Supported CuCl/Γ-Al2O3 for Friedel–Crafts Acylation with Effective Inhibition of Defluorination

Article abstract of DOI:10.1007/s10562-017-2070-0

A series of supported CuCl/γ-Al₂O₃ with different Cu content (2, 6, 10, 13, 16 and 17?wt%) were prepared by the method of wet impregnation and spontaneous monolayer dispersion techniques. These co-catalysts were analyzed by XRD, BET, SEM, XPS and ICP techniques and were found to inhibit the formation of defluorination byproducts during the phenylacetylation of fluorobenzene catalyzed by AlCl₃, allowing the using of fluorobenzene as the reaction solvent that could be easily recycled. The best inhibitor was 10?wt% Cu supported CuCl/γ-Al₂O₃, which still had good catalytic performance after being recycled for five times. Graphical Abstract: [Figure not available: see fulltext.].

Full text of DOI:10.1007/s10562-017-2070-0

Catal Lett (2017) 147:2225–2231
DOI 10.1007/s10562-017-2070-0
Supported CuCl/γ-Al2O3 for Friedel–Crafts Acylation
with Effective Inhibition of Defluorination
Yanhong Wang1 · Jiahong Wang1 · Zhongzhu Long2 · Shuihong Cai3 ·
Qiaochun Wang1
Received: 26 December 2016 / Accepted: 2 May 2017 / Published online: 1 June 2017
© Springer Science+Business Media New York 2017
Abstract A series of supported CuCl/γ-Al2O3 with dif-
ferent Cu content (2, 6, 10, 13, 16 and 17 wt%) were pre-
pared by the method of wet impregnation and spontaneous
monolayer dispersion techniques. These co-catalysts were
analyzed by XRD, BET, SEM, XPS and ICP techniques
and were found to inhibit the formation of defluorination
byproducts during the phenylacetylation of fluorobenzene
catalyzed by AlCl3, allowing the using of fluorobenzene as
the reaction solvent that could be easily recycled. The best
inhibitor was 10 wt% Cu supported CuCl/γ-Al2O3, which
still had good catalytic performance after being recycled
for five times.
Graphical Abstract
Keywords Supported catalysts · CuCl · Al2O3 · Friedel–
Electronic supplementary material The online version of this
article (doi:10.1007/s10562-017-2070-0) contains supplementary
material, which is available to authorized users.
Crafts acylation · Defluorination
1 Introduction
* Qiaochun Wang
qcwang@ecust.edu.cn
Friedel–Crafts acylation is a common method to synthe-
size aromatic ketones, which are widely used as intermedi-
ates in manufacturing special chemicals such as perfumes,
pharmaceutical, dyes and agricultural chemicals [1–4].
However, it has been reported that defluorination would
occur during the Friedel–Crafts acylation when fluoro-
aromatic substrates were used [5]. Although the content
of defluorinated by-product is low in many cases, it has
1
Key Laboratory for Advanced Materials and Institute of Fine
Chemicals, School of Chemistry & Molecular Engineering,
East China University of Science & Technology, Shanghai,
People’s Republic of China
2
College of Pharmacy, East China University of Science &
Technology, Shanghai, People’s Republic of China
3
Shanghai Dongyue Biochem. Co. Ltd., Shanghai 200237,
People’s Republic of China
Vol.:(0123456789)
1 3

2226
Y. Wang et al.
great impacts on the quality of the final product, especially
for those fluorinated pharmaceuticals. For example, M2
(Scheme 1) is the key intermediate for synthesis of atorv-
astatin calcium-one of the famous lipid-lowering agents
used for the prevention of cardiovascular disease―and is
usually synthesized by phenylacetylation of fluorobenzene,
accompanied by the formation of defluorinated byprod-
uct 1 [6, 7], whose content was required to be lower than
0.2 wt%. In order to achieve high-grade M2, the method
of using fluorobenzene as both substrate and reaction sol-
vent is unacceptable because excess fluorobenzene can
hardly be recycled for two reasons, one is that the formed
benzene cannot be removed from fluorobenzene by sim-
ple distillation (the difference of their boiling points is so
small), and the other is that benzene is easier to be acylated
than fluorobenzene. As a result, other solvents apart from
fluorobenzene were used [7–9], and once they are recy-
cled, benzene should be completely removed, which greatly
increases the cost of production.
Stoichiometric Na2SO3 aqueous solution was subsequently
added dropwise to the mixture under stirring. The obtained
mixture was stirred for 2 h and then 2 equiv. of H2SO3 solu-
tion was added. The resulting mixture was centrifuged suf-
ficiently and the supernate was removed. The left wet sol-
ids were dried for 4 h at 50°C in a vacuum oven to avoid
air contact. The dried CuCl/γ-Al2O3 mixture was finally
transferred into a tube furnace and calcinated at 380°C for
4 h. The obtained CuCl/γ-Al2O3 co-catalyst was donated as
Cu@Al (X); where X stands for the content of the total Cu
in the catalysts.
2.2 The Phenylacetylation Reaction of Fluorobenzene
The phenylacetylation reaction of fluorobenzene was then
carried out at 0–5°C with the existence of anhydrous AlCl3
and CuCl/γ-Al2O3. Target product M2 and the by-product
1 were obtained by column chromatography on silica gel.
1
M2 H-NMR (DMSO, 400 MHz, TMS) δ 8.18–8.09 (m,
It is consequently worthwhile to develop efficient inhibi-
tors to restrain the occurrence of defluorinated byproduct
during the acylation of fluorobenzenes. Herein we describe
the using of supported CuCl/γ-Al2O3 as inhibitor for phe-
nylacetylation of fluorobenzene. The optimization of the
inhibitor was carried out and 10 wt% Cu supported CuCl/γ-
Al2O3 was found to be the best one. It still kept excellent
defluorination inhibiting performance after being recycled
for five times.
2H), 7.41–7.20 (m, 7H), 4.39 (s, 2H); EI-TOF MS: m/z
calculated for [C14H11OF]: 214.0794, found: 214.0793. By
1
product 1 H-NMR (DMSO, 400 MHz, TMS) δ 8.09–8.02
(m, 2H), 7.68–7.19 (m, 8H), 4.40 (s, 2H); EI-TOF MS: m/z
calculated for [C14H12O]: 196.0888, found: 196.0886.
3 Results and Discussion
3.1 The Determination of Using CuCl/γ-Al2O3
as Defluorination Inhibitor
2 Experimental
The defluorination phenomenon has been commonly found
in the synthesis of organic fluorinated compounds such as
1-(4-fluorophenyl)-2-phenyleththanone [5], antidepressant
paroxetine hydrochloride [14], ezetimibe [15], double-
hexyl-trifluoro-alkyl benzene [16] and so on. It is believed
that the formation of defluorination by-product is mainly
attributed to the cleavage of C–F bond [15]. There are three
main pathways that can cause the cleavage of C–F bond,
namely, electron transfer, oxidative addition and nucleo-
philic pathways, respectively [17, 18]. The electron transfer
pathway often involves homolytic splitting of C–F bond via
2.1 Preparation of Supported CuCl/γ-Al2O3
The CuCl/γ-Al2O3 inhibitors (in the 2–17 wt% Cu range,
detected by ICP) were prepared according to the wet
impregnation method [10] and spontaneous monolayer
dispersion techniques [11–13]. The γ-Al2O3 powder was
firstly activated in a tube furnace under dry nitrogen atmos-
phere at 500°C for about 4 h before being used. Then
an exact amount of the activated γ-Al2O3 powder was
added to a CuCl2 aqueous solution with vigorous stirring.
Scheme 1 Phenylacetylation of fluorobenzene
1 3

Supported CuCl/γ-Al2O3 for Friedel–Crafts Acylation with Effective Inhibition of…
2227
a single-electron transfer process which is mediated either
by electrochemically or low-valent metals. It often pro-
ceeds by initial electron transfer from a reductant, followed
by scission of the C–F bond to give a fluoride ion and a
carbon radical. The oxidative addition pathway means
employing transition metals for C–F bond cleavage through
oxidative addition of the C–F bond to the transition metal
center. The nucleophilic pathway refers to nucleophilic
substitution reaction in which fluorine atom is substituted
by nucleophile such as OH−, SH− and so on.
surface area, and as a result could reduce the amount of
inhibitors. Inspired by the fact that a number of suitable
supports, such as SiO2 [21, 22], activated carbons [23,
24], zeolite [25, 26] and γ-Al2O3 [27–29], could be used
to prepare high efficient immobilized catalysts, and the
work of Tang and coworkers, where monolayered CuCl
was dispersed on γ-Al2O3 [11], a series of γ-Al2O3 sup-
ported CuCl of different Cu contents (Cu@Al (X), X = 2,
6, 10, 13, 16 and 17%) were prepared to serve as the co-
catalysts for inhibiting the defluorination during the phe-
nylacetylation of fluorobenzene.
Among these three mechanisms, the electron transfer
pathway should be the most reasonable way in our work,
because that there is no existence of transition metal or
nucleophile, while there is free aluminum impurities (which
meets the requirement of the presence of low-valent metals
in this mechanism) in anhydrous AlCl3 during its industrial
manufaction [19, 20]. The side reactions during the acyla-
tion can thus be deduced as (1) fluorobenzene was firstly
reduced to anion radical by gaining one electron from the
trace free Al in AlCl3; (2) the anion radical dissociated as
a benzene radical and a F−; (3) the benzene radical gener-
ates benzene by hydrogen abstraction reaction; and (4) the
acylation of the formed benzene, which has higher reactiv-
ity to be acylated than fluorobenzene, leads to the forma-
tion of the by-product 1. In this way, the occurrences of by-
product 1 and benzene in Scheme 1 are thus explained.
Then it is hoped that Cu2+ or Cu+ could have defluori-
nation inhibiting effect because of their electron-accepting
nature, in the way of being reduced to Cu+ or Cu by gaining
the electron, which would no longer come to fluobenzene
again from Al. Before verifying the defluorination inhibit-
ing effect of Cu2+/Cu+ during the acylation of fluoroben-
zene, the content of benzene among the starting fluoroben-
zene was identified as 0.011% by GC measurement. Then
phenylacetylation of fluorobenzene under AlCl3 catalyst
in the presence of CuCl, CuO, CuCl2 and the blank con-
trol were then carried out. The contents of benzene among
the first run of the recycled fluorobenzene (Cben) with 1.5
equiv. CuCl, CuO, CuCl2 and blank control were 0.011,
0.032, 0.030 and 0.030%, respectively. It can be seen that
CuCl had inhibiting effect for defluorination while CuO
and CuCl2 has no effect, indicating that CuCl could be used
as the defluorination inhibitor, probably due to its covalent
character.
3.2 The Characterizations of the Supported CuCl/
γ-Al2O3
3.2.1 XRD
XRD patterns of supported CuCl/γ-Al2O3 inhibitors are
shown in Fig. 1. The diffraction peaks corresponding to
free CuCl could not be observed in Cu@Al (2%), Cu@
Al (10%) and Cu@Al (13%), while the increasing reflec-
tions corresponding to crystalline CuCl at 2θ = 28.6°,
47.5° and 56.3° were clearly observed in Cu@Al (16%)
and Cu@Al (17%). These results indicate that those co-
catalysts whose Cu content lower than 16% formed mon-
olayer CuCl on the γ-Al2O3 surface, while CuCl began to
stack when the Cu content becomes equal to or greater
than 16%, this phenomenon is in accordance with the
reported result that 0.33 g of CuCl salt was necessary to
form a monolayer onto the surface of 1 g γ-Al2O3 [30]
(Cu content was calculated as 15.9%).
Then different molar ratios of CuCl were then used
to evaluate the inhibiting effect. GC results of recycled
solvents were shown in Table S1. It was found that the
Cben decreased with the addition of CuCl and became
the same as that of the starting fluorobenzene when 1.5
equiv. of CuCl was added. Considering that industrial
CuCl were nonuniform in size and has small surface area,
supported CuCl was supposed to have better defluorina-
tion-inhibiting performance for possessing a much larger
Fig. 1 XRD patterns of γ-Al2O3(a), Cu@Al (2%) (b), Cu@Al (10%)
(c), Cu@Al(13%) (d), Cu@Al(16%) (e), and Cu@Al(17%) (f)
1 3

2228
Y. Wang et al.
3.2.2 BET
3.3 The Defluorination-Inhibiting Performance
of the CuCl/γ-Al2O3
The surface area of samples was calculated by using BET
method; the pore volume distributions were calculated by
BJH method, the results were listed in Table 1. It can be
seen that the BET surface area and pore volume of γ-Al2O3
were correspondingly 178.2 m2 g−1 and 0.253 cm3 g−1, and
they decreased gradually to 69.4 m2 g−1 and 0.137 cm3 g−1
respectively with the increasing contents of Cu from 2 to
17%. This is no doubt because that the increasing loading
of CuCl onto the supporter would reduce the specific sur-
face area and the pore volume.
The defluorination-inhibiting performance experiments
using the Cu@Al(X) were then carried out and the GC
results are listed in Table 1. It can be found that the using
of CuCl/γ-Al2O3 indeed restrained the defluorination
because all the Cbens were lower than that of the blank con-
trol (0.030%). The Cben became equal to that of the starting
fluorobenzene (0.011%) when the content of X reached to
6%, then it came to a lower content of 0.008% when X is
10%, and finally kept nearly unchanged in spite of X came
to 17%.
3.2.3 XPS
It should be pointed out that the removal of the CuCl/γ-
Al2O3 by filtration using filter paper was still facile when
X≤10% but became quite difficult when X reached to 13%.
The SEM images of CuCl/γ-Al2O3 might offer the reason,
as illustrated in Fig. 3. It can be seen that the loading of
CuCl caused the collapse of Al2O3 particles. The mor-
phology of γ-Al2O3 was regular with an average particle
diameter of ~50 μm, and became less regular with a size
distribution from ~20 to 50 μm when X=10%, with a size
distribution from ~10 μm to 40 μm when X=13%, and
finally with a size distribution from several micrometers
to 20 μm when X=16%. The bad filterability should be
attributed to the decreasing size of the co-catalyst with the
increasing X.
The obtained XPS spectra were dealt with XPS Peak Fit-
ting Program, which made it possible to calculate the ratio
of Cu+:CuO:Cu2+. The peak fitting results are shown in
Fig. 2. The ratios of Cu+:CuO:Cu2+ in Cu@Al (2%), Cu@
Al (6%),Cu@Al (10%), Cu@Al (13%), Cu@Al (16%) and
Cu@Al (17%) were 1.2:2.1:1.0, 1.2:1.0:1.0, 1.4:1.0:1.0,
1.6:1.1:1.0, 1.7:1.2:1.0 and 1.8:1.2:1.0, respectively. Tak-
ing into account that Cu2+ and CuO has no contributions
for defluoronation, the percentages of Cu+ among the total
copper in these inhibitors would reflect their inhibiting
activities. The Cu+ contents were then calculated as 27.9,
37.5, 41.2, 43.2, 43.6 and 45.0%, respectively. It can be
found that the percentage of Cu+ among the total copper
(Cu+/Cu%) in these co-catalysts increased sharply before
the Cu content reaches 10%, while it grown little when
Cu+/Cu% exceed 10%.
Taking into account the inhibiting performance, the ease
of filtration and XPS result, Cu@Al (10%) was deemed
to be the best inhibitor and the optimal amount of Cu@
Al (10%) was then investigated, as illustrated in Table S2.
The defluorination-inhibiting performance continued to
improve until the amount of co-catalyst reached to 0.5
equiv. (based on Cu), and then remained at this level even
0.8 equiv. of Cu@Al (10%) was added. The using of 0.5
equiv. amount of the inhibitor was consequently adopted
for the next recycling experiments.
Table 1 BET data of the co-catalysts and the respective GC data of
recycled fluorobenzene
Entry Samples
BET sur-
face area
Pore
Recycled fluoroben-
zene
volume
(m2 g−1
)
(cm3 g−1
)
3.4 The Recycling of Cu@Al (10%) Inhibitor
Fluoroben- Cben (%)
zene (%)
The recycling experiments using 0.5 equiv. of Cu@Al
(10%), along with the blank controls were carried out and
the corresponding GC results of recycled fluorobenzene
and HPLC data of obtained M2 are listed in Tables 2 and 3.
It can be seen that Cu@Al (10%) still had good defluorina-
tion-inhibiting performance after being used for five times.
The Cbens were restrained lower than that of the starting
fluorobenzene; the contents of defluorinated 1 in M2 were
kept at a satisfactory level of lower than 0.05%; and the
yields of M2 were the same level as the blank controls.
1
2
Al2O3
178.2
157.4
0.253
0.238
–
–
Cu@Al
99.80
0.013
(2%)
3
4
5
6
7
Cu@Al
(6%)
120.3
98.5
81.8
72.7
69.4
0.198
0.160
0.142
0.138
0.137
99.98
99.96
99.72
99.85
99.90
0.011
0.008
0.009
0.008
0.009
Cu@Al
(10%)
Cu@Al
(13%)
Cu@Al
(16%)
Cu@Al
(17%)
1 3

Supported CuCl/γ-Al2O3 for Friedel–Crafts Acylation with Effective Inhibition of…
2229
Fig. 2 XPS spectra and the fitting Cu+:CuO:Cu2+ ratios of Cu@Al (2%) (a), Cu@Al (6%) (b), Cu@Al (10%) (c), Cu@Al (13%) (d), Cu@Al
(16%) (e) and Cu@Al (17%) (f)
1 3

2230
Y. Wang et al.
Fig. 3 The SEM images of γ-Al2O3 (a), Cu@Al (10%) (b), Cu@Al (13%) (c) and Cu@Al (16%) (d)
Table 2 GC data of recycled fluorobenzene and HPLC results of M2
products among the recycling experiments with 0.5 equiv. amount
Cu@Al (10%) co-catalyst
the form of monolayer onto γ-Al2O3 when mass content of
Cu was lower than 16%. These CuCl/γ-Al2O3 all exhibited
good defluorination-inhibiting nature during the phenyla-
cetylation of fluorobenzene. The best inhibitor was Cu@Al
(10%) and the optimal amount added was 0.5 equiv. basing
on immobilized Cu. The Cu@Al (10%) inhibitor kept its
good performance after being reused for at least five times
and could be recycled through simple filtration, making the
using of fluorobenzene as both substrate and reaction sol-
vent possible during the acylation.
Entry
Recycled fluorobenzene
Products
M2 (%)
Yield (%)
Fluoroben-
zene (%)
Cbens (%)
1 (%)
Run1
Run2
Run3
Run4
Run5
Run6
99.96
99.91
99.94
99.89
99.82
99.98
0.008
0.008
0.009
0.010
0.009
0.007
98.93
97.77
99.43
99.21
99.63
99.29
0.044
0.030
0.026
0.028
0.031
0.039
76
74
78
76
75
78
Acknowledgements This work was financially supported by NSFC/
China (21572063, 21372076), the Science Fund for Creative Research
Groups (21421004), and the Programme of Introducing Talents of
Discipline to Universities (B16017).
Table 3 GC data of the recycled fluorobenzene and HPLC results of
M2 products in the recycling experiments of blank control
References
Entry
Recycled fluorobenzene
Products
M2 (%)
Yield(%)
1. Liu T, Cheng H, Sun L, Zhang C (2016) Appl Catal A 512:9
2. Madhavi S, Sreenivasulu R, Yazala J-P, Raju R-R (2017) Saudi
Pharm J 25:275
Fluoroben-
zene (%)
Cbens (%)
1 (%)
3. Xu G, Guo J, Zhang Y, Guo Q (2015) Chem Catal Chem 7:2485
4. Balta D-K, Karahan O, Avci D, Arsu N (2015) Prog Org Coat
78:200
Run1
Run2
Run3
99.95
99.95
99.88
0.030
0.041
0.062
98.79
98.70
98.70
0.152
0.341
0.465
76
77
76
5. Xing Y, Chen S, Zhou Y, Liu N, Chen L (2015) Synth Commun
45:2832
6. Wen J, Qi H, Kong X, Chen L, Yan X (2014) Synth Commun
44:1893
By comparison in those blank controls (Table 3), Cben
kept increasing to over 0.06% with the increasing runs of
recycling, and the content of 1 in M2 began to exceed the
required 0.2% at the second run and kept increasing to
0.46% after three times of recycling.
7. Laube M, Tondera C, Wuest F, Pietzsch J, Kniess T (2014) RSC
Adv 4:38726
8. Yu J-W, Shuai M, Wang Y-Q (2015) Tetrahedron Lett 56:1575
9. Klein D-J, Modarelli D-A, Harris F-W (2001) Macromolecules
34:2427
10. Khan N-A, Hasan Z, Min K-S, Paek S-M, Jhung S-H (2013) Fuel
Process Technol 116:265
11. Gui L, Guo Q, Xie Y, Tang Y (1984) Sci Chin B 5:445
12. Xie Y, Tang Y (1990) Adv Catal 37:1
13. Thewis J (1962) Encyclopaedic dictionary of physics. Pergamon
Press, Oxford
4 Conclusions
14. Wang F, Sun X, Yu W (2015) Guangdong Chem Ind (CN)
42:250
In this work, supported CuCl/γ-Al2O3 with 2–17% mass
contents of Cu were prepared by using wet impregnation
method and spontaneous monolayer dispersion techniques.
These Cu@Al (X)s were characterized by XRD, BET,
SEM, XPS and ICP techniques. CuCl could be dispersed in
15. Hao C, Song S, Feng Y, Cheng M, Zhao D (2015) Chin J Med
Chem 25:211
16. Su X, Huo X, Qian X, Yang C (2010) Chin J Liq Cryst Displays
25:316
1 3

Supported CuCl/γ-Al2O3 for Friedel–Crafts Acylation with Effective Inhibition of…
2231
17. Burdeniuc J, Jedlicka B, Crabtree R-H (1996) Chem Ber 130:145
18. Stahl T, Klare H-F, Oestreich M (2013) ACS Catal 3:1578
19. Chinese standard GB/T3959-2008 industrial anhydrous alu-
minum chloride
24. Asfaram A, Ghaedi M, Agarwal S, Gupta V-K (2015) RSC Adv
5:18438
25. Dai Q, Bai S, Lou Y, Wang X, Guo Y, Lu G (2016) Nanoscale
8:9621
20. Yan S, Shi R (2007) Inner Mongolia Petrochem Ind (CN) 3:54
21. Jeong E-S, Park C-I, Jin Z, Hwang I-H, Son J-K, Han S-W
(2015) Catal Lett 145:971
26. Jin Z, Liu S, Wang X (2013) Appl Catal A 453:295
27. Kumar P-A, Ha H-P (2010) Catal Lett 136:177
28. Ling P, Li D, Wang X (2012) J Mol Catal A 357:112
29. Chaieb T, Delannoy L, Louis C, Thomas C (2016) Catal Lett
146:2622
22. Kalyoncu S, Duzenli D, Onal I, Seubsai A, Noon D (2015) Catal
Lett 145:596
23. Martin-Martinez M, Gomez-Sainero L-M, Palomar J, Omar S,
Rodriguez J-J (2016) Catal Lett 146:2614
30. Hernández-Maldonado A-J, Qi G, Yang R(2005) Appl Catal B
61:212
1 3