Metal- and Additive-Free Oxidation of Sulfides into Sulfoxides by Fullerene-Modified Carbon Nitride with Visible-Light Illumination

Article abstract of DOI:10.1002/cssc.201800450

Photocatalytic selective oxidation has attracted considerable attention as an environmentally friendly strategy for organic transformations. Some methods have been reported for the photocatalytic oxidation of sulfides into sulfoxides in recent years. However, the practical application of these processes is undermined by several challenges, such as low selectivity, sluggish reaction rates, the requirement of UV-light irradiation, the use of additives, and the instability of the photocatalyst. Herein, a metal-free C₆₀/graphitic carbon nitride (g-C₃N₄) composite photocatalyst was fabricated by a facile method, and well characterized by TEM, SEM, FTIR spectroscopy, XRD, X-ray photoelectron spectroscopy, diffuse reflectance spectroscopy, and photoluminescence spectroscopy. The C₆₀/g-C₃N₄ catalyst exhibited a high photocatalytic activity at room temperature for the selective oxidation of sulfides into the corresponding sulfoxides in the presence of other functional groups, due to the synergetic roles of C₆₀ and g-C₃N₄. Several important parameters have been screened, and this method afforded good to excellent yields of sulfoxides under optimal conditions. The superoxide radical (^.O₂^?) and singlet oxygen (¹O₂) were identified as the oxidative species for the oxidation of sulfides into sulfoxides by exploring EPR experiments, and hence, a plausible mechanism for this oxidation was proposed. Moreover, the C₆₀/g-C₃N₄ catalyst can be easily recovered by filtration and then reused at least four times without loss in activity.

Full text of DOI:10.1002/cssc.201800450

DOI: 10.1002/cssc.201800450
Full Papers
Metal- and Additive-Free Oxidation of Sulfides into
Sulfoxides by Fullerene-Modified Carbon Nitride with
Visible-Light Illumination
Xi Chen, Kejian Deng,* Peng Zhou, and Zehui Zhang*[a]
Photocatalytic selective oxidation has attracted considerable
attention as an environmentally friendly strategy for organic
transformations. Some methods have been reported for the
photocatalytic oxidation of sulfides into sulfoxides in recent
years. However, the practical application of these processes is
undermined by several challenges, such as low selectivity, slug-
gish reaction rates, the requirement of UV-light irradiation, the
use of additives, and the instability of the photocatalyst.
Herein, a metal-free C60/graphitic carbon nitride (g-C3N4) com-
posite photocatalyst was fabricated by a facile method, and
well characterized by TEM, SEM, FTIR spectroscopy, XRD, X-ray
photoelectron spectroscopy, diffuse reflectance spectroscopy,
and photoluminescence spectroscopy. The C60/g-C3N4 catalyst
exhibited a high photocatalytic activity at room temperature
for the selective oxidation of sulfides into the corresponding
sulfoxides in the presence of other functional groups, due to
the synergetic roles of C60 and g-C3N4. Several important pa-
rameters have been screened, and this method afforded good
to excellent yields of sulfoxides under optimal conditions. The
ꢀ
1
C
superoxide radical ( O2 ) and singlet oxygen ( O2) were identi-
fied as the oxidative species for the oxidation of sulfides into
sulfoxides by exploring EPR experiments, and hence, a plausi-
ble mechanism for this oxidation was proposed. Moreover, the
C60/g-C3N4 catalyst can be easily recovered by filtration and
then reused at least four times without loss in activity.
Introduction
The oxidation reaction represents one of the most vital proto-
cols for upgrading raw starting materials into high-value-
added products in organic transformations.[1–4] Among various
types of oxidation reactions, the oxidation of sulfides into sulf-
oxides (sulfoxidation) is of great importance in organic synthe-
sis,[5] because sulfoxides are important intermediates or build-
ing blocks in the construction of pharmaceuticals, agrochemi-
cals, and other valuable fine chemicals. For example, optically
active sulfoxides are useful intermediates in medicinal and
pharmaceutical chemistry to prepare therapeutic agents.[6]
Therefore, interest in the selective oxidation of sulfides into
sulfoxides has attracted great interest for a long time. Tradi-
tionally, the oxidation of sulfides into sulfoxides has mainly
been performed through thermal oxidation processes by the
use of several kinds of oxidants, such as trifluoroperacetic acid,
hydrogen peroxide, tert-butyl hydroperoxide, and iodobenzene
diacetate.[7–11] However, these methods suffer from several
drawbacks, such as low selectivity, the high input of energy,
and the release of highly toxic waste. Therefore, the search for
new routes for the efficient and environmentally friendly oxida-
tion of sulfides into sulfoxides has continued to be of interest
in chemical research. Photocatalytic transformations, which uti-
lize sunlight as an abundant and readily available source of
energy for driving chemical reactions, have been considered to
be an important pathway towards a sustainable future.[12,13]
Recently, the photocatalytic oxidation of sulfides into sulfox-
ides with high selectivity has received increasing attention be-
cause it allows oxidation reactions to occur under mild condi-
tions through the use of solar-light irradiation. For instance,
Zhang and co-workers reported the photocatalytic oxidation of
the sulfides into sulfoxides over carbonyl ruthenium(II) porphy-
rin complexes by the use of iodobenzene diacetate (PhI(OAc)2)
as the oxygen source.[14] An inorganic semiconductor material
as a photosensitizer and a ruthenium–aqua complex as a cata-
lyst were also reported to be robust for sulfide oxidation.[15]
Additionally, an iron(IV)–oxo species has been used to oxidize
radical sulfide cations into the corresponding sulfoxides
through in situ generated [RuIII(bpy)3]3+ (bpy=2,2’-bipyri-
dine).[16] Nevertheless, these photocatalytic systems require a
large excess of sacrificial electron acceptors, such as
[Co(NH3)5Cl]Cl2 and Na2S2O8. From the viewpoint of green and
sustainable chemistry, O2 is the most attractive oxidant be-
cause it is an environmentally friendly and readily available oxi-
dant, with harmless water as the only byproduct generated in
the reaction.[17] However, the triplet ground state of oxygen is
inactive for photocatalytic oxidation, and activation of the trip-
let ground state of oxygen to the oxidative species for the se-
lective oxidation of sulfides to the corresponding sulfoxides is
a challenge. There are some reports on the use of molecular
[a] X. Chen, Prof. K. Deng, P. Zhou, Prof. Z. Zhang
Key Laboratory of Catalysis and Materials Sciences of the
Ministry of Education, South-Central University for Nationalities
Wuhan, 430074 (PR China)
E-mail: dengkj@mail.scuec.edu.cn
zehuizh@mail.ustc.edu.cn
Supporting Information and the ORCID identification number(s) for the
author(s) of this article can be found under https://doi.org/10.1002/
cssc.201800450.
ChemSusChem 2018, 11, 1 – 10
1
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
oxygen for the photocatalytic oxidation of sulfides into sulfox-
ides. Chen and co-workers reported three photocatalytic sys-
tems that achieved the aerobic oxidation of sulfides into sulf-
oxides by employing TiO2 as the catalyst.[18–20] The use of TiO2
alone cannot promote this reaction, which requires an addi-
tional redox mediator, such as (2,2,6,6-tetramethylpiperidin-1-
yl)oxyl (TEMPO) and anchored organic dyes or triethylamine on
the surface of TiO2 to promote the oxidation of sulfides into
sulfoxides. These processes were performed with methanol as
the solvent, and methanol was also oxidized into formalde-
hyde. Li and co-workers performed the photocatalytic oxida-
tion of sulfides on Pt/BiVO4 in water under visible-light illumi-
nation, but the conversion efficiency was poor.[21]
From the standpoint of green and sustainable chemistry, the
development of metal-free catalytic systems for the oxidation
of sulfides is needed because of inevitable metal residue con-
tamination of the products through the use of a homogeneous
complex or the loss of the metal component from heterogene-
ous catalysts.[22,23] There is one reported case of the metal-free
and photocatalytic oxidation of sulfides into sulfoxides by the
use of graphitic carbon nitride (g-C3N4) with oxygen as the oxi-
dant.[24] However, two equivalents of isobutyraldehyde was re-
quired, and the real oxidant was proposed to be a peroxide
radical and peracid, which were produced from the oxidation
of isobutyraldehyde. Additionally, NH4HF2 was added for the
preparation of mesoporous graphitic carbon nitride (mpg-
C3N4). Clearly, the development of new photocatalytic systems
for the environmentally friendly and metal-free oxidation of
sulfides into sulfoxides with molecular oxygen without any ad-
ditive is required.
Figure 1. XRD patterns of C60, g-C3N4, and 4 wt% C60/g-C3N4 composite.
clearly observed, and there is also a weak XRD peak at 2q=
13.08. The two peaks are characteristic XRD peaks of graphite-
like g-C3N4, corresponding to the (100) and (002) planes, and
suggest that g-C3N4 was successfully prepared. For C60, it exhib-
its four diffraction peaks at 2q=10.7, 17.7, 20.7, and 21.78, cor-
responding to the (111), (220), (311), and (222) planes, respec-
tively, and can be indexed to the cubic phase of C60 (JCPDS no.
44-0558).[25] For the C60/g-C3N4 composite, the intensity and po-
sition of the characteristic peak at 2q=27.28 barely changes
relative to those of bare g-C3N4, which suggests that the depo-
sition of C60 on the surface of g-C3N4 does not influence the
lattice structure of g-C3N4. More importantly, characteristic
peaks for C60 were also clearly observed in the XRD pattern of
the 4 wt% C60/g-C3N4 composite, which suggested that C60 had
been successfully immobilized on the surface of g-C3N4.
Fullerenes (C60) represent another allotrope of carbon with
unique electronic properties.[25] C60 is reported to be favorable
for efficient electron-transfer reduction because it has a closed-
shell configuration, which consists of 30 bonding molecular or-
bitals with 60 p electrons.[25] The unique structure of C60 serves
as an excellent electron acceptor, which efficiently causes rapid
photoinduced charge separation and relatively slow charge re-
combination. Wang and co-workers have summarized recent
significant advances achieved in the field of oxidation and hy-
drogenation as realized by g-C3N4-based catalytic systems;[26,27]
they also found the combination of carbon nitride and carbon
nanotubes showed synergistic effects in optoelectronic conver-
sion.[28] Thus, g-C3N4 has emerged as a class of 2D nanomaterial
analogous to graphite, and its unique structure renders exten-
sive potential use as a metal-free semiconductor for photocata-
lytic reactions.[25,26] Therefore, herein, we anticipated to take
advantages of the properties of fullerene and g-C3N4 to design
a hybrid metal-free catalyst for the photocatalytic oxidation of
sulfide under visible-light irradiation.
Representative SEM and TEM images of g-C3N4 and 4 wt%
C60/g-C3N4 composite are shown in Figure 2. As observed from
Figure 2a, C3N4 shows clear mesopores in lamellar structures.
In Figure 2b, there are some dark spots with lower transmis-
Results and Discussion
Catalyst characterization
The XRD patterns of C60, C3N4, and 4 wt% C60/g-C3N4 composite
are shown in Figure 1. From the XRD pattern of g-C3N4, one
characteristic XRD peak, with a strong intensity at 2q=27.28, is
Figure 2. TEM images of a) g-C3N4 and b) 4 wt% C60/g-C3N4 composite. SEM
images of c) g-C3N4 and d) 4 wt% C60/g-C3N4 composite (scale: 1 mm).
&
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
2
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
sion that indicate perturbation of C60 nanoparticles, without
changing the porous structure of C3N4, because C60 nanoparti-
cles are so small and easy to wrap in g-C3N4 nanosheets.
In Figure 2c, the SEM image of C60/g-C3N4 composite also
displays a 2D lamellar structure, which is in good agreement
with the TEM observation. Additionally, the SEM image of the
C60/g-C3N4 composite shows more mesopores than those in
the SEM image (Figure 2d); this may be due to ultrasonic treat-
ment.[29]
X-ray photoelectron spectroscopy (XPS) measurements were
performed to determine the chemical composition and valence
state of various species. The peak positions in all XPS spectra
are calibrated with C1s at 284.6 eV. Figure 3a displays the XPS
survey spectra of g-C3N4 and the 4 wt% C60/g-C3N4 composite.
There are only carbon and nitrogen elements in the two
samples, which suggest that there are no impurities in the two
samples. The C1s XPS spectrum can be fitted into two peaks
with different binding energies at 284.6 and 287.9 eV (Fig-
ure 3b). The peak with the binding energy at 287.9 eV is attrib-
uted to sp2-bonded carbon (NꢀC=N),[30] whereas the other
weak peak located at 284.6 eV can be assigned to the CꢀC
bond. By comparing the C1s XPS spectrum of g-C3N4 with that
of 4 wt% C60/g-C3N4, the binding energy for the sp2-bonded
carbon (NꢀC=N) peak shifted from 287.9 eV for g-C3N4 to
287.7 eV for the 4 wt% C60/g-C3N4 catalyst, which suggested
that there was an interaction between C60 and g-C3N4.
As far as the high-resolution N1s XPS spectrum of g-C3N4 is
concerned, the N1s peak can be deconvoluted into three
peaks at 398.1, 398.8, and 400.2 eV, which correspond to
C=NꢀC, Nꢀ(C)3, and NꢀH, respectively (Figure 3c); the very
weak peak at 404.2 eV corresponds to p excitation.[30,31] Similar-
ly, the peaks also shifted to a lower binding energy for the
4 wt% C60/g-C3N4 catalyst. As a hybrid photocatalyst, the
change in the light absorption of the UV/Vis region is a direct
way to check the efficiency of the hybrid photocatalyst. There-
fore, the samples of g-C3N4 and 4 wt% C60/g-C3N4 composite
were further characterized by UV/Vis diffuse reflectance ab-
sorption spectroscopy. As shown in Figure 4, both g-C3N4 and
the C60/g-C3N4 composite exhibit a sharp and wide absorption
edge at lꢁ435 nm, which are assigned to the intrinsic band
gap absorption of g-C3N4.[32] According to the equation Eg =
1240/lg (in which Eg is the band gap energy of the semicon-
ductor and lg is the optical absorption edge of the semicon-
ductor), the band gap of g-C3N4 is estimated to be 2.85 eV. As
shown in Figure 4, the absorption bands of the C60/g-C3N4
composites in the visible region become stronger with an in-
crease in the loading of C60 from 1 to 5 wt%, which suggests
that C60 enhances the light absorption of g-C3N4; this is benefi-
cial for photocatalytic reactions. The difference in absorption is
significant for a larger difference in the amount of C60, for ex-
ample, by comparing the spectra for 1 and 5 wt%, but the in-
distinctive overall change may be due to the small amount of
C60.
Figure 3. XPS spectra of g-C3N4 and the 4 wt% C60/g-C3N4 composite: a) XPS
survey spectrum, b) high-resolution C1s spectrum, and c) high-resolution
N1s spectrum.
1574 cmꢀ1 are attributed to aromatic CꢀN stretching vibration
modes.[33] In addition, a sharp band was also observed at n˜ =
812 cmꢀ1, which was attributed to the s-triazine ring system.[34]
The stretching modes of terminal NꢀH groups at the defect
sites of the aromatic ring exhibit a shoulder band near n˜ =
3170 cmꢀ1.[32] For C60, the FTIR spectrum is weak, but some
characteristic bands at n˜ =578, 1172, and 1427 cmꢀ1 are visible;
these are attributed to the internal modes of the C60 mole-
cule.[35] After the deposition of C60, there is no apparent struc-
tural difference between g-C3N4 and 4 wt% C60/g-C3N4. Howev-
er, the characteristic bands for g-C3N4 from n˜ =1200 to
The samples were further characterized by FTIR spectrosco-
py (Figure 5). The absorption band at n˜ =1627 cmꢀ1 should be
ascribed to the stretching vibration modes of the C=N bond,
whereas the four strong bands at n˜ =1253, 1328, 1419, and
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
3
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 1. Photocatalytic oxidation of 1a under various conditions.[a]
Entry
Catalyst
Conversion
[%]
Selectivity [%]
1b
1c
1
2
3
4
1% C60/g-C3N4
2% C60/g-C3N4
3% C60/g-C3N4
4% C60/g-C3N4
5% C60/g-C3N4
–
27.6
46.0
50.9
63.9
51.5
–
20.2
3.4
–
100
100
100
100
100
–
98.1
64.3
–
0
0
0
0
0
5
6[b]
7
–
g-C3N4
1.9
35.7
–
–
0
8
C60 (1.5 mg)
4% C60/g-C3N4
4% C60/g-C3N4
4% C60/g-C3N4
4% C60/g-C3N4
9[c]
10[d]
11[e]
12[f]
13[g]
Figure 4. UV/Vis diffuse reflectance absorption spectra of g-C3N4 and C60/g-
C3N4 with different compositions.
–
–
34.41
94.35
22.1
100
100
98.0
0
2.0
C
60 +g-C3N4
[a] Reaction conditions: 1a (0.2 mmol), catalyst (30 mg), CH3OH (5 mL),
oxygen balloon, room temperature, Xe lamp (1 Wcmꢀ2, l>400 nm) irra-
diation for 2 h. [b] Without catalyst. [c] In the dark. [d] Filled with N2. [e] In
air, with irradiation for 2 h. [f] In air, with irradiation for 10 h. [g] C60 and
C3N4 were added in amounts corresponding to those of entry 9.
(Table 1, entry 5). These results suggested that C60 played a cru-
cial role in the photocatalytic activity of the C60/g-C3N4 cata-
lysts. With increasing C60 content, a larger number of active
catalytic sites are provided, but, once a limiting value is
reached, the adsorption of substrate molecules and light will
be restrained.
Figure 5. FTIR spectra of g-C3N4, C60, and the 4 wt% C60/g-C3N4 composite.
Furthermore, the reactions were conducted in different sol-
vents to study its influence on the photocatalytic oxidation of
1a (Table 2). Interestingly, the solvent had a great effect on the
photocatalytic activity of the 4 wt% C60/g-C3N4 catalyst towards
the oxidation of 1a. No reaction took place in THF (Table 2,
entry 1), and the reactions in ethanol and benzotrifluoride pro-
duced low conversions of 22.5 and 30.4% (Table 2, entries 2
and 3), respectively. Acetonitrile is considered to be the best
solvent for photocatalytic reactions in some cases,[36,37] and it
also produced a high conversion of 83.9% (Table 2, entry 4);
however, the selectivity of methyl phenyl sulfoxide (1b) was
poor. It has been reported that Pt/BiVO4 systems exhibit excel-
1700 cmꢀ1 are shifted, which suggests that there is an interac-
tion between g-C3N4 and C60, which may benefit electron trans-
fer and enhance the photocatalytic activity of composite mate-
rials.
It has been reported that the delocalized 60 p-electron
structure of C60 facilitates the separation of photoinduced elec-
trons because it can serve as an excellent electron acceptor.[35]
C60 has a 2D p-conjugated structure, which is a good electron-
accepting material that effectively hinders electron–hole re-
combination.
Photocatalytic oxidation of sulfides
Table 2. Photocatalytic oxidation of sulfides in different solvents.[a]
The photocatalytic activity of the C60/g-C3N4 catalysts was eval-
uated by the oxidation of methyl phenyl sulfide (1a) as the
model reaction. The photocatalytic reactions were performed
under continuous illumination with visible light from a 300 W
xenon lamp equipped with a l=400 nm cutoff filter under an
oxygen balloon. First, the catalytic activity of the series C60/g-
C3N4 catalysts was investigated. It was observed that the con-
version of 1a increased with increasing C60 weight percentage
from 1 to 4 wt% (Table 1, entries 1–4). However, the photoca-
talytic performance of the C60/g-C3N4 catalyst decreases upon
further increasing the weight percentage of C60 to 5 wt%
Entry
Solvent
t
[h]
Conversion
[%]
Selectivity [%]
1b
1c
1
2
3
4
5
6
THF
6
6
6
6
6
6
<1
–
–
CH3CH2OH
benzotrifluoride
CH3CN
H2O
CH3OH
22.5
30.4
83.9
99.5
99.9
36.1
47.8
68.3
55.7
63.9
53.2
31.7
44.3
0
100
[a] Reaction conditions: 1a (0.2 mmol), 4 wt% C60/g-C3N4 (30 mg), solvent
(5 mL), oxygen balloon, room temperature, Xe lamp (1 Wcmꢀ2
400 nm).
, l>
&
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
4
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
lent performance for the oxidation of sulfides in water, com-
pared with in organic solvents.[21]
The time course of the product distribution was recorded
for the photocatalytic oxidation of 1a in methanol over the
4 wt% C60/g-C3N4 catalyst. During the reaction process, the
concentration of 1a gradually decreased, whereas the concen-
tration of 1b gradually increased (Figure 6). No intermediates
or other byproducts were observed during the reaction pro-
cess. After light irradiation for 6 h, 1a was quantitatively con-
verted into the target product of 1b.
In our catalytic system, a quantitative conversion of 1a was
observed in water after 6 h, but the reaction in water pro-
duced an almost equal mixture of 1b and methyl phenyl sul-
fone (Table 2, entry 5). Pleasingly, methanol was found to be
the best solvent because 1a was quantitatively transformed
into 1b after 6 h in methanol under visible-light irradiation
(Table 2, entry 6). Therefore, methanol was the best solvent for
the photocatalytic oxidation of 1a, and could stabilize the oxi-
dation product of 1b to prevent further oxidation. Compound
1b in water could be further oxidized into methyl phenyl sul-
fone; this was possibly because 1b could easily dissolve in
water, as a result of its polarity, to promote further oxidation.
In addition, the huge difference between ethanol and metha-
nol in the activity of the C60/g-C3N4 catalyst was possibly be-
cause the hydrophilic C60/g-C3N4 catalyst could be much better
dispersed in methanol. According to the experimental results,
the 4 wt% C60/g-C3N4 catalyst was the best choice for the pho-
tocatalytic oxidation of sulfides.
To obtain more information on the photocatalytic oxidation
of 1a, additional experiments were also performed. For com-
parison, a control experiment was also conducted in the ab-
sence of catalyst under the same reaction conditions (Table 1,
entry 6); the oxidation of 1a did not occur, which suggested
that the reaction was promoted by the catalyst. In addition, a
low conversion of 20.2% was observed in the presence of g-
C3N4 with a high selectivity of 98.1% to 1b (Table 1, entry 7).
However, the single use of C60 produced a very low conversion
of 1a (Table 1, entry 8). The conversion was not increased if C60
was added with g-C3N4 (Table 1, entry 13 vs. entry 8), which in-
dicated that the merely mixing with C60 had little effect on the
catalytic efficiency of g-C3N4. These results suggested that C60
and g-C3N4 had a synergetic effect on the photocatalytic oxida-
tion of 1a. The dispersion of C60 on the surface of g-C3N4, to
give rise the hybrid C60/g-C3N4 catalyst, might enhance the ab-
sorption of visible light (Figure 4) and promote the electron-
transfer ability (Figure 9, below), resulting in enhanced photo-
catalytic activity. In addition, no reaction took place in the dark
or under a nitrogen atmosphere (Table 1, entries 9 and 10).
These results show that the C60/g-C3N4photocatalyst, light ir-
radiation, and oxygen are all crucial factors in the photocatalyt-
ic oxidation of sulfides into sulfoxides. Although there has
been an example of photocatalytic oxidation of sulfide into
sulfoxide with g-C3N4, this catalytic system required the use of
isobutyraldehyde as an additive.[24] To the best of our knowl-
edge, this is the first example of the successful oxidation of
sulfide to sulfoxide with a metal-free heterogeneous catalyst
without any additive under mild conditions (258C, 1 atm
(=101325 Pa) O2). Clearly, it is more sustainable to perform
chemical reactions without additives. Photocatalytic oxidation
in air resulted in a relatively low conversion efficiency, com-
pared with that if the system was filled with oxygen (Table 1,
entry 4 vs. entry 11); this was due to the low concentration of
oxygen in the reaction solution. By prolonging the reaction
time to 10 h, compound 1b was attained in a high yield of
94.4% in air (Table 1, entry 12).
Figure 6. Time course of the products distribution for the photocatalytic oxi-
dation of 1a (MPS) into 1b (MPSO).
Substrate scope
The substrate scope of the developed method was extended
to a series of structurally diverse sulfides, and the reactions
were performed in methanol over the 4 wt% C60/g-C3N4 cata-
lyst. As shown in Table 3, this method with the 4 wt% C60/g-
C3N4 catalyst exhibits a good tolerance for a wide range of sul-
fides. The sulfides were successfully oxidized into the corre-
sponding sulfoxides with good to excellent yields. For all
cases, the selectivity of the sulfoxides reached nearly 100%.
However, the activity of the substrates was greatly influenced
by their structures. The electronic properties of the substitu-
ents in the aryl sulfides played an important role in the sub-
strate activity (Table 3, entries 1–3 vs. 4–6).
Generally, the substrates with electron-donating groups
demonstrate higher activity than those substrates with elec-
tron-withdrawing groups; the latter require longer reaction
times to give high yields of the corresponding sulfoxides
(Table 3, entries 1–3 vs. 4–6). The substrates with electron-do-
nating groups more easily donate electrons to sulfur atoms
than the substrates with electron-withdrawing groups, which
would be beneficial for the oxidation reactions. Additionally,
the steric hindrance of the groups also showed a great influ-
ence on the substrate activity (Table 3, entries 1 vs. 7, 5 vs. 6,
and 8 vs. 9). The large steric hindrance of the substrates
caused them difficulties in accessing the active sites. For exam-
ple, diphenyl sulfide, with two aromatic rings connected by a
sulfur atom, has a high electron density on the sulfur atom,
but it demonstrated the lowest activity of all tested substrates
(Table 3, entries 8 and 9), owing to the greatest steric hin-
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
5
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 3. Substrate scope of the oxidation of sulfides into sulfoxides.[a]
Entry
Sulfide
t [h]
Conversion
[%]
Selectivity
[%]
1
2
3
4
5
6
6
100
100
100
88.2
90.6
100
100
100
99.6
97.7
6
12
12
6
16
85.7
96.6
7
8
6
81.4
70.8
96.7
98.2
Figure 7. Recycling experiments of the 4 wt% C60/g-C3N4 catalyst. Reaction
conditions: 1a (0.2 mmol), 4 wt% C60/g-C3N4 (30 mg), CH3OH (5 mL), 258C,
oxygen balloon, visible-light irradiation (>400 nm).
20
9
12
99.8
99.6
[38]
10
11
6
6
82.9
85.3
99.7
94.9
C
in methanol, methanol had the ability to quench OH. There-
C
fore, OH should not be the oxidative species for the photoca-
12
6
86.8
98.5
talytic oxidation of sulfides. Furthermore, extra experiments
were performed with the use of scavengers to detect oxidative
[a] Reaction conditions: substrates (0.2 mmol), 4 wt% C60/g-C3N4 (30 mg),
CH3OH (5 mL), 258C, 1 atm O2, visible-light irradiation (>400 nm).
species. p-Benzoquinone (BQ), beta-carotene, and KI were
ꢀ
C
added as scavengers to capture superoxide radicals ( O2 ), sin-
glet oxygen (1O2), and photoinduced holes (h+) during the
photocatalytic reaction process, respectively (Table 4).[39,40] High
conversion of 1a and high selectivity of 1b were attained
without any additive (Table 4, entry 1). The reaction stopped if
KI was added (Table 4, entry 2), which indicated that photoin-
duced electron–hole separation was crucial for activation.[41]
drance. Based on the above analysis, it can be concluded that
the presence of electron-donating groups accelerates the con-
version of sulfide into sulfoxide, whereas a large steric hin-
drance effect is unfavorable for the photocatalytic oxidation of
sulfide. More importantly, this method was also effective for
aliphatic sulfides and a substrate with a heteroatom ring
(Table 3, entries 10–12).
Table 4. The effect of scavengers on the oxidation of sulfides.[a]
Entry Quencher
Quenching group Conversion [%] Selectivity [%]
Catalyst recycling experiments
1
2
3
4
–
KI
–
99.5
<1
26.7
99
–
99
99
h+
One of the most important merits of heterogeneous catalysts
is that they can be recycled and reused. Thus, the recyclability
of the 4 wt% C60/g-C3N4 catalyst was investigated for the oxi-
dation of 1a. After the first catalytic run, the catalyst was sepa-
rated by centrifugation at 8000 rpm, then leached and washed
with methanol three times, and dried at 508C under vacuum.
The spent 4 wt% C60/g-C3N4 catalyst was subjected to a
second run under the same conditions. As shown in Figure 7,
the 4 wt% C60/g-C3N4 catalyst demonstrated excellent recycla-
bility. After six consecutive runs, the conversion of 1a and the
selectivity of 1b were maintained without significant decreas-
es.
beta-carotene 1O2ꢀ
C
BQ
O2
33.68
[a] Reaction conditions: 1a (0.2 mmol), 4 wt% C60/g-C3N4 (30 mg), metha-
nol (5 mL), oxygen balloon, visible-light irradiation (l>400 nm) for 6 h.
The conversion of 1a dropped from 100 to 26.7% in the
presence of beta-carotene without loss in the selectivity of 1b
(Table 4, entry 3), which suggested that O2 was present in the
1
reaction system and served as an oxidative species for the
photocatalytic oxidation of sulfides.[5,42] Similarly, the conver-
sion also decreased to 33.7% with BQ as a scavengers to cap-
ꢀ
C
ture O2 (Table 4, entry 4). The results in Table 4 clearly indicat-
Proposed mechanism for the oxidation of sulfides
ꢀ
1
C
ed that O2 and O2 were the oxidative species for the photo-
The conversion of sulfides into sulfoxides was greatly en-
hanced with the use of C60/g-C3N4. Generally speaking, super-
catalytic oxidation of sulfides. The EPR spin-trapping technique
1
ꢀ
C
was then used to directly to probe O2 and O2
.
ꢀ
1
C
oxide radicals ( O2 ), singlet oxygen ( O2), and hydroxyl radicals
TEMPO and dimethyl pyridine N-oxide (DMPO) were em-
ployed to probe the reactive oxygen species generated during
visible-light irradiation. After irradiation, the characteristic sig-
C
( OH) were reported to be the active species for photocatalytic
reactions. Because the photocatalytic reaction was performed
&
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
6
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
ꢀ
nals of the DMPO– O2 and TEMPO–1O2 adducts were clearly
observed (Figure 8). For comparison, these signals were not
observed in the dark, which further confirmed that these oxi-
dative species were generated by light irradiation in the pres-
ence of the C60/g-C3N4 catalyst. To confirm enhanced conver-
sion with a combination of C60 and C3N4, the same experiment
was conducted with only g-C3N4. As shown in Figures S1 and
S2 in the Supporting Information, the only active species of O2
C
in the reaction is O2ꢀ; the improved conversion of sulfide
C
1
mainly results from oxidative O2.
Scheme 1. Possible mechanism for the oxidation of sulfides into sulfoxides
in methanol. ISC=intersystem crossing.
Figure 9. Room-temperature PL excitation and emission spectra of g-C3N4
and C60/g-C3N4 photocatalysts (lex =370 nm).
composite could lead to the separation of photogenerated
electron–hole pairs.
On one hand, one oxidative species, O2ꢀ, was generated
C
through the activation of molecular oxygen (O2) by receiving
one electron from C60; on the other hand, the other oxidative
Figure 8. Changes in the ESR spectra for systems containing DMPO or
1
TEMPO and 4 wt% C60/g-C3N4 in methanol under visible-light irradiation for
species, O2, could be produced through triplet energy transfer
-
1
C
3
from photoexcited C60* to ground-state oxygen (3O2).[10,43,44]
12 min (l>400 nm). a) DMPO– O2 without substrate and b) TEMPO– O2 with-
out substrate.
1
ꢀ
C
The O2 and O2 oxidative species then attack the sulfur atom
in the sulfides to produce the peroxysulfoxide reactive inter-
mediate with dipolar (A1) and diradical (A2) structures
(Scheme 1), which were both present in the reaction system.[10]
During this process, one electron was simultaneously released
and then transferred to the photoinduced holes (h+). Methanol
could stabilize the persulfoxides through hydrogen-bonding
interactions to generate intermediate B.[44] Finally, the addition
of another sulfide molecule to intermediate B afforded two
molecules of the sulfoxide product.
On the basis of the above results, a plausible mechanism
was proposed for the photocatalytic oxidation of sulfides. As
shown in Scheme 1, photoinduced holes (h+) in the valence
band and electrons in the conduction band (eꢀ) were generat-
ed with g-C3N4 under light irradiation. The generated electrons
then transferred from g-C3N4 to C60 due to the p structure and
excellent electron receptivity of C60; this caused the separation
of h+ and eꢀ on the surface of C3N4 more efficiently.[26] As
shown in Figure 9, the intensity of the photoluminescence (PL)
signal for the 4 wt% C60/g-C3N4 composite is much lower and
exhibits a clear blueshift from l=444 to 439 nm, in compari-
son with g-C3N4. This result indicates that the 4 wt% C60/g-C3N4
catalyst has a lower recombination rate of electrons and holes
under visible-light irradiation, and verifies that the C60/g-C3N4
Conclusions
We have developed a metal-free heterogeneous catalytic
system for the selective photo-oxidation of sulfides to sulfox-
ides. The 4 wt% C60/g-C3N4 catalyst demonstrated high catalyt-
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
7
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
the reaction, the content of each compound was identified based
on GC-MS analysis and quantified based on the internal standard
method with bromobenzene as the internal standard. The conver-
sion of sulfide and the selectivity of sulfoxide were calculated by
using Equations (1) and (2), respectively:
ic activity in the oxidation of sulfides to sulfoxides with O2
under visible-light illumination due to the synergetic effect be-
tween C60 and g-C3N4. The reaction solvent was found to be
crucial for the activity of the 4 wt% C60/g-C3N4 catalyst, as well
as product selectivity, and methanol proved to be the best sol-
vent. This catalytic method could smoothly promote the trans-
formation of various structurally diverse sulfides to the corre-
sponding sulfoxides. According to the ESR investigation and
Conversion ½%ꢂ ¼ ½ðC0ꢀCsulfideÞ=C0ꢂ ꢃ 100
Selectivity½ð%ꢂ ¼ ½Csulfoxide=ðC0ꢀCsulfideÞꢂ ꢃ 100
ð1Þ
ð2Þ
some control experiments, a possible mechanism was pro-
ꢀ
in which C0 was the initial concentration of sulfide, and Csulfoxide and
Csulfide were the concentrations of substrate sulfides and the corre-
sponding sulfoxides, respectively.
C
posed for the photocatalytic oxidation of sulfides with O2
1
and O2 as the oxidative species. More importantly, the 4 wt%
C60/g-C3N4 catalyst demonstrates excellent stability without the
loss of catalytic activity. This photocatalytic system offers the
potential to utilize sunlight for effective and economical pro-
duction of value-added chemicals through oxidative reactions.
Determining the active oxygen species
To determine the reactive active species involved in the reaction,
we performed extra experiments to trap the active oxygen species.
Sacrificial agents, BQ, beta-carotene, and KI, were used to capture
Experimental Section
superoxide radical ( O2ꢀ), singlet oxygen (1O2), and photoinduced
C
holes (h+), respectively. Moreover, the active oxygen species were
probed by means of EPR analysis. Samples containing catalysts
(5 mg) and DMPO (100 mm) or TEMPO(100 mm) were vacuumized
and oscillated to suspend the catalyst in methanol evenly, then the
mixtures were added to an EPR quartz tube filled with O2, and irra-
diated with a 300 W Xe lamp (CERAMAX LX-300) equipped a l=
400 nm cutoff filter. The settings of the ESR spectrometer for sin-
glet oxygen were as follows: center field=337.245 mT, microwave
frequency=9437.606 mGHz, sweep width=4 G, modulation fre-
quency=100 kHz, and power=0.998 mW; the setting of the ESR
spectrometer of superoxide radicals were as follows: center field=
322.767 mT, microwave frequency=9054.648 mGHz, sweep
width=5 G, modulation frequency=100 kHz, and power=
0.998 mW.
Materials
The chemicals used in this study were purchased from Aladdin
Chemicals Co. Ltd. (Beijing, P.R. China). The solvents used in this
study were supplied by Sinopharm Chemical Reagent Co., Ltd.
(Shanghai, P.R. China). All chemicals and solvents were used direct-
ly without any purification. Preparation of the catalyst was accord-
ing to the previous report.[26,45] In a typical process, urea (10 g) was
placed in a crucible and heated at 5008C for 3 h at a heating rate
of 58Cminꢀ1 in a Muffle furnace. The resulting yellow powder was
ground and collected. The C60/g-C3N4 composite was prepared as
follows: A mixture of C60 and as-prepared g-C3N4 with a total
weight of 500 mg was added to toluene (50 mL) under stirring and
ultrasonic treatment for 60 min. After toluene was removed under
vacuum, the powder was washed with ethanol at least three times,
and dried under vacuum overnight to obtain the gray C60-hybrid-
ized g-C3N4. A series of C60/g-C3N4 hybrids with different weight
ratios of C60 and g-C3N4 from 1 to 5 wt% were prepared.
Acknowledgements
We gratefully acknowledge financial support from the National
Natural Science Foundation of China (grant nos. 21772237 and
21272281).
Catalyst characterization
Powder XRD (Bruker D8 Advance; CuKa =1.5404 ꢁ) was used to an-
alyze the crystalline phase of the catalysts, operating at a scanning
rate of 0.058sꢀ1 over the 2q range of 10 to 508. The chemical state
of the product was measured by XPS (VGMultilab 2000 photoelec-
tron spectrometer) by using AlKa radiation as the excitation source
under 2ꢂ10ꢀ6 Pa vacuum. UV/Vis diffuse reflectance absorption
spectra were recorded on a Shimadzu UV-2600 spectrophotometer
equipped with an integrating sphere by using BaSO4 as the refer-
ence sample. FTIR spectra of the samples were recorded on a
NEXUS 470 spectrometer as conventional KBr pellets. PL spectra
were measured at room temperature on an F-7000 fluorescence
spectrophotometer with an excitation wavelength of 370 nm.
Conflict of interest
The authors declare no conflict of interest.
Keywords: fullerenes · heterogeneous catalysis · oxidation ·
photocatalysts · sulfur
[1] Y. C. Ling, G. M. Wang, J. Reddy, C. C. Wang, J. Z. Zhang, Y. Li, Angew.
Chem. Int. Ed. 2012, 51, 4074–4079; Angew. Chem. 2012, 124, 4150–
4155.
[2] N. Dimitratos, J. A. Lopez-Sanchez, G. J. Hutchings, Chem. Sci. 2012, 3,
20–44.
[3] M. Bordeaux, A. Galarneau, J. Drone, Angew. Chem. Int. Ed. 2012, 51,
10712–10723; Angew. Chem. 2012, 124, 10870–10881.
[4] S. R. Zhang, L. Nguyen, Y. Zhu, S. H. Zhan, C. K. Tsung, F. Tao, Acc. Chem.
Res. 2013, 46, 1731–1739.
[5] P. Kowalski, K. Mitka, K. Ossowska, Z. Kolarska, Tetrahedron 2005, 61,
1933–1953.
[6] S. Caron, R. W. Dugger, S. G. Ruggeri, J. A. Ragan, D. H. B. Ripin, Chem.
Rev. 2006, 106, 2943–2989.
General procedure for the photocatalytic oxidation of sul-
fides
In general, catalyst (30 mg) was suspended in methanol (5 mL)
containing substrate (0.2 mmol sulfides) in a homemade quartz re-
actor equipped with cold-water condensation. The suspension was
degassed and filled with dioxygen (1 atm) in a balloon. Then the
suspension was magnetically stirred and irradiated under visible
light (1 Wcmꢀ2 Xe lamp with a l=400 nm UV cutoff filter). After
&
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
8
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
[7] C. G. Venier, T. G. Squires, Y. Y. Chen, G. P. Hussmann, J. C. Shei, B. F.
Smith, J. Org. Chem. 1982, 47, 3773–3777.
[27] P. Zhang, J. Deng, J. Mao, H. Li, Y. Wang, Chin. J. Catal. 2015, 36, 1580–
1586.
[28] Y. Gong, J. Wang, Z. Wei, P. Zhang, H. Li, Y. Wang, ChemSusChem 2014,
7, 2303–2309.
[29] X. Bai, L. Wang, Y. Wang, W. Yao, Y. Zhu, Appl. Catal. B 2014, 152, 262–
270.
[30] D. J. Martin, K. Qiu, S. A. Shevlin, A. D. Handoko, X. Chen, Z. Guo, J.
Tang, Angew. Chem. Int. Ed. 2014, 53, 9240–9245; Angew. Chem. 2014,
126, 9394–9399.
[31] G. Bi, J. Wen, X. Li, W. Liu, J. Xie, Y. Fang, W. Zhang, RSC Adv. 2016, 6,
31497–31506.
[32] B. Chai, T. Peng, J. Mao, K. Li, L. Zan, Phys. Chem. Chem. Phys. 2012, 14,
16745–16748.
[33] F. Dong, L. Wu, Y. Sun, M. Fu, Z. Wu, S. C. Lee, J. Mater. Chem. 2011, 21,
15171–15174.
[34] Y. Wang, R. Shi, J. Lin, Y. Zhu, Energy Environ. Sci. 2011, 4, 2922–2928.
[35] J. Yu, T. Ma, G. Liu, B. Cheng, Dalton Trans. 2011, 40, 6635–6644.
[36] Z. Chai, T.-T. Zeng, Q. Li, L.-Q. Lu, W.-J. Xiao, D. Xu, J. Am. Chem. Soc.
2016, 138, 10128–10131.
[37] F. Raza, J. H. Park, H.-R. Lee, H.-I. Kim, S.-J. Jeon, J.-H. Kim, ACS Catal.
2016, 6, 2754–2759.
[38] Y. Xu, Z. C. Fu, S. Cao, Y. Chen, W. F. Fu, Catal. Sci. Technol. 2017, 7, 587–
595.
[39] W. Lu, T. Xu, Y. Wang, H. Hu, N. Li, X. Jiang, W. Chen, Appl. Catal. B 2016,
180, 20–28.
[40] C. S. Foote, R. W. Denny, J. Am. Chem. Soc. 1968, 90, 6233–6235.
[41] Q. Cai, Z. Hu, Q. Zhang, B. Li, Z. Shen, Appl. Surf. Sci. 2017, 403, 151–
158.
[42] J. Kyriakopoulos, A. T. Papastavrou, G. D. Panagiotou, M. D. Tzirakis, K. S.
Triantafyllidis, M. N. Alberti, K. Bourikas, C. Kordulis, M. Orfanopoulos, A.
Lycourghiotis, J. Mol. Catal. A 2014, 381, 9–15.
[43] M. R. Fraelich, R. B. Weisman, J. Phys. Chem. 1993, 97, 11145–11147.
[44] E. L. Clennan, S. E. Hightower, A. Greer, J. Am. Chem. Soc. 2005, 127,
11819–11826.
[8] S. Wolfe, C. F. Ingold, J. Am. Chem. Soc. 1983, 105, 7755–7757.
[9] F. Gasparrini, M. Giovannoli, L. Maresca, G. Natile, G. Palmieri, Synth.
Commun. 1984, 14, 1111 – 1117.
[10] F. Jensen, A. Greer, E. L. Clennan, J. Am. Chem. Soc. 1998, 120, 4439–
4449.
[11] L. Xu, J. Cheng, M. L. Trudell, J. Org. Chem. 2003, 68, 5388–5391.
[12] B. Zhou, J. Song, Z. Zhang, Z. Jiang, P. Zhang, B. Han, Green Chem.
2017, 19, 1075–1081.
[13] W. Huang, B. C. Ma, H. Lu, R. Li, L. Wang, K. Landfester, K. Zhang, ACS
Catal. 2017, 7, 5438–5442.
[14] T.-H. Chen, Z. Yuan, A. Carver, R. Zhang, Appl. Catal. A 2014, 478, 275–
282.
[15] X. Zhou, F. Li, X. Li, H. Li, Y. Wang, L. Sun, Dalton Trans. 2015, 44, 475–
479.
[16] A. Company, G. Sabenya, M. Gonzꢃlez-Bꢄjar, L. Gꢅmez, M. Clꢄmancey, G.
Blondin, A. J. Jasniewski, M. Puri, W. R. Browne, J. M. Latour, L. Que, Jr.,
M. Costas, J. Pꢄrez-Prieto, J. Lloret-Fillol, J. Am. Chem. Soc. 2014, 136,
4624–4633.
[17] Q. Wang, M. Zhang, C. C. Chen, W. H. Ma, J. C. Zhao, Angew. Chem. Int.
Ed. 2010, 49, 7976–7979; Angew. Chem. 2010, 122, 8148–8151.
[18] X. Lang, J. Zhao, X. Chen, Angew. Chem. Int. Ed. 2016, 55, 4697–4700;
Angew. Chem. 2016, 128, 4775–4778.
[19] X. Lang, W. Hao, W. R. Leow, S. Li, J. Zhao, X. Chen, Chem. Sci. 2015, 6,
5000–5005.
[20] X. Lang, W. R. Leow, J. Zhao, X. Chen, Chem. Sci. 2015, 6, 1075–1082.
[21] B. Zhang, J. Li, B. Zhang, R. Chong, R. Li, B. Yuan, S.-M. Lu, C. Li, J. Catal.
2015, 332, 95–100.
[22] I. Esteve-Adell, B. Crapart, A. Primo, P. Serp, H. Garcia, Green Chem.
2017, 19, 3061–3068.
[23] B. C. Patra, S. Khilari, R. N. Manna, S. Mondal, D. Pradhan, A. Pradhan, A.
Bhaumik, ACS Catal. 2017, 7, 6120–6127.
[24] P. Zhang, Y. Wang, H. Li, M. Antonietti, Green Chem. 2012, 14, 1904–
1905.
[45] B. Chai, X. Liao, F. Song, H. Zhou, Dalton Trans. 2014, 43, 982–989.
[25] X. Chen, H. Chen, J. Guan, J. Zhen, Z. Sun, P. Du, Y. Lu, S. Yang, Nano-
scale 2017, 9, 5615–5623.
Manuscript received: March 5, 2018
[26] Y. Gong, M. Li, H. Li, Y. Wang, Green Chem. 2015, 17, 715–736.
Version of record online: && &&, 0000
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
9
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
X. Chen, K. Deng,* P. Zhou, Z. Zhang*
Unrestricted catalysis: A composite
catalyst of C60 fullerene and graphitic
carbon nitride (g-C3N4) exhibits a high
photocatalytic activity for the selective
oxidation of sulfides into the corre-
sponding sulfoxides, in the presence of
other functional groups, under visible-
light illumination at room temperature.
&& – &&
Metal- and Additive-Free Oxidation of
Sulfides into Sulfoxides by Fullerene-
Modified Carbon Nitride with Visible-
Light Illumination
&
ChemSusChem 2018, 11, 1 – 10
www.chemsuschem.org
10
ꢀ 2018 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!