
mpg-C N and generated the alkoxy radical. This would be
an approach that complements the method of other
previous work. And further applications of this catalytic
system to other oxidation processes are under investigation
in our laboratory.
to the following formula
：
Conversion = (n
0 1 0
-n )/n * 100%
Yield = n(actual yield)/n(theoretical yield) * 100%
: mol of substrate before reaction
: mol of substrate after reaction
n
0
n
1
Experimental Section
The apparent quantum efficiency (QE) was measured
Materials and instrumentation
under the same photocatalytic reaction condition. The QE
was calculated according to the following formula[
37]
:
All of the reagents and solvents were of analytical grade,
purchased from commercial sources (Alfa Aesar, sigma,
Adamas-beta, Energy Chemical, TCI and Aladdin) and
used without further purification. The Photo Reaction Setup
was purchased from Anhui Kemi machinery technology Co.,
Ltd. Gas chromatographic (GC) analysis was acquired on a
Shimadzu GC-2014 Series GC System equipped with a
flame-ionization detector. 1H-NMR and 13C-NMR spectra
were recorded on a Bruker Avance 400 spectrometer at
ambient temperature. The following abbreviations are used
to explain the multiplicities: s = singlet, d = doublet, t =
triplet, q = quartet, and m = multiplet. The powder X-ray
diffraction (XRD) patterns of the catalysts were measured
by an X’pert (PANalytical) diffractometer, using CuKα
radiation at 40 kV and 40 mA,with a 2θ range of 20-80°.
FTIR spectra were recorded on a FTIR/Raman Thermo
Nicolet 6700 spectrometer, in ATR mode. UV-Vis spectrum
was measured by UV-3600.
number of reacted electrons
QE 
 100%
number of incident electrons
number of converted substrates
number of incident electrons

 100%
General information for DFT calculations
Gaussian 16 package[38] were used for all DFT
calculations. The B3LYP functional[39-40] and 6-31G*[41] basis
set were employed for geometry optimization of reactants,
intermediates, and products. For the 4×4 ONIOM model of
catalyst, B3LYP/6-31G* was used for high layer calculation,
HF/6-31G* was used for medium layer, and PM6 was used
for the low layer. Then frequency analysis was performed at
the same level. The adsorption energy was calculated as
follows: Eads = E(reactant + catalyst) – E(free reactant
molecule) – E(free catalyst). In addition, the vibrational
frequencies were used to compute zero-point energy (ZPE)
corrections.
Preparation of mpg-C
The mesoporous graphitic carbon nitride (mpg-C
synthesized according to a modified method in the previous
3 4
N Catalysts
3 4
N ) was
[
36]
work . Briefly, about 40 g urea were dissolved in a Acknowledgements
solution of 0.2 M HCl solution (60 mL) and ethanol (52 mL)
under vigorous stirring, and tetraethyl orthosilicate (32 mL)
was then slowly added to the above solution drop by drop.
After stirring vigorously at room temperature for 3 h, the
mixture was heated under vacuum for the solvent
evaporation and then dried at 100 °C for about 12 h. The
obtained white solid was calcined in a muffle furnace at
This work was supported by the National Natural Science
Foundation of China (21572212, 51821006, 51961135104), the
National Key R&D Program of China (2018YFB1501604), the
Major Science and Technology Projects of Anhui Province
(18030701157), the Users with Excellence Program of Hefei
Science Center CAS (Grant 2019HSC-UE017), the Strategic
Priority Research Program of the CAS (XDA21060101), and the
550 °C for 4 h (heating rate: 2.5 °C/min) under nitrogen
7
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
[
36]L. Shi, L. Liang, F. X. Wang, M. S. Liu, K. L. Chen, K. N. Sun, N. Q. Zhang,
J. M. Sun, ACS Sustainable Chem. Eng. 2015, 3, 3412−3419.
37]J. Yu, J. Zhang, M. Jaroniec, Green Chem. 2010, 12, 1611-1614.
38]Gaussian 16, Revision A.01 (Gaussian Inc., Wallingford, CT, 2016).
Local Innovative and Research Teams Project of Guangdong
Pearl River Talents Program (2017BT01N092). The authors
thank the Hefei Leaf Biotech Co., Ltd. and Anhui Kemi
Machinery Technology Co., Ltd. for free samples and equipment
that benefited our ability to conduct this study.
[
[
[39]A.D. Becke, Phys. Rev. A. 1988, 38, 3098.
[
[
40]A. D. Becke, J. Chem. Phys.1993, 98, 1372-1377.
41]R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980,
7
2, 650-654.
42] Y. Chen, B.Wang, S. Lin, Y. Zhang, X. Wang, J. Phys. Chem. C 2014,
18, 29981-29989.
43] M. Zhou, S. B. Wang, P. J. Yang, C. J. Huang, X. C. Wang, ACS Catal.
018, 8, 4928−4936.
[
[
1
Keywords: aerobic oxidative cleavage • diols • carboxylic acids
•
metal-free photocatalysis • mesoporous carbon nitride
2
[
1] A. W. H. Wong, T. K. M. Shing, in Comprehensive Organic Synthesis II,
Second Edition), Vol. 7 (Eds.: P. Knochel, G. A. Molander), Elsevier,
(
Amsterdam, 2014, p. 801.
[2] H. C. Zhu, Y. Zhang, X. G. Yang, H. Y. Liu, X. M. Zhang, J. M. Yao, Ind.
Eng. Chem. Res. 2015, 54, 2825-2829.
[
[
3] K. S. Hanson, R. T. Baker, Acc. Chem. Res. 2015, 48, 2037-2048.
4] F. Napoly, L. Jean-Gérard, C. Goux-Henry, M. Draye, B. Andrioletti, Eur. J.
Org. Chem. 2014, 781-787.
[
5] A. Verdoliva, P. Bellofiore, V. Rivieccio, S. Catello, M. Colombo, C.
Albertoni, A. Rosi, B. Leoni, A. M. Anastasi, R. De Santis, J. Biol. Chem.
2010, 285, 9090-9099.
[
[
[
[
6] L. Malaprade, Bull. Soc. Chim. Fr. 1934, 3, 833-852.
7] R. Criegee, Ber. Dtsch. Chem. Ges. 1931, 64, 260-266.
8] M. Frigerio, M. Santagostino, Tetrahedron Lett, 1994, 35, 8019–8022.
9] J. N. Moorthy, N. Singhal, K. Senapati, Org. Biomol. Chem. 2007, 5, 767-
7
71.
10]A. Sudalai, A. Khenkin, R. Neumann, Org. Biomol. Chem. 2015, 13, 4374–
394.
11]J. I. C. Lena, O. Sesenoglu, N. Birlirakis, S. Arseniyadis, Tetrahedron Lett.
001, 42, 21–24.
12]E. Baer, J. M. Grosheintz, H. O. L. Fischer, J. Am. Chem. Soc. 1939, 61,
607–2609.
13]J. I. C. Lena, O. Sesenoglu, N. Birlirakis, S. Arseniyadis, Tetrahedron Lett.
001, 42, 21–24.
[
[
[
[
4
2
2
2
[
[
14]G. D. Du, L. K. Woo, J. Porphyrins Phthalocyanines 2005, 9, 206.
15]M. Kirihara, K. Yoshida, T. Noguchi, S. Naito, N. Matsumoto, Y. Ema, M.
Torii, Y. Ishizuka, I. Souta, Tetrahedron Lett. 2010, 51, 3619-3622
16]S. RiaÇo, D. Fernández, L. Fadini, Catal. Commun. 2008, 9, 1282-1285.
17]V. Escande, C. H. Lam, C. Grison, P. T. Anastas, ACS Sustainable Chem.
Eng. 2017, 5, 3214-3222.
[
[
[18]V. Escande, C. H. Lam, P. Coish, P. T. Anastas, Angew. Chem. Int. Ed.
2017, 56, 9561-9565.
[19]E. Amadio, J. González-Fabra, D. Carraro, W. Denis, B. Gjoka, C. Zonta,
K. Bartik, F. Cavani, S. Solmi, C. Bo and G. Licini, Adv. Synth. Catal. 2018,
3
60, 3286–3296.
20]H. Luo, L. Wang, S. Shang, J. Niu, S. Gao, Commun. Chem. 2019, 2, 17-
6.
21]Z. Zhou, M. Liu, L. Lv and C.-J. Li, Angew. Chem., Int. Ed. 2018, 57, 2616-
620.
[
[
2
2
[
[
22]S. M. Kim, D. W. Kim, J. W. Yang, Org. Lett. 2014, 16, 2876−2879.
23]N. García, R. Rubio-Presa, P. García-García, M. A. Fernández-Rodríguez,
M. R. Pedrosa, F. J. Arnáiz, R. Sanz, Green Chem. 2016, 18, 2335-2340.
24]J. Schwarz, B. König, Chem. Commun. 2019, 55, 486-488.
25]S. Solmi, E. Rozhko, A. Malmusi, T. Tabanelli, S. Albonetti, F. Basile, S.
Agnoli, F. Cavani, Appl. Catal. A Gen. 2018, 557, 89-98.
[
[
[26]N. Obara, S. Hirasawa, M. Tamura, Y. Nakagawa, K. Tomishige,
ChemCatChem 2016, 8, 1732–1738.
[
[
27]J. Li, N. Wu, Catal. Sci. Technol. 2015, 5, 1360-1384.
28]G. Liao, Y. Gong, L. Zhang, H. Gao, G. Yang , B. Fang, Energy Environ.
Sci. 2019, 12, 2080-2147.
[
[
[
[
29]H. Liu, H. Li, J. Lu, S. Zeng, M. Wang, N. Luo, S. Xu, F. Wang, ACS Catal.
2018, 8, 4761−4771.
30]F. Su, S. C. Mathew, G. Lipner, X. Fu, M. Antonietti, S. Blechert, X. Wang,
J. Am. Chem. Soc. 2010, 132, 16299−16301
31]F. Su, S. C. Mathew, L. Möhlmann, M. Antonietti, X. Wang, S. Blechert,
Angew. Chem., Int. Ed. 2011, 50, 657−660.
32]W. Schilling, D. Riemer, Y. Zhang, N. Hatami, S. Das, ACS Catal. 2018, 8,
5425−5430.
[
[
33]A. Hu, J. Guo, H. Pan, Z. Zuo, Science 2018, 361, 668–672.
34]Y. Wang, J. Yao, H. Li, D. Su, M. Antonietti, J. Am. Chem. Soc. 2011, 133 ,
2
362−2365.
35]A. Hu, Y. Chen, J. Guo, N. Yu, Q. An, Z. Zuo, J. Am. Chem. Soc. 2018,
40, 13580−13585.
[
1
8
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.202001466
FULL PAPER
Entry for the Table of Contents
HO
OH
Ph
O
Ph
Ph
Ph
Ph
Ph
C
N
OH
O
HO
OH
O
HO
No base;
No metal;
Wide substrates scope;
Mild reaction conditions;
A heterogeneous metal-free photocatalytic system was developed for oxidative cleavage of C–C bonds in 1,2-diols at mild conditions
visible light, room temperature, no base, O balloon). The reaction mechanism was studied through a series of control experiments
(
2
and DFT calculations. In addition, the catalytic system has a broad substrate scope, including aliphatic, benzylic, alkenyl 1,2-diols
and α-hydroxy acids, such as lactic acid.
9
This article is protected by copyright. All rights reserved.