Highly efficient transformation of ethylbenzene into acetophenone catalyzed by NHPI/Co(II) using molecular oxygen in hexafluoropropan-2-ol

Article abstract of DOI:10.1016/j.mcat.2020.111244

Acetophenone is an important industrial intermediate and generally produced by the Friedel-Crafts acylation reaction, suffering from a low reactivity and serious equipment corrosion. Direct oxidation of ethylbenzene to acetophenone by molecular oxygen will be benign and cost-effective. The catalytic performance of NHPI/Co(II) herein was investigated by selective oxidation of ethylbenzene to acetophenone in different solvents at room temperature. The solvent hexafluoropropan-2-ol (HFIP) was found to markedly enhance the transformation efficiency from ethylbenzene to acetophenone in comparison with acetic acid, pyridine and ethanol, and the ethylbenzene conversion and the selectivity to acetophenone was high up to 87.9 % and 61.2 %, respectively. A higher concentration of phthalimide-N-oxyl (PINO) radicals was observed by an in situ electron paramagnetic resonance spectrometer (EPR) in HFIP with respect to other solvents, suggesting that HFIP may facilitate the generation of the N-oxyl radical and thus promote the selective oxidation of ethylbenzene to acetophenone. Furthermore, the benzylic carbon radical (PhCHC^[rad]H₃) from ethylbenzene was trapped by tetramethylpiperidine N-oxyl radical (TEMPO) and observed by a high resolution mass spectrometer. The findings of both PINO and PhCHC^[rad]H₃ under reaction conditions indicated that the selective oxidation of ethylbenzene to acetophenone catalyzed by NHPI/Co(II) should proceed via a radical mode. The selective oxidation of ethylbenzene to acetophenone using molecular oxygen by NHPI/Co(II) in HFIP exhibited an important industrial application prospect.

Full text of DOI:10.1016/j.mcat.2020.111244

Molecular Catalysis 498 (2020) 111244
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.journals.elsevier.com/molecular-catalysis
Highly efficient transformation of ethylbenzene into acetophenone
catalyzed by NHPI/Co(II) using molecular oxygen in hexafluoropropan-2-ol
,
Sihao Xua, Guojun Shia *, Ya Fenga, Chong Chenb, Lijun Jia
a School of Chemistry and Chemical Engineering, Yangzhou University, Yangzhou 225002, Jiangsu Province, People’s Republic of China
b Analysis Test Center, Yangzhou University, Yangzhou 225002, Jiangsu Province, People’s Republic of China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Acetophenone is an important industrial intermediate and generally produced by the Friedel-Crafts acylation
reaction, suffering from a low reactivity and serious equipment corrosion. Direct oxidation of ethylbenzene to
acetophenone by molecular oxygen will be benign and cost-effective. The catalytic performance of NHPI/Co(II)
herein was investigated by selective oxidation of ethylbenzene to acetophenone in different solvents at room
temperature. The solvent hexafluoropropan-2-ol (HFIP) was found to markedly enhance the transformation ef-
ficiency from ethylbenzene to acetophenone in comparison with acetic acid, pyridine and ethanol, and the
ethylbenzene conversion and the selectivity to acetophenone was high up to 87.9 % and 61.2 %, respectively. A
higher concentration of phthalimide-N-oxyl (PINO) radicals was observed by an in situ electron paramagnetic
resonance spectrometer (EPR) in HFIP with respect to other solvents, suggesting that HFIP may facilitate the
generation of the N-oxyl radical and thus promote the selective oxidation of ethylbenzene to acetophenone.
Furthermore, the benzylic carbon radical (PhCHCH3) from ethylbenzene was trapped by tetramethylpiperidine
N-oxyl radical (TEMPO) and observed by a high resolution mass spectrometer. The findings of both PINO and
PhCHCH3 under reaction conditions indicated that the selective oxidation of ethylbenzene to acetophenone
catalyzed by NHPI/Co(II) should proceed via a radical mode. The selective oxidation of ethylbenzene to ace-
tophenone using molecular oxygen by NHPI/Co(II) in HFIP exhibited an important industrial application
prospect.
Ethylbenzene
Molecular oxygen
Selective oxidation
NHPI
HFIP
Cobalt(II) acetate
Introduction
the air was used as the oxidant directly. And the reactions via vapor
oxidation can be carried out using a fixed-bed flowing reactor without
–
The activation and selective oxidation of CH bonds are of great
importance in industrial transformation of inexpensive hydrocarbons
into oxygenated compounds with a higher value. The selective oxidation
of ethylbenzene to acetophenone is a typical reaction for selective
fear of the separating and recovering of catalysts. However, the high
temperature reaction in vapor phase oxidation often leads to the pro-
duction a lot of products, such as styrene, benzaldehyde, benzoic acid,
CO and CO2 [8,9,10]. Compared to the vapor phase oxidation, the liquid
phase oxidation of ethylbenzene to acetophenone can be performed in
milder reaction conditions, and exhibited a higher ethylbenzene con-
version and a lower selectivity to overoxidized products like CO2,
showing a promising industrial application prospect [11,12]. Hetero-
geneously catalytic transformation of ethylbenzene into acetophenone
facilitates the separation of the catalyst, but the conversion of ethyl-
benzene and the selectivity to acetophenone are generally low [13,14].
The homogeneous catalysts can be molecularly dispersed in solvent,
permitting a good contact with ethylbenzene molecules and thus
exhibiting higher reactivity and selectivity [15].
–
oxidation reaction of benzyl CH bonds [1,2]. As an important chem-
ical intermediate, acetophenone is widely used in industrial production
fields, such as perfumes [3], soaps [4], resins [5] and drugs [6].
Industrially, acetophenone is synthesized by Friedel-Crafts acylation
reaction of benzene. However, the process suffers from a low reactivity,
serious equipment corrosion and a high cost of wastewater treatment
[7]. Therefore, a cost-efficient and benign technology for selective
oxidation of ethylbenzene to acetophenone will be crucial.
Different strategies were developed to convert ethylbenzene into
acetophenone via vapor phase or liquid phase oxidation in the past
years. There are no solvents needed in vapor phase oxidation, in which
Molecular oxygen is an environmentally friendly and cheap oxidant,
* Corresponding author.
E-mail address: gjshi@yzu.edu.cn (G. Shi).
https://doi.org/10.1016/j.mcat.2020.111244
Received 18 December 2019; Received in revised form 18 September 2020; Accepted 19 September 2020
Available online 18 October 2020
2468-8231/© 2020 Elsevier B.V. All rights reserved.

S. Xu et al.
Molecular Catalysis 498 (2020) 111244
and widely used in catalytic reactions in the oxidation of amines, alco-
hols and alkane [16,17,18]. However, it is necessary to activate the
Catalytic reactions and product analysis
–
triplet state of molecular oxygen and/or ground CH bond to realize the
Liquid phase oxidation of ethylbenzene was performed in a Schlenk
tube. Ethylbenzene (2 mmol), Co(OAc)24H2O (0.01 mmol), NHPI (0.05
mmol) and solvent (1 mL) were added to the Schlenk tube. Nitrobenzene
of 0.5 mmol was added as an internal standard after catalytic tests. The
rest of the tube was repeatedly purged by O2 (99.999 %) before mea-
surements, then the reaction was carried out at 303 K and a flowing
atmospheric O2 of 50 ml/min under a continuous stirring. The resulting
reaction mixture was analyzed by an off-line gas chromatograph (SHI-
MADZU GC-2014) with a SGE AC-10 capillary column and a flame
ionization detector.
oxygen functionalization of hydrocarbon due to the spin ꢀ flip restric-
tion between them [11]. Ishii’s group [19] developed a NHPI/Co(II)
catalytic system to oxidize toluene and other substrates into value-added
products in liquid phase under mild conditions for the first time, indi-
–
cating the aerobic oxidation of CH bonds under mild conditions was
possible by NHPI/Co(II) catalysts.
N-hydroxyphthalimide (NHPI) was employed to catalyze the acti-
–
vation and functionalization of CH bonds in many reaction systems.
Wang et al. [20] achieved a selective aerobic oxidation of cyclohexane to
ε
-caprolactone under mild conditions in the presence of NHPI and
aldehyde. Carboxylic functionalized β-carbolines were successfully
synthesized by aerobic oxidation in the presence of NHPI and transition
metal salts using molecular oxygen at room temperature [21]. Li et al.
Characterization
Electron paramagnetic resonance (EPR) tests were performed at
room temperature on an EPR spectrometer (A300ꢀ 10/12) with a field
modulation of 100 kHz. The microwave frequency was maintained at
9.848 GHz. The resulting reaction solution of ethylbenzene oxidation
was transferred from the Schlenk tube into a capillary quartz tube under
the reaction condition, then analyzed immediately.
–
[22] successfully achieved the chlorination of benzylic CH bond of
toluene by NHPI and 2,3-dichloro-5,6-dicyano-benzoquinone (DDQ).
NHPI is also used in liquid phase oxidation of ethylbenzene to ace-
tophenone [11,12,23,24]. Miao et al. [11] successfully achieved highly
selective oxidation of ethylbenzene to acetophenone (yield 70 %) in the
presence of Fe(NO3)3 and NHPI. However, the reaction was completed
in a longer period of time, probably due to the relatively low reactivity of
Ethylbenzene benzylic carbon radical (PhCHCH3) was determined by
a high resolution mass spectrometer (HRMS) (maXis, Bruker). TEMPO (2
mmol) was added to the reaction mixture after 30 min of reaction. Then,
the HRMS experiments are carried out immediately to detect the
captured intermediates.
–
the catalyst. Zhang et al. [12] realized controllable activation of CH
bond in the presence of -Fe2O3 and NHPI, the conversion of ethyl-
α
benzene and the selectivity to acetophenone in 4 h were 55 % and 96 %,
respectively. In addition, di-dodecyl-dimethyl ammonium bromide
(DDDAB) [23] and ion liquids [24] were found significant promoting
role on solvent-free oxidation of ethylbenzene to acetophenone cata-
lyzed by NHPI/Co(II). All of the above researches implied that NHPI is
active for oxidation of ethylbenzene to acetophenone in liquid phase,
but the efficiency of producing acetophenone currently is relatively low.
Therefore, an enhanced efficiency of oxidation of ethylbenzene to ace-
tophenone in liquid phase is raised here to meet increasing needs of
industrial application.
Results and discussion
Table 1 compares the effects of different solvents on the liquid phase
oxidation of ethylbenzene. There were no products from ethylbenzene
oxidation detected when ethanol was used as the solvent (entry 1), and
very low ethylbenzene conversion was observed in pyridine (entry 2).
Under the same reaction conditions, an ethylbenzene conversion of 21.2
%, a selectivity to acetophenone of 73.5 %, a selectivity to 1-phenyletha-
nol of 22.7 % and a selectivity to benzaldehyde of 2.4 % were observed
when acetic acid (HOAc) was used as the solvent (entry 3). It is
worthwhile to note that the conversion of ethylbenzene was signifi-
cantly increased to 87.8 % when HFIP was used as the solvent, and the
selectivities to acetophenone and 1-phenylethanol was 61.2 % and 34.7
%, respectively, and a small amount of benzaldehyde was observed
(entry 4). The results indicated that HFIP not only promotes the acti-
1,1,1,3,3,3-Hexafluoropropan-2-ol (HFIP) is
a non-nucleophilic
polar solvent with weak acidity, high dielectric constants and ioniza-
tion power [25]. In addition, HFIP is a strong hydrogen-bond donor that
pairs with hydrogen-bond acceptor groups, thereby can interfere with
the catalytic reactions cycle, promote the kinetics of polar reactions and
significantly increase the substrate conversion and selectivity to the
desired product [26,27,28]. Pappo et al. [29] developed a simple and
efficient method for selective oxidation of toluene to benzaldehyde
using HFIP as the solvent. Based on Ishii and his co-author’s picture
[19], Pappo’s group increased significantly the yield of benzaldehyde (>
90 %).
–
vation of the CH bonds of ethylbenzene and the selective formation of
–
acetophenone, but also leads to the cleavage of the CC bonds of the
ethyl groups in ethylbenzene. It was suggested that the distinctive HFIP
might anticipate and markedly promote some reaction by stabilizing
radical intermediates [30,31]. Compared with the catalytic performance
reported by Zhang et al. [12], a lower selectivity to acetophenone was
observed, and a part of ethylbenzene was converted into 1-phenyletha-
nol. The decrease in the selectivity to acetophenone might be related to
the lower capability of homogeneous NHPI/Co in comparison to
NHPI/Fe2O3 for the transformation of 1-phenylethanol to the desired
acetophenone.
In the present work, we employed HFIP as solvent in liquid phase
oxidation of ethylbenzene in the presence of NHPI and Co(II) in mo-
lecular oxygen at room temperature. The concentrations of PINO radi-
cals in different solvents were investigated by electron paramagnetic
resonance spectrometer (EPR). Further, the benzylic carbon radical was
trapped by tetramethylpiperidine N-oxyl (TEMPO) radicals and detected
by high resolution mass spectrometry (HRMS). Accordingly, the possible
mechanism of ethylbenzene oxidation was proposed.
Table 1
Experimental section
Aerobic autoxidation of ethylbenzene in different solventsa.
Selectivity (%)
Chemicals
Entry
Solvent
Conv. (%)
AP
1-PEO
BA
Others
Ethylbenzene, acetic acid, absolute ethanol, cobalt acetate tetrahy-
drate and N-hydroxyphthalimide were purchased from China Pharma-
ceutical Group Chemical Reagents Co., Ltd. 1,1,1,3,3,3-
Hexafluoropropan-2-ol was purchased from Aladdin-Reagent. All re-
agents were not purified before use.
1
2
3
4
Ethanol
Pyridine
HOAc
0
–
–
–
0
–
0
0.1
21.2
87.8
100.0
73.5
61.2
0
22.7
34.7
2.4
3.2
1.4
0.9
HFIP
a
Reaction conditions: ethylbenzene (2 mmol), NHPI (0.05 mmol), Co
(OAc)24H2O (0.01 mmol), O2 (1 atm), solvent (1 mL), 30 ◦C, 4 h. AP: Aceto-
phenone, 1-PEO: 1-Phenylethanol, BA: Benzaldehyde.
2

S. Xu et al.
Molecular Catalysis 498 (2020) 111244
The dependence of ethylbenzene reactivity in different solvents on
Table 2
The product distribution of the aerobic oxidation of ethylbenzene catalyzed by
NHPI/Co(OAc)2 at room temperature in HFIP in 4 h.
the reaction time is shown in Fig. 1. No products from ethylbenzene
oxidation were detected in 4 h when ethanol was used as the solvent.
The ethylbenzene conversion increased slightly with reaction time in the
case of HOAc as the solvent. It is surprising that a markedly enhanced
reactivity was observed when HFIP was used as the solvent, and the
ethylbenzene conversion presented a sharp rise in the initial stage of the
reaction (0–40 min) and then was retained at 80–90 % after 1 h.
The detailed product distribution from the aerobic oxidation of
ethylbenzene catalyzed by NHPI/Co(II) in HFIP in 4 h is listed in Table 2.
In the initial stage of the reaction, acetophenone was the main product
(entry 1). In a reaction time of 2 min, 1-phenylethanol (49.2 %) and a
small amount of benzaldehyde (6.3 %) were observed (entry 2). In the
first 40 min of the reaction time (entry 1–6), the ethylbenzene conver-
sion increased rapidly, indicating a high reactivity. However, the reac-
tion rate presented an apparent decrease after 40 min of reaction,
leading to a slow increase of the ethylbenzene conversion (entry 7–11).
The selectivity to 1-phenylethanol and benzaldehyde decreased
monotonously with the reaction time, the selectivity to acetophenone,
however, exhibited a monotonously increasing trend. After 4 h of re-
action (entry 11), the conversion of ethylbenzene was up to 87.8 %, and
the selectivity of ethylbenzene to acetophenone, 1-phenylethanol and
benzaldehyde were 61.2 %, 34.7 %, and 3.2 %, respectively. However, it
is worthwhile to note that the selectivity to acetophenone nearly
stopped increasing after 1 h. NHPI will be converted into N-oxyl anion
instead of N-oxyl radical in the presence of water, which is a strong
acceptor of proton [32]. Here, water was produced when ethylbenzene
or 1-phenylethanol was converted into acetophenone and its concen-
tration continuously increased, which might lead to the fact that the
selectivity to acetophenone stopped increasing.
Selectivity (%)
Entry
Reaction time (min)
Conv. (%)
AP
1-PEO
BA
Others
1
0
0.6
44.9
39.5
47.8
50.1
54.4
49.6
54.2
58.8
60.9
60.1
61.2
0
0
55.1
5.2
3.0
0.8
1.3
2.5
4.1
1.1
0.7
0.6
0.9
2
2
9.7
49.2
44.8
45.6
39.9
43.6
37.4
36.6
34.7
36.0
34.7
6.3
4.4
3.5
4.4
4.3
4.3
3.5
3.7
3.3
3.2
3
10
20
30
40
50
60
120
180
240
34.7
53.6
58.4
71.5
75.0
77.8
83.4
84.4
87.8
4
5
6
7
8
9
10
11
Reaction conditions: ethylbenzene (2 mmol), NHPI (0.05 mmol), Co(OAc)24H2O
(0.01 mmol), O2 (1 atm), HFIP (1 mL), 30 ◦C. AP: Acetophenone, 1-PEO: 1-
Phenylethanol, BA: Benzaldehyde.
oxidation. It was found that Co(acac)2 presented a slightly lower activity
and similar selectivity to acetophenone or 1-phenylethanol in compar-
ison with Co(OAc)24H2O. However, vitamin B12 gave a much lower
activity for ethylbenzene oxidation. This can be attributed to the higher
chemical state of cobalt ions in vitamin B12 in comparison with those in
Co(acac)2 and Co(OAc)24H2O.
Table S4 (ESI) demonstrates the tendency of the conversion and
selectivities of ethylbenzene versus reaction time in HOAc. It was shown
that the conversion of ethylbenzene slowly increased with the reaction
time compared to that observed in HFIP. After 4 h of reaction, the
conversion of ethylbenzene was ca. 21.2 %. In the first 90 min of reac-
tion time, acetophenone was only the product. 1-Phenylethanol and
benzaldehyde were gradually observed when the longer reaction time
was applied. The selectivity to acetophenone decreased monotonously
with the reaction time, while selectivity to 1-phenylethanol showed a
contrary trend after 2 h of reaction. Under the applied reaction condi-
tions, the selectivity to benzaldehyde decreased monotonously with the
reaction time.
In order to probe the influence of other factors on oxidation of eth-
ylbenzene to acetophenone, the reaction temperature, the concentration
of NHPI and the kind of cobalt salts were chosen as the variations and
the resulting catalytic performance of the homogeneous NHPI/Co are
shown in Table S1, S2 and S3. It was found that the ethylbenzene
conversion increased from 54.0%–81.8% when increasing the reaction
temperature varied from 20 ◦C to 40 ◦C with other reaction conditions
unchanged (Table S1). At the same time, an increase in the selectivity to
acetophenone and a decrease in the selectivity to 1-phenylethanol were
observed (Table S1), suggesting that the enhanced reaction temperature
promoted the conversion of 1-phenylethanol into acetophenone.
Furthermore, the higher NHPI concentration was found favorable for the
transformation from ethylbenzene to acetophenone, which might be
ascribed to the enhancement of the concentration of active PINO radi-
cals (Table S2).
–
Previous researches suggested that CH bonds can be activated by
PINO radical generated from NHPI to abstract hydrogen and form car-
bon radical [19,29,33,34,35,37]. Here, electron paramagnetic reso-
nance (EPR) was used to explore the PINO radicals and their
concentration. As far as we know, there was no report that confirmed
that both PINO and PhCHCH3 were present in the oxidation of ethyl-
benzene. When ethanol was used as the solvent, there was no the signal
of PINO radical observed (Fig. S1, ESI). Comparatively, the g-factor of
this triplet signal was detected in 2.0070 in HFIP and HOAc, indicating
the generation of the PINO radical (g = 2.0070) (Fig. 2A) [12,19,36,38].
Based on the facts that there was no products generated from ethyl-
benzene oxidation in ethanol, it was concluded that PINO radical should
be related to the ethylbenzene oxidation. Furthermore, compared the
concentration of PINO radical in the HFIP reaction solution to that in
acetic acid, it was found that the conversion of ethylbenzene and the
concentration of PINO were well in agreement in different solvents.
Fig. 2 (B) further explored the change of PINO radical concentration
versus reaction time in HFIP. It was found that the concentration of
PINO radical increased before 20 min and then decreased with reaction
time. The signal of PINO radical nearly disappeared after 2 h. Compared
with the trend of ethylbenzene conversion versus reaction time, it can be
concluded that the higher concentration of PINO radicals promoted the
ethylbenzene conversion and the reaction rate, and PINO concentration
were well positively correlated.
Table S3 shows effect of the kind of cobalt salts on the ethylbenzene
Tetramethylpiperidine N-oxyl radicals (TEMPO) were added to the
reaction solution to trap the possible PhCHCH3 radical, the possible
adduct generated by the coupling of PhCHCH3 radical and TEMPO were
detected by high resolution mass spectrometry (HRMS) (Scheme 1)[39].
Fig. 1. Effects of different solvents on oxidation of ethylbenzene.
3

S. Xu et al.
Molecular Catalysis 498 (2020) 111244
Fig. 2. PINO determine by EPR: (A) in different solvents at 2 min; (B) in HFIP with different reaction time.
Scheme 1. Trapping of the PhCHCH3 radicals by TEMPO.
The HRMS of the reaction solution after adding the trapping agent
believed that ca. 6.3 % of ethyl phenyl hydrogen peroxide (EPHP) was
produced under the optimized experimental conditions, which was
oxidized to acetophenone under the normal analytical conditions by gas
chromatography.
are shown in Fig. 3. TEMPO was added to the reaction solution after 30
min of reaction, the MS signal at m/z 262.2154 (C17H27NO + H+) was
observed. This signal is related to the formation of the adduct of
PhCHCH3 with the trapping agent TEMPO, implying that the hydrogen
Based on the PINO radical, PhCHCH3 and ethyl phenyl hydrogen
peroxide observed under the experimental conditions and the activation
–
atom of benzylic CH bonds in ethylbenzene was abstracted and the
–
–
bonds proposed by other researchers [19,29,
CH bonds of ethylbenzene was activated (Fig. 3A). After 4 h of reac-
mechanism of CH
tion, the signal at m/z 262.2154 was very weak (Fig. 3B). No signal was
observed to show the presence of the adduct between PhCHCH3 and
TEMPO in the absence of TEMPO (Fig. 3C). Combined the results from
catalytic test and EPR analysis, it is suggested that PINO radicals may
33–37], a possible reaction route for liquid phase oxidation of ethyl-
benzene under our experimental conditions was proposed (Fig. S3, ESI).
First, the initiator Co2+ reacts with O2 to form cobalt oxygen complexes.
The complexes abstract a hydrogen atom of NHPI to generate PINO
radical. Then, PINO radical activates ethylbenzene to form PhCHCH3.
PhCHCH3 reacts with O2 to form PhCH(OO)CH3 and then abstract a
hydrogen atom from NHPI to become PhCH(OOH)CH3. The generated
hydroperoxide will decompose and produce acetophenone and
1-phenylethanol.
–
activate the benzylic CH bond by abstracting hydrogen from the
benzylic carbon to generate carbon radical. At the initial stage of the
reaction (30 min) in HFIP, the high concentration of PINO radical led to
a high concentration of PhCHCH3 radical. No PINO radical was detected
after 4 h of reaction, while the concentration of PhCHCH3 was also very
low that was determined by HRMS analysis. The above results indicated
that the concentration of PINO radical may be crucial to the activation of
The liquid phase oxidation reaction network of ethylbenzene in
different solvents was further investigated using acetophenone and 1-
phenylethanol as starting materials under the same reaction condi-
tions, and the results are shown in Table S6 (ESI). No further oxygen-
ated product from acetophenone was detected in both HOAc and HFIP,
while 1-phenylethanol can be further oxidized to acetophenone under
the same conditions. The conversion of 1-phenylethanol was much
higher in HFIP than that in HOAc (91.5 % vs. 64.3 %) with a similar
selectivity to acetophenone of ca. 100 %. Based on the above results and
the fact that ethyl phenyl hydrogen peroxide (EPHP) was confirmed as
the intermediate of the products (Table S5, ESI), the reaction network of
the ethylbenzene oxidation in HFIP as solvent was proposed (Fig. 4).
Benzaldehyde and ethyl phenyl hydrogen peroxide (EPHP) were directly
generated from ethylbenzene. 1-phenylethanol was directly generated
–
benzylic carbon CH bonds. In addition, the signal at m/z 262.2158 was
also detected when HOAc was used as solvent (Fig. S2, ESI), indicating
that the PhCHCH3 is the intermediate of ethylbenzene oxidation in both
acetic acid and HFIP.
In order to clarify how ethylbenzene is converted into acetophenone
and 1-phenylethanol, the possible intermediate ethyl phenyl hydrogen
peroxide (EPHP) was analyzed using triphenylphosphine (TPP). The
results are shown in Table S5 in ESI. An increase of ca. 6.3 % in the
selectivity to 1-pheylethanol (1-PEO) was observed after adding TPP to
reaction mixture. At the same time, a roughly same decrease in the
selectivity to acetophenone was found. However, 1-PEO is stable under
the analytical conditions by gas chromatography. Therefore, it is
4

S. Xu et al.
Molecular Catalysis 498 (2020) 111244
Fig. 3. HRMS of the reaction solution after adding TEMPO at (A) 30 min, (B) 4 h and (C) without TEMPO.
acetophenone was up to 61.2 %. The analysis via EPR indicated that the
solvent HFIP is beneficial to the rapid generation of PINO radical. The
PINO radical may be the actual catalyst for the reaction, which activates
–
the ethylbenzene benzylic CH bonds by abstracting hydrogen on the
benzylic carbon to produce PhCHCH3, which was confirmed by HRMS
detection. The ethylbenzene oxidation reaction in HFIP was suggested to
proceed according to the picture proposed by Ishii.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
CRediT authorship contribution statement
Sihao Xu: Conceptualization, Methodology, Funding acquisition,
Writing - review & editing, Conceptualization, Methodology, Funding
acquisition, Writing - review & editing. Guojun Shi: Investigation,
Writing - original draft. Ya Feng: Investigation. Chong Chen: Re-
sources, Investigation. Lijun Ji: Methodology.
Fig. 4. Possible reaction network of the ethylbenzene oxidation catalyzed by
Acknowledgements
NHPI/Co(II) in HFIP.
The authors gratefully acknowledge the project funded by the Pri-
ority Academic Program Development of Jiangsu Higher Education
Institutions.
from 1-phenylethyl hydrogen peroxide, but acetophenone may be from
oxidation of 1-phenylethanol and/or decomposition of ethyl phenyl
hydrogen peroxide.
Appendix A. Supplementary data
Conclusions
Supplementary material related to this article can be found, in the
online version, at doi:https://doi.org/10.1016/j.mcat.2020.111244.
In summary, the solvent HFIP markedly improved the trans-
formation of ethylbenzene into acetophenone in liquid phase using
molecular oxygen. Under the experimental conditions, the conversion of
ethylbenzene was as high as 87.8 % in 4 h, and the selectivity to
5

S. Xu et al.
Molecular Catalysis 498 (2020) 111244
[22] Z.H. Li, B. Fiser, B.L. Jiang, J.W. Li, B.H. Xu, S.J. Zhang, Org. Biomol. Chem. 17
(2019) 3403–3408.
References
[23] R.D. Patil, B. Fuchs, N. Taha, Y. Sasson, ChemistrySelect 1 (2016) 3791–3796.
[24] (a) S. Mahmood, B.-H. Xu, T.-L. Ren, Z.-B. Zhang, X.-M. Liu, S.-J. Zhang, Mol.
Catal. 447 (2018) 90–96;
[1] R.F. Xie, G.L. Fan, L. Yang, F. Li, Chem. Eng. J. 288 (2016) 169–178.
[2] Y. Chen, S.F. Zhao, Z.G. Liu, Phys. Chem. Chem. Phys. 17 (2015) 14012–14020.
[3] A.R. Faraji, S. Mosazadeh, F. Ashouri, J. Colloid Interface Sci. 507 (2017) 10–26.
[4] S.S. Jie, X. Lin, Z.S. Chao, Z.G. Liu, Mater. Chem. Phys. 214 (2018) 239–246.
[5] S. Devika, M. Palanichamy, V. Murugesan, Appl. Catal. A Gen. 407 (2011) 76–84.
[6] A.L.G. Pinheiro, A.P.S. Oliveira, F.F. de Sousa, J.M. Soares, G.D. Saraiva, A.
C. Oliveira, R. Lang, Catalysts 8 (2018).
(b) G. Dobras, M. Sitko, M. Petroselli, M. Caruso, M. Cametti, C. Punta,
B. Orlinska, ChemCatChem. 12 (2020) 259–266.
[25] (a) T. Dohi, N. Yamaoka, Y. Kita, Tetrahedron 66 (2010) 5775–5785;
(b) H. Hamamoto, G. Anilkumar, H. Tohma, Y. Kita, Chem. - Eur. J. 8 (2002)
5377–5383;
[7] H.J. Sanders, H.F. Keag, H.S. McCulloug, Ind. Eng. Chem. 45 (1953) 2–16.
[8] A. Selvamani, M. Selvaraj, P.S. Krishnan, M. Gurulakshmi, K. Shanthi, Appl. Catal.
A Gen. 495 (2015) 92–103.
(c) E. Gaster, Y. Vainer, A. Regev, S. Narute, K. Sudheendran, A. Werbeloff,
H. Shalit, D. Pappo, Angew. Chem., Int. Ed. 54 (2015) 4198–4202.
[26] I.A. Shuklov, N.V. Dubrovina, A. Boerner, Synthesis-stuttgart 19 (2007)
2925–2943.
[9] C.A. Korologos, M.D. Nikolaki, C.N. Zerva, C.J. Philippopoulos, S.G. Poulopoulos,
J. Photochem. Photobiol., A 244 (2012) 24–31.
[27] J. Graton, Z. Wang, A.M. Brossard, D.G. Monteiro, J.Y. Le Questel, B. Linclau,
Angew. Chem., Int. Ed. 51 (2012) 6176–6180.
[10] S. Larrondo, G. Alvaro, B. Irigoyen, N. Amadeo, Catal. Today 107 (2005) 444–450.
[11] C.X. Miao, H.Q. Zhao, Q.Y. Zhao, C.G. Xia, W. Sun, Catal. Sci. Technol. 6 (2016)
1378–1383.
[28] A. Berkessel, J.A. Adrio, D. Huettenhain, J.M. Neudorfl, J. Am. Chem. Soc. 128
(2006) 8421–8426.
[12] C.F. Zhang, Z.P. Huang, J.M. Lu, N.C. Luo, F. Wang, J. Am. Chem. Soc. 140 (2018)
2032–2035.
[29] E. Gaster, S. Kozuch, D. Pappo, Angew. Chem., Int. Ed. 56 (2017) 5912–5915.
[30] T. Dohi, N. Yamaoka, Y. Kita, Tetrahedron 66 (2010) 5775–5785.
[31] S.P. de Visser, J. Kaneti, R. Neumann, S. Shaik, J. Org. Chem. 68 (2003)
2903–2912.
[13] L.L. Fu, S.F. Zhao, Y. Chen, Z.G. Liu, Chem. Commun. 52 (2016) 5577–5580.
[14] W. Jiang, J. Yang, Y.Y. Liu, J.F. Ma, Chem. Commun. 52 (2016) 1373–1376.
[15] A.R. Faraji, F. Ashouri, Z. Hekmatian, S. Heydari, S. Mosazadeh, Polyhedron 157
(2019) 90–106.
[32] Y. Cai, N. Koshino, B. Saha, J.H. Espenson, J. Org. Chem. 70 (2005) 238–243.
[33] W.Y. Zhou, X. Dai, Y. Chen, F.A. Sun, M.Y. He, Q. Chen, ChemistrySelect 3 (2018)
566–572.
[16] M. Varyani, P.K. Khatri, S.L. Jain, Tetrahedron Lett. 57 (2016) 723–727.
[17] S. Debnath, S.K. Saxena, V. Nagabhatla, Catal. Commun. 84 (2016) 129–133.
[18] N. Benferrah, M. Hammadi, C. Philouze, F. Berthiol, F. Thomas, Inorg. Chem.
Commun. 72 (2016) 17–22.
[34] D. Habibi, A.R. Faraji, M. Arshadi, H. Veisi, A. Gil, J. Mol. Catal. A Chem. 382
(2014) 41–54.
[35] Y.L. Zhou, S.S. Lin, D.H. Xia, Y.Z. Xiang, Chin. Chem. Lett. 23 (2012) 895–898.
[36] S. Fan, W.J. Dong, X.B. Huang, H.Y. Gao, J.J. Wang, Z.K. Jin, J. Tang, G. Wang,
ACS Catal. 7 (2017) 243–249.
[19] Y. Yoshino, Y. Hayashi, T. Iwahama, S. Sakaguchi, Y. Ishii, J. Org. Chem. 62 (1997)
6810–6813.
[20] L.Y. Wang, Y.B. Zhang, R.F. Du, H.R. Yuan, Y.T. Wang, J. Yao, H.R. Li,
ChemCatChem 11 (2019) 2260–2264.
[37] B.J. Gao, S.Q. Meng, X.L. Yang, Org. Process Res. Dev. 19 (2015) 1374–1382.
[38] F. Recupero, C. Punta, Chem. Rev. 107 (2007) 3800–3842.
[39] X.D. Jia, Y.Z. Zhu, Y. Yuan, X.W. Zhang, S.W. Lu, L. Zhang, L.L. Luo, ACS Catal. 6
(2016) 6033–6036.
[21] H.D. Qiu, W.D. Liang, G.J. Zhang, M.M. Lin, W.M. Liu, Z.H. Gao, W.T. Wei, C.
L. Tang, H.X. Jin, H.Z. Liang, X.J. Yan, ChemistrySelect 3 (2018) 12363–12366.
6