Metal and solvent-free oxidation of α-isophorone to ketoisophorone by molecular oxygen

Article abstract of DOI:10.1016/j.catcom.2010.02.010

The metal and solvent-free oxidation of α-isophorone to ketoisophorone (KIP) using N-hydroxy phthalimide (NHPI) as the catalyst under atmospheric oxygen was studied, eliminating the isomerization process of α-isophorone to β-isophorone. The effect of co-catalyst, temperature, time, amount of NHPI, and recycling on the oxidation was investigated. The oxidation of α-isophorone proceed well without any co-catalyst under 60 °C for 10 h, which was different with the traditional oxidation of α-isophorone at high temperature. Furthermore, NHPI can be easily separated and reused with a slight loss on its activity. This aerobic oxidation process by NHPI provides a good alternative way for the industrial synthesis of KIP.

Full text of DOI:10.1016/j.catcom.2010.02.010

Catalysis Communications 11 (2010) 758–762
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Metal and solvent-free oxidation of
molecular oxygen
a-isophorone to ketoisophorone by
*
Congmin Wang, Guoliang Wang, Jianyong Mao, Zhen Yao, Haoran Li
Department of Chemistry, Zhejiang University, Hangzhou 310027, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
The metal and solvent-free oxidation of
a
-isophorone to ketoisophorone (KIP) using N-hydroxy phthali-
Received 24 November 2009
Received in revised form 4 February 2010
Accepted 9 February 2010
mide (NHPI) as the catalyst under atmospheric oxygen was studied, eliminating the isomerization pro-
cess of -isophorone to b-isophorone. The effect of co-catalyst, temperature, time, amount of NHPI,
and recycling on the oxidation was investigated. The oxidation of -isophorone proceed well without
any co-catalyst under 60 °C for 10 h, which was different with the traditional oxidation of -isophorone
a
a
Available online 12 February 2010
a
at high temperature. Furthermore, NHPI can be easily separated and reused with a slight loss on its activ-
ity. This aerobic oxidation process by NHPI provides a good alternative way for the industrial synthesis of
KIP.
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
suffered the drawbacks such as relatively harsh conditions, metal-
lic toxicity, and high expense. Therefore, selective metal-free oxi-
Selective allylic oxidation of cyclic olefins is a fundamental and
important functional transformation in synthetic chemistry [1]. As
a typical allylic oxidation, the oxidation of 3,5,5-trimethyl-2-
dation of a-IP (1) to KIP (3) with molecular oxygen under
relatively mild condition becomes particularly desirable.
Recently, aerobic oxidation catalyzed by metal-free catalyst
such as N-hydroxyphthalimide (NHPI) was attractive because of
its high efficiency and relatively mild reaction condition [18–22].
It was found that NHPI combined with metal or organic co-catalyst
to be a remarkable catalytic system, which exhibited high activity
in the oxidation of hydrocarbons [22,23], and some olefins such as
cyclohexene and styrene [24–27]. However, to the best of our
knowledge, the metal and solvent-free aerobic oxidation of the ole-
cyclohexen-1-one (a-isophorone, a-IP, 1) to 3,5,5-trimethyl-2-
cyclohexen-1,4-dione (ketoisophorone, KIP, 3) has received consid-
erable attention (Scheme 1), because KIP is a key intermediate for
the synthesis of various carotenoids and flavoring fine chemicals
[2]. The conventional method for the production of KIP (3) in
industry is the homogeneous liquid oxidation of 3,5,5-trimethyl-
3-cyclohexen-1-one (b-isophorone, b-IP, 2), which is prepared by
the isomerization process of
zation temperature is very high, and the conversion of
IP (2) is no more than 2% [10]. The selective oxidation of
interesting route to prepare KIP, eliminating the isomerization step
of -IP (1) to b-IP (2).
Accordingly, a variety of catalytic systems including phospho-
a
-IP (1) [3–9]. However, the isomeri-
-IP (1) to b-
-IP is an
fins such as a-IP (1) and 2-cyclohexen-1-one, which is conjugated
with the carbonyl group, has not been reported.
a
a
Herein, we report a metal and solvent-free aerobic oxidation of
a-IP (1) to KIP (3) with molecular oxygen in the presence of NHPI.
a
The effect of co-catalyst, temperature, time, amount of NHPI, and
recycling on the conversion and selectivity was investigated. The
molybdic acid [11,12], Pd(OAc)2 [13], molybdovanadophosphate
[14,15], and metal supported MgAl hydrotalcite [16,17] have been
possible oxidation pathways of a-IP (1) catalyzed by NHPI were
estimated. This metal and solvent-free aerobic oxidation process
provides a good alternative way for the industrial synthesis of
KIP (3).
extensively investigated for the oxidation of
a-IP (1) to KIP (3). For
the selective oxidation of -IP (1) to KIP (3), molecular oxygen as
a
an oxidant is more convenient for both economical and environ-
mental benefits. Murphy and Baiker [11] described an efficient
2. Experimental section
and selective aerobic oxidation of
a-IP (1) to KIP (3) catalyzed by
homogeneous phosphomolybdic acid system. However, these oxi-
dation processes were mainly catalyzed by metallic compounds
contained molybdenum, vanadium, chromium, copper, etc., which
2.1. Materials
The commercial-grade
a-IP (1), b-IP (2), and KIP (3) were
supplied by Zhejiang NHU Co., Ltd. NHPI, benzoyl peroxide (BPO),
tetrabutylammonium bromide (TBAB), anthraquinone (AQ), Ethyl
benzoate, and dichloromethane were analytical reagents
* Corresponding author. Fax: +86 571 8795 1695.
E-mail address: lihr@zju.edu.cn (H. Li).
1566-7367/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.catcom.2010.02.010

C. Wang et al. / Catalysis Communications 11 (2010) 758–762
759
and selectivity were determined using GC equipped with column
m). Ethyl benzoate (99.9%) was
applied as internal standard for quantitative analysis.
O
O
ATSE-54 (50 m  0.53 mm  1
l
O2
3. Results and discussion
O
3
1
The co-catalysts are generally indispensable in the aerobic oxi-
dation of hydrocarbons catalyzed by NHPI as a function of initiator
[25,28–30]. Therefore, the aerobic oxidation of
O2
O
a-IP (1) catalyzed
by NHPI with different co-catalysts at 80 °C for 10 h was investi-
gated at first. It was seen in Table 1 that the selectivity and conver-
sion did not increase remarkably when different co-catalysts were
used. For example, the KIP (3) selectivity of 47.5%, 42.9%, and con-
version of 41.0%, 39.6%, were obtained when benzoyl peroxide
(BPO) and anthraquinone (AQ) were used as the co-catalysts, while
the KIP (3) selectivity of 48.3% and conversion of 39.0% were ob-
tained in the absence of co-catalyst. Hence, the results showed that
2
Scheme 1. Catalytic oxidation of isophorone to ketoisophorone with dioxygen.
purchased from commercial sources, and were all used as received
unless otherwise stated.
the co-catalyst was not necessary in the aerobic oxidation of a-IP
(1) catalyzed by NHPI. This feature may be related to the structure
of KIP (3), which is similar to that of quinone [31]. The study on the
reason why the oxidation did not need co-catalyst is in progress.
2.2. Experimental methods
Different with the traditional oxidation of
metallic catalyst, which carried out at high temperature (>100 °C)
because of the thermodynamic stability of a-IP (1) [12], the oxida-
a-IP (1) catalyzed by
In a typical oxidation procedure, a batch reactor, which com-
posed of a four-neck flask connected with a reflux condenser, gas
bubbler, mechanical stirrer, and a water bath equipped with a ther-
mostat, was used for catalytic reactions. A mixture of NHPI (0.8 g,
5 mmol) and
a-IP (13.6 g, 100 mmol) was placed in the flask and
Table 2
heated at 60 °C under vigorous stirring, then a stream of dioxygen
was conducted into the reaction solution for 10 h. After the reac-
tion finished, the reaction mixture cooled down to about 0 °C.
The resulting finely crystalline precipitate was removed by centri-
fuging to produce the recycled NHPI (0.68 g). The filtrate was evap-
orated, treated, and analyzed regularly to monitor the oxidation
reaction by gas chromatography. The oxidation products and inter-
mediates were identified by HP5973 GC–MS with DP17 column
The effect of temperature on the oxidation of a-isophorone catalyzed by NHPI.
Entry
Temperature (°C)
Conversion (%)
Selectivity (%)
KIP
HIP
1
2
3
4
5
80
70
60
50
40
39.0
26.1
15.4
10.4
4.1
48.3
56.0
77.2
53.2
45.5
5.5
5.4
4.5
4.2
3.7
(30 m  0.25 mm  0.25
lm) by comparing retention times and
fragmentation patterns with authentic samples. The conversion
Reaction conditions: 100 mmol
a-IP, 5 mmol NHPI, 10 h.
Table 1
Table 3
The effect of different catalyst on the oxidation of
a-isophorone.
The effect of time and the amount of NHPI on the oxidation of a-isophorone.
Entry
Catalyst
Conversion (%) Selectivity (%)
Entry
Time (h)
NHPI (mol%)
Conversion (%)
Selectivity (%)
KIP
HIP
KIP
HIP
1
2
3
4
5
NHPI/BPO
NHPI/TBAB
NHPI/AQ
NHPI
41.0
20.0
39.6
39.0
Trace
47.5
46.2
42.9
48.3
–
4.8
4.0
4.5
5.5
–
1
2
2
3
4
5
6
7
8
5
5
5
5
5
5
3
4
6
7
10.9
15.4
17.3
24.8
29.2
10.1
11.9
15.8
16.1
81.7
77.2
68.0
57.7
51.8
70.0
75.7
67.4
65.8
4.9
4.5
4.0
3.4
3.0
4.2
4.2
4.4
4.6
10
15
20
25
10
10
10
10
–
Reaction conditions: 100 mmol
10 h.
BPO: benzoyl peroxide.
TBAB: tetrabutylammonium bromide.
AQ: anthraquinone.
a-IP, 5 mmol NHPI, 0.3 mmol co-catalyst, 80 °C,
Reaction conditions: 100 mmol a-IP, 60 °C.
+
O
O
O
O
O
O
O
O
NHPI
O2
O
+
+
+
+
OOH
O
OH
OOH
O
3
5
6
1
4
7
Scheme 2. The oxidation of a-isophorone catalyzed by NHPI with dioxygen.

760
C. Wang et al. / Catalysis Communications 11 (2010) 758–762
tion catalyzed by NHPI proceed well under relatively mild temper-
ature (Scheme 2). It was clear in Table 2 that the effect of temper-
ature on the oxidation of -IP (1) was significant. When the
Recycled
a
temperature of the oxidation was higher than 60 °C, the selectivity
would decrease remarkably because of the formation of the unde-
sired byproducts such as dimer and polymers, which were de-
tected by ESI-MS. On the other hand, the KIP (3) selectivity
would reduce from 77.2% to 45.5% when the temperature of the
oxidation decreased from 60 °C to 40 °C because the active inter-
mediate PINO was not easy to produce at low temperature [32].
In contrast, the desired KIP (3) was obtained with high selectivity
of 77.2% at 60 °C. An important feature of the oxidation is that
there is no evidence for the formation of over-oxidation products
such as formyisophorone, which may be related to the low temper-
ature [14]. During the oxidation, besides the main product KIP (3),
4-hydroxy-3,5,5-trimethyl-2-cyclohexen-1-one (HIP) (5) of about
4% and 4,4,6-trimethyl-7-oxobicyclo [4.1.0]-heptan-2-one (OIP)
(6) of about 2% were also produced, which were identified by
GC–MS.
Fresh
4000 3500 3000 2500 2000 1500 1000 500
Wavelength (cm-1)
Fig. 1. The comparison of IR spectrum between the fresh NHPI and recycled NHPI.
, fresh NHPI; , recycled NHPI.
It should be noted that the time had strong effect on the oxida-
tion, which was shown in Table 3. The conversion of
a-IP (1) in-
Table 4
creased from 10.9% to 24.8%, and selectivity of KIP (3) decreased
continuously from 81.7% to 57.7% with the increasing of the time
from 4 h to 20 h. Selectivity of KIP (3) decreased remarkably when
the time of the oxidation was longer than 15 h because of the for-
The effect of reused NHPI on the oxidation of a-isophorone.
Entry
Recycling
Conversion (%)
Selectivity (%)
KIP
HIP
mation of the byproducts. Considering the separation of
a-IP (1)
1a
2
3
4
5
0
1
2
3
4
15.4
14.2
14.7
14.5
14.1
77.2
69.0
67.0
70.0
69.6
4.5
4.2
4.4
4.3
4.0
and KIP (3) is easy, 10 h might be the best for the oxidation cata-
lyzed by NHPI at 60 °C. In addition, we also studied the influence
of amount of NHPI on the oxidation. It can be seen in Table 3 that
the conversion and selectivity increased when amount of NHPI en-
hanced from 3 mol% to 5 mol%. On the other hand, the conversion
changed very little, but the selectivity decreased remarkably as the
Reaction conditions: 100 mmol
a
-IP, 60 °C, 10 h.
a
NHPI (5 mmol).
O
O
NOH
O
.
NO
O
NHPI
PINO
O
O
O
O
O
O
PINO
O2
NHPI
+
.
.
OO
OOH
OH
O
8
1
9
4
3
5
+1
O
O
O
O
O
O2
O
O
NHPI
O
O
O
O
O
+
.
OO
O
.
OOH
OH
O
O
6
5
10
11
7
Scheme 3. Proposed reaction pathways for a-isophorone oxidation with dioxygen in the presence of NHPI.

C. Wang et al. / Catalysis Communications 11 (2010) 758–762
761
Table 5
The comparison of different allylic oxidation in the presence of NHPI.
Entry
Substance
Time (h)
Conversion (%)
Selectivity (%)
Ketone
Alcohol
Epoxy
Other
1
a
b-IP
Cyclohexene
Cyclohexenone
-IP
10
10
5
15.4
45.6
64.8
4.4
77.2
91.1
55.4
46.4
4.5
1.4
9.7
6.8
2.6
Trace
3.6
15.7
1.0
31.3
41.4
2a
3
4b
10
5.4
Reaction conditions: 100 mmol substrate, NHPI 5.0 mmol, 60 °C.
a
Pyridine (40 ml) was used as solvent.
BPO (0.3 mmol) was added as co-catalyst.
b
amount of NHPI increases up to 7 mol%. Therefore, 5 mol% is con-
sidered as a preferable amount of NHPI.
gated with carbonyl group, which is the reason why the selectivity
of KIP (3) in the oxidation of
of the time [37].
a-IP (1) decreased with the increasing
The separation of NHPI and KIP (3) is very easy because NHPI is
immiscible with a-IP (1) and KIP (3) at low temperature. When the
oxidation finished, after cooling down the reaction mixture to 0 °C,
NHPI was separated from the organic phase by centrifuging, which
could be used in the subsequent reaction. The structure of recycled
NHPI was confirmed using IR spectra, 1H NMR, and ESI-MS, which
was shown in Fig. 1. It was clear that there was not obvious differ-
ence between the recycled NHPI and the fresh NHPI. In order to ac-
cess the influence of the recycled NHPI on the oxidation, five
recycling runs were carried out under the same condition. It was
seen in Table 4 that the NHPI could be recycled with a slight loss
of its activity. It seemed that the NHPI could have the potential
to be used more than five times.
4. Conclusion
In summary,
a-IP (1) can be directly oxidized to KIP (3) with
atmospheric oxygen by NHPI. The effect of co-catalyst on the oxi-
dation was investigated and the results showed that the co-cata-
lyst was not necessary in the aerobic oxidation of
catalyzed by NHPI. Especially, the oxidation of -IP (1) catalyzed
by NHPI was carried out well under 60 °C for 10 h, while the tradi-
tional oxidation of -IP (1) was usually at high temperature. Fur-
a-IP (1)
a
a
thermore, the separation and recycling of NHPI were easy with a
slight loss on its activity. Possible reaction pathways for the oxida-
Based on previous reports and the observed reaction products,
tion of
a-IP (1) were presented, which indicated that the byprod-
the possible oxidation pathways of
a-IP catalyzed by NHPI could
ucts such as dimer and polymers produced easily when the
double bond is conjugated with carbonyl group. Considering the
advantage such as aerobic oxidation, metal and solvent-free pro-
cess, this oxidation catalyzed by NHPI provided a potentially new
way for the synthesis of KIP (3) in industry.
be postulated, which was shown in Scheme 3. Firstly, NHPI can
be converted to phthalimide N-oxylradical (PINO). The hydrogen
atom of
(8), which further led to the formation of
and the corresponding -IP hydroperoxide (4). Then
a
-IP (1) was abstracted by PINO to produce
-IP peroxy radical (9)
-IP hydro-
a-IP radical
a
a
a
peroxide (4) was directly decomposed to KIP (3) and HIP (5). It is
well known that alkyl hydroperoxide (4) is the initial intermediate
Acknowledgement
in the oxidation of hydrocarbon [33–35]. In the oxidation of
KIP catalyzed by NHPI, it was confirmed by the addition of tri-
phenyl phosphate (TPP) because -IP hydroperoxide (4) could be
quantitatively reduced to the HIP (5) by TPP at room temperature
[34,35]. On the other hand, -IP peroxy radical (9) can add to the
double bond in -IP to produce the more stable b-peroxy radical
(10) [36]. The b-peroxy radical (10) reacted with O2 to give the
new peroxy radical (11) and further converted to the dimer hydro-
peroxide (7) and polymer, which was detected by ESI-MS. It is re-
ported that the greater the stability of b-peroxy alkyl radical, the
more likely it is to react with oxygen rather than to undergo uni-
molecular decomposition to produce OIP (6) [37]. Thus the dimer
hydroperoxide (7) and polymer are main products, and OIP (6) is
a-IP to
This work was supported by National Natural Science Founda-
tion of China (Nos. 20990221, 20704035).
a
References
a
a
[1] E.F. Murphy, T. Mallat, A. Baiker, Catal. Today 57 (2000) 115.
[2] H. Ernst, Pure Appl. Chem. 74 (2002) 2213.
[3] J.Y. Mao, N. Li, H.R. Li, X.B. Hu, J. Mol. Catal. A: Chem. 258 (2006) 178.
[4] S.B. Halligudi, N.K.K. Raj, S.S. Deshpande, S. Gopinathan, J. Mol. Catal. A: Chem.
157 (2000) 9.
[5] N. Ito, T. Etoh, H. Hagiwara, M. Kato, Synthesis (1997) 153.
[6] M. Beyrhouty, A.B. Sorokin, S. Daniele, L.G. Hubert-Pfalzgraf, New J. Chem. 29
(2005) 1245.
[7] M. Costantini, A. Dromard, M. Jouffret, B. Brossard, J. Varagnat, J. Mol. Catal. 7
(1980) 89.
[8] J.Y. Mao, X.B. Hu, H. Li, Y. Sun, C.M. Wang, Z.R. Chen, Green Chem. 10 (2008)
827.
trace in the oxidation of a-IP (1).
For further verifying of the mechanism of catalytic oxidation,
the oxidation of b-IP (2), cyclohexene, and cyclohexenone cata-
lyzed by NHPI was also carried out, which was given in Table 5.
It was clear that the selectivity of KIP (3) was more than 90%,
and the other byproducts such as polymer were only about 1%
when b-IP (2) was oxidized in the same condition, while the selec-
tivity of KIP (3) was only 77.2%, and the other byproducts such as
[9] X.B. Hu, J.Y. Mao, Y. Sun, H. Chen, H.R. Li, Catal. Commun. 10 (2009) 1908.
[10] C. Beck, T. Mallat, A. Baiker, J. Catal. 195 (2000) 79.
[11] E.F. Murphy, A. Baiker, J. Mol. Catal. A: Chem. 179 (2002) 233.
[12] E.F. Murphy, M. Schneider, T. Mallat, A. Baiker, Synthesis (2001) 547.
[13] T. Joseph, S.B. Halligudi, C. Satyanarayan, D.P. Sawant, S. Gopinathan, J. Mol.
Catal. A: Chem. 168 (2001) 87.
[14] E.F. Murphy, T. Mallat, A. Baiker, M. Schneider, Appl. Catal., A 197 (2000) 295.
[15] A. Hanyu, Y. Sakurai, S. Fujibayashi, S. Sakaguchi, Y. Ishii, Tetrahedron Lett. 38
(1997) 5659.
[16] D. Kishore, A.E. Rodrigues, Catal. Commun. 8 (2007) 1156.
[17] D. Kishore, A.E. Rodrigues, Appl. Catal., A 345 (2008) 104.
[18] F. Recupero, C. Punta, Chem. Rev. 107 (2007) 3800.
[19] F. Rajabi, B. Karimi, J. Mol. Catal. A: Chem. 232 (2005) 95.
[20] B. Karimi, J. Rajabi, Synthesis (2003) 2373.
[21] Y. Ishii, S. Sakaguchi, T. Iwahama, Adv. Synth. Catal. 343 (2001) 393.
[22] Y. Ishii, T. Iwahama, S. Sakaguchi, K. Nakayama, Y. Nishiyama, J. Org. Chem. 61
(1996) 4520.
polymer were more than 15% for
a-IP (1). The difference between
a-IP (1) and b-IP (2) is the position of carbon–carbon double bond.
The former is conjugated with carbonyl group, but the latter is not.
Similarly, the higher selectivity of ketone and the lower other
byproducts were obtained for cyclohexene, compared with that
of cyclohexenone. Therefore, the results showed that the addition
which resulted in the dimer and polymer is favored over the
abstraction of an allylic hydrogen when the double bond is conju-
[23] R.A. Sheldon, I. Arends, Adv. Synth. Catal. 346 (2004) 1051.
[24] R.A. Sheldon, I. Arends, A. Dijksman, Catal. Today 57 (2000) 157.

762
C. Wang et al. / Catalysis Communications 11 (2010) 758–762
[25] X.L. Tong, J. Xu, H. Miao, G.Y. Yang, H. Ma, Q.H. Zhang, Tetrahedron 63 (2007)
7634.
[32] Q.H. Zhang, C. Chen, H. Ma, H. Miao, W. Zhang, Z.Q. Sun, J. Xu, J. Chem. Technol.
Biotechnol. 83 (2008) 1364.
[26] S.M. Silvestre, J.A.R. Salvador, Tetrahedron 63 (2007) 2439.
[27] C.M. Wang, W.J. Zhao, H.R. Li, L.P. Guo, Green Chem. 11 (2009) 843.
[28] X.L. Tong, J. Xu, H. Miao, J. Gao, Tetrahedron Lett. 47 (2006) 1763.
[29] G.Y. Yang, Q.H. Zhang, H. Miao, X.L. Tong, J. Xu, Org. Lett. 7 (2005) 263.
[30] X.L. Tong, J. Xu, H. Miao, Adv. Synth. Catal. 347 (2005) 1953.
[31] Z.T. Du, Z.Q. Sun, W. Zhang, H. Miao, H. Ma, J. Xu, Tetrahedron Lett. 50 (2009)
1677.
[33] G.Y. Yang, Y.F. Ma, J. Xu, J. Am. Chem. Soc. 126 (2004) 10542.
[34] S. Evans, J.R.L. Smith, J. Chem. Soc., Perkin Trans. 2 (2000) 1541.
[35] S. Evans, J.R.L. Smith, J. Chem. Soc., Perkin Trans. 2 (2001) 174.
[36] F.R. Mayo, J. Am. Chem. Soc. 80 (1958) 2497.
[37] K.U. Ingold, Acc. Chem. Res. 2 (1969) 1.