Polyethylene glycol radical-initiated benzylic C-H bond oxygenation in compressed carbon dioxide

Article abstract of DOI:10.1039/b908993d

The PEG radical originating from the thermal/oxidative degradation of PEG in dense CO₂ was successfully applied to the oxygenation of benzylic hydrocarbons under organic solvent-free conditions. In addition, in our study, dense CO₂ could improve the oxygenation reaction.

Full text of DOI:10.1039/b908993d

View Article Online / Journal Homepage / Table of Contents for this issue
LETTER
www.rsc.org/njc | New Journal of Chemistry
Polyethylene glycol radical-initiated benzylic C–H bond oxygenation
in compressed carbon dioxidew
Jin-Quan Wang and Liang-Nian He*
Received (in Montpellier, Frnace) 6th May 2009, Accepted 15th June 2009
First published as an Advance Article on the web 25th June 2009
DOI: 10.1039/b908993d
The PEG radical originating from the thermal/oxidative
degradation of PEG in dense CO was successfully applied
to the oxygenation of benzylic hydrocarbons under organic
2
dense CO was also successfully applied to the oxygenation of
2
benzylic hydrocarbons under organic solvent-free conditions.
We first chose the oxygenation of ethyl benzene to investigate
this PEG radical chemistry, and carried out a reaction
with 0.7 mmol PEG-1000 (molecular weight: 1000 Da) and
solvent-free conditions. In addition, in our study, dense CO
could improve the oxygenation reaction.
2
2
.5 MPa O at 100 1C for 12 h. Under these conditions, no
2
The selective catalytic oxygenation of hydrocarbons with O
2
oxidation product was detected (Table 1, entry 1). However, a
as the terminal oxidant is a fundamental industrial technology,
and its success largely depends on the use of metal catalysts to
promote both the rate of reaction and selectivity to partial
catalytic amount of Co(OAc) , remarkably, could promote the
2
reaction, and thus a 15% yield of acetophenone with excellent
selectivity was achieved (Table 1, entry 2), concomitant with
1-phenylethanol (2%). The reaction was further improved by
1
oxidation products. The Wacker oxidation of alkenes to
2
carbonyl compounds catalyzed by palladium salts, the
adding 7.5 MPa CO
unusual ‘‘expanded effect’’ of PEG in dense CO
lowered melting points, raised diffusion rates and lowered
2
(Table 1, entry 3), presumably due to the
production of ethylene oxide from ethylene in the presence
3
of silver salts, and terephthalic acid from p-xylene catalyzed
4
by manganese and cobalt salts are representative commercial
2
, due to the
7,10
solvent viscosity.
through a window-equipped high-pressure reactor, we found
that PEG and its derivatives are expandable with CO
In addition, a lower temperature led to a decrease in
yield (Table 1, entry 5). Moreover, PEG and O (Table 1,
Indeed, as judged by visual inspection
processes. Recently, Ishii, Recupero, Sheldon and others have
developed an elegant and effective system for C–H bond
oxygenation, even under mild conditions, by hydrogen
.
2
5
abstraction from N-hydroxyphthalimide (NHPI). However,
2
in order to maintain the solubility of NHPI and activity of the
catalytic system, the oxygenation reaction is generally carried
entries 4 and 6) were found to be indispensable for this
transformation. These facts indicate that PEG oxidative/thermal
degradation plays a critical role in the oxygenation of ethyl
out in CH COOH or toxic PhCN. Despite the usefulness of
3
this method for the oxidation of C–H bonds, the development
of new and environmentally more benign catalytic systems for
2 2
benzene to acetophenone, and that ‘‘PEG/O /CO ’’ possibly
serves as a radical initiator, oxidant and solvent, which makes
this methodology efficient and more environmentally friendly.
Note that metal salts could promote this radical reaction.
We next screened commercially available metals salts (Table 1,
entries 7–13). The results show that Co and Mn salts, which
6
oxidation is a sustained goal in organic synthesis.
Polyethylene glycol (PEG) and its derivatives are commonly
known to be inexpensive, thermally stable, have an almost
negligible vapor pressure, and be toxicologically innocuous
and environmentally benign media for chemical reactions and
5
a
could accelerate the generation of free radicals, generally
improved the yield of acetophenone (Table 1, entries 7–9),
while the other salts, such as of Cu, Ni and Pd, demonstrated
no positive effect on the oxygenation reaction (Table 1, entries
10–13). Moreover, acetates showed generally higher activities
than halides in this study (Table 1, entries 3, 7–9). In order to
understand this process in more depth, parallel experiments
7
phase transfer catalysts. We have also developed several
methodologies for organic transformations using PEG as a
7
d–g
solvent or support for a catalyst.
On the other hand, the
8
oxidative thermal degradation of PEG was also encountered
because PEG is susceptible to free radical oxidative attack in
the presence of O at elevated temperatures over 70 1C. In this
2
respect, PEG and O
2
can react to form PEG peroxide in excess
were conducted with and without Co(OAc) in the oxygenation
2
air through a so-called random chain scission process.
Recently, we reported that the PEG radicals from oxidative
degradation could be used to induce the oxidation of benzylic
of ethyl benzene, and the spent PEG was detected by ESI-MS
techniques (Fig. S1 in the ESIw). The results hint that
Co(OAc)2 indeed accelerates the oxidative degradation of
9
1
2
alcohols. We would like to disclose here that the PEG radical
PEG, as previously reported, and that it was also beneficial
for the oxygenation reaction (Table 1, entries 1 and 2). Based
originating from the thermal/oxidative degradation of PEG in
2
on these results, Co(OAc) in this study can be assumed to
have acted as an initiator for generating PEG radicals and to
have promoted the decomposition of intermediate hydro-
State Key Laboratory of Elemento-Organic Chemistry, Nankai
University, Tianjin 300071, P. R. China. E-mail: heln@nankai.edu.cn;
Fax: +86 22 23504216; Tel: +86 22 23504216
w Electronic supplementary information (ESI) available: General
information, EPR spectrum, GC chart, ESI-MS spectrum and NMR
charts. See DOI: 10.1039/b908993d
peroxides into the product, as similarly reported in radical
5
reactions catalyzed by NHPI in the presence of Co(OAc) .
2
Moreover, when the reaction was carried out at an elevated
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1637–1640 | 1637

View Article Online
a
Table 1 Screening reaction conditions for the oxidation of ethyl benzene
b
Entry
Metal salt
None
T/1C
Conversion (%)
Yield of acetophenone (%)
c
1
2
100
100
100
100
80
100
100
100
100
100
100
100
100
100
120
0
0
c
Co(OAc)
Co(OAc)
Co(OAc)
Co(OAc)
Co(OAc)
2
2
2
2
2
17
24
0
16
0
12
12
14
0
0
0
5
3
66
15
22
0
15
0
11
11
13
0
0
0
5
0
65
3
4
d
5
6
e
7
8
9
CoCl
MnCl
Mn(OAc)
CuCl
2
Á6H
2
O
2
Á4H
2
O
2
Á4H
2
O
10
11
12
13
14
15
CuCl
PdCl
NiCl
Co(OAc)
Co(OAc)
2
Á2H
2
O
2
2
Á6H
2
O
f
2
2
a
b
, 12 h. Yield based on GC-MS,
Reaction conditions: 1.93 mmol ethyl benzene, 2% metal salts, 0.7 mmol PEG-1000, 2.5 MPa O
d
2
, 7.5 MPa CO
2
c
using biphenyl as an internal standard. Without CO
11 f
industrial methods employing simple cobalt salts as homogeneous catalysts.
e
2
.
In the absence of O
2
.
CH
3
COOH was used in place of PEG, as in the principal
In the presence of TEMPO (0.193 mmol).
temperature (120 1C), the yield of the desired product could be
further improved up to 65% (Table 1, entry 15). However, in
order to differentiate the activity of the substrates and to
validate whether the oxidation could be performed under
milder conditions, 100 1C was selected as the reaction
temperature for further investigations. On the other hand,
the oxygenation reaction was completely suppressed by
TEMPO (2,2,6,6-tetramethyl-piperidine-1-oxyl), which supports
the free radical process (Table 1, entry 14).
The generality of this methodology was evaluated under
typical reaction conditions, as shown in Table 3. A variety of
benzylic C–H bonds could be oxygenized into their corres-
ponding oxygenated products (Table 3). Substrates favorable
for forming relatively stable radicals showed a high reactivity,
and selectively gave carbonyl compounds as major products.
In particular, fluorene, 1,2,3,4-tetrahydronaphthalene and
acenaphthene furnished moderate to high yields (Table 3,
entries 1–3), possibly due to their relatively active C–H bonds
being affected by ring strain and more stable radicals being
formed in situ. Ethyl benzene, cumene and butyl benzene could
also be oxidized with a high selectivity, although with a low
reactivity so far (Table 3, entries 4–6). Even toluene and its
derivatives successfully underwent benzylic C–H oxidations to
generate the corresponding carboxylic acids in reasonable
yields, alongside aldehydes as major by-products (Table 3,
entries 7–9). We were pleased to find that this method was
applicable even to cycloalkanes; for example, cyclohexane
gave an 8% conversion to afford 4% cyclohexanone and 3%
cyclohexanol (Table 3, entry 10). Nevertheless, the current
industrial process shows just 3–7% conversion of cyclohexane
Since the oxygenation might be initiated by PEG radicals
from the oxidative degradation of PEG, we next evaluated the
effect of the molecular weight of PEG on the oxygenation of
ethyl benzene (Table 2). Indeed, molecular weights from
300–6000 gave a similar yield under otherwise identical
conditions. However, PEG-20 000 was inactive (Table 2,
entry 5), presumably due to the increased mass transport
1
3
limitation of gaseous O
2
in highly viscous long chain PEGs.
9
Notably, the effect of the molecular weight of the PEG
9
Table 2) and the influence of the quantity of PEG on the
(
reaction support the notion that the degraded PEG accumulates
and cannot interfere with the reaction.
1
4
at 160 1C.
Based on aforementioned findings, a plausible mechanism
Table 2 The oxidation of ethyl benzene in different molecular weight
PEGs
a
for this protocol may be considered to involve a PEG radical
8b
b
formed in situ during the thermal degradation of PEG.
A
Entry
PEG
Conversion (%)
Yield (%)
mechanistic scheme is therefore proposed, as delineated in
1
2
3
4
5
300
600
1 000
6 000
20 000
23
23
24
25
0
21
21
22
23
0
Scheme 1. The reaction of PEG with O is prone to generate
2
PEG radicals, which induce the C–H bond oxygenation. The
2
radical process could also be promoted by Co(OAc) . The
proposed radical mechanism is supported by evidence, including
the production of trace amounts of formylated products of the
degraded PEG (Fig. S1), a broadening PEG molecule weight
distribution (Fig. S2 in the ESIw), the generating of a peroxide
intermediate, as detected by a KI/starch test, detected radical
a
2
Reaction conditions: 1.93 mmol ethyl benzene, 2% mmol Co(OAc) ,
b
0
.7 mmol PEG-1000, 2.5 MPa O
were determined by GC-MS with respect to an internal standard
biphenyl).
2 2
, 7.5 MPa CO , 100 1C, 12 h. Yields
(
1
638 | New J. Chem., 2009, 33, 1637–1640
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009

View Article Online
Table 3 The generality of PEG radical-initiated benzylic C–H bond
oxygenation
chemistry of PEG thermal degradation will become an even
more powerful tool for organic synthesis.
a
b
We are grateful for the financial support from the National
Natural Science of Foundation of China (No. 20421202,
Entry
1
Substrate
Product
Conversion (%) Yield (%)
2
0672054, 20872073), the 111 Project (B06005) and the
100
83
Committee of Science and Technology of Tianjin.
2
3
65
86
58
73
Experimental
Safety warning
Experiments using large amounts of compressed gas, especially
molecular O
and must only be carried out using the appropriate equipment
under rigorous safety precautions. In particular, CO is
2
and supercritical fluids, are potentially hazardous
4
5
6
24
14
14
20
13
12
25
2
introduced into the substrate-loaded reactor before O2 is
added. In other words, the following order should be used:
O after CO after substrate to avoid an explosive regime; in
2
2
addition, there could be peroxides in the present reaction
system. The exact level of risk associated with the handling
of oxidized PEG is currently unknown.
d
d
Experimental procedure
7
8
35
38
All experiments were carried out with substrate (1.93 mmol),
d
PEG-1000 (0.7 g, 0.7 mmol) and Co(OAc)
5 mL autoclave equipped with an inner glass tube. 3 MPa
CO and 2.5 MPa O were introduced into the autoclave and
2
(10 mg, 3%) in a
28
20
2
2
2
c
d
9
24
8
heated to the reaction temperature. The final pressure was then
adjusted to the desired pressure at the reaction temperature by
d
4
1
0
introducing an amount of CO . The mixture was stirred
2
continuously for the designed reaction time. Then, the reactor
a
Reaction conditions: 1.93 mmol substrate, 2% mmol Co(OAc)
2
,
2
was placed into ice water, and CO was released slowly by
0
b
.7 mmol PEG-1000, 2.5 MPa O , 7.5 MPa CO , 100 1C, 12 h.
2
2
c
d
passing it through a cold trap containing diethyl ether to
absorb the trace amount of reactant and product entrained
2
Isolated yield based on substrate. 3 MPa O , 4 h. Determined
by GC-MS.
2
by the CO . After depressurization, diethyl ether was added
into the reactor. The products were then extracted by diethyl
ether and analyzed by a gas chromatograph (Shimadzu-2014)
equipped with a capillary column (HP-5 30 m  0.25 mm)
using a flame ionization detector. The structure and purity
of the products were further identified using NMR
(Bruker-300 MHz, Varian-400 MHz), GC-MS (HP G1800A),
HPLC-MS (LCQ Advantage) and GC-HPLC by comparing
retention times and fragmentation patterns with those of
authentic samples. The EPR experiments were performed on
a Bruker EMX-6 instrument.
Scheme 1 The proposed mechanism.
species from electron paramagnetic resonance experiments
Fig. S3 in the ESIw), and the suppressing effect of TEMPO
on the reactions (Table 1, entry 14).
References
(
1
(a) E. N. Jacobson, Adv. Synth. Catal., 2004, 346, 109;
b) T. Punniyamurthy, S. Velusumy and J. Iqbal, Chem. Rev.,
005, 105, 2329.
(a) J. Smidt, W. Hafner, R. Jira, J. Sedlmeier, R. Sieber,
R. Ruttinger and H. Kojer, Angew. Chem., 1959, 71, 176;
b) J. Smidt, W. Hafner, R. Jira, R. Sieber, J. Sedlmeier and
(
2
In conclusion, the PEG radical, from the oxidative
degradation of PEG, has been successfully used to oxygenate
benzylic hydrocarbons in a mild and environmentally-friendly
2
¨
(
2
manner; dense CO further improved the oxidation reaction.
A. Sabel, Angew. Chem., Int. Ed. Engl., 1962, 1, 80.
(a) T. E. Lefort, Fr. Pat., 729 952, 1931; (b) T. E. Lefort,
U.S. Pat. 1 998 878, 1935.
These findings provide a useful example to further develop
free radical chemistry starting from PEG thermal oxidative
degradation. Further broadening of this protocol to other
organic reactions that require radical chemistry and the
possibility of PEG recycling are currently being investigated.
Given further time for in-depth investigations, the free radical
3
4 T. Yoshioka, Petrotech (Tokyo, Jpn), 1978, 1, 932.
5
For reviews and an article, see: (a) Y. Ishii, S. Sakaguchi and
T. Iwahama, Adv. Synth. Catal., 2001, 343, 395; (b) F. Recupero
and C. Punta, Chem. Rev., 2007, 107, 3800; (c) R. A. Sheldon and
I. W. C. E. Arends, Adv. Synth. Catal., 2004, 346, 1051;
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009 New J. Chem., 2009, 33, 1637–1640 | 1639

View Article Online
(
1
d) G. Yang, Y. Ma and J. Xu, J. Am. Chem. Soc., 2004, 126,
0542.
(a) J. M. Thomas and R. Raja, Catal. Today, 2006, 117, 22;
b) S. Stahl, Angew. Chem., Int. Ed., 2004, 43, 3400;
Angew. Chem., 1958, 709, 251; (d) J. Glastrup, Polym. Degrad.
Stab., 1996, 52, 217; (e) J. R. Conder, N. A. Fruitwala and
M. K. Shingari, J. Chromatogr., 1983, 269, 171;
(f) E. A. Altwicker, Chem. Rev., 1967, 67, 475.
9 J.-Q. Wang, L.-N. He and C.-X. Miao, Green Chem., 2009, 11,
1013.
6
(
´
(
1
c) J. Wo
¨
999, 5, 1460; (d) J. Zhu, A. Robertson and S. C. Tsang, Chem.
ltinger, J. E. Backvall and A. Zsigmond, Chem.–Eur. J.,
¨
Commun., 2002, 2044; (e) R. Neumann, A. M. Khenkin and
I. Vigdergauz, Chem.–Eur. J., 2000, 6, 875; (f) F. Loeker and
W. Leitner, Chem.–Eur. J., 2000, 6, 2011; (g) R. Bal, M. Tada,
T. Sadaki and Y. Iwasawa, Angew. Chem., Int. Ed., 2006, 45, 448;
10 For reviews and representative examples, see: (a) Chemical Synthesis
Using Supercritical Fluids, ed. P. G. Jessop and W. Leitner,
Wiley-VCH, Weinheim, 1999; (b) E. J. Beckman, J. Supercrit.
Fluids, 2004, 28, 121; (c) A. Baiker, Chem. Rev., 1999, 99, 453;
(d) G. Musie, M. Wei, B. Subramaniam and D. H. Busch, Coord.
Chem. Rev., 2001, 219–221, 789; (e) T. Sakakura, J. Choi and
H. Yasuda, Chem. Rev., 2007, 107, 2365; (f) P. G. Jessop and
B. Subramaniam, Chem. Rev., 2007, 107, 2666.
(h) F. Wang, J. Xu, X. Q. Li, J. Gao, L. P. Zhou and R. Ohnishi, Adv.
Synth. Catal., 2005, 347, 1987; (i) M. Christmann, Angew. Chem., Int.
Ed., 2008, 47, 2740; (j) X. B. Li and E. Iglesia, Angew. Chem., Int. Ed.,
2007, 46, 8649; (k) C. Limberg, Angew. Chem., Int. Ed., 2005, 44, 6102.
7
(a) J. Chen, S. K. Spear, J. G. Huddleston and R. D. Rogers, Green
Chem., 2005, 7, 64; (b) D. J. Heldebrant and P. G. Jessop, J. Am.
Chem. Soc., 2003, 125, 5600; (c) Z. S. Hou, N. Theyssen,
A. Brindmann and W. Leitner, Angew. Chem., Int. Ed., 2005, 44,
11 For examples, see: (a) M. P. Suresh and G. K. Bub, Ind. Eng.
Chem. Prod. Res. Dev., 1981, 20, 481; (b) W. Partenheimer, Chem.
Ind., 1990, 40, 321.
12 For representative examples, see: (a) A. Daminska, H. Kaczmarek
´
1
2
346; (d) Y. Du, Y. Wu, A.-H. Liu and L.-N. He, J. Org. Chem.,
008, 73, 4709; (e) J.-Q. Wang, F. Cai, E. Wang and L.-N. He,
and J. Kowalonek, Polymer, 1999, 40, 5781; (b) A. M. Sakharov,
L. I. Mazaletskaya and I. P. Skibida, Kinet. Katal., 2001, 42, 730;
(c) B. Q. Chen, J. R. Evans and S. Holding, J. Appl. Polym. Sci.,
2004, 94, 548.
Green Chem., 2007, 9, 882; (f) Y. Du, J.-Q. Wang, J.-Y. Chen,
F. Cai, J.-S. Tian, D. L. Kong and L. N. He, Tetrahedron Lett.,
2006, 47, 1271; (g) X.-Y. Dou, J.-Q. Wang, Y. Du, E. Wang and
L.-N. He, Synlett, 2007, 3058.
For typical examples, see: (a) E. Bortel, S. Hodorowicz and
R. Lamot, Makromol. Chem., 1979, 180, 2491; (b) S. Han,
C. Kim and D. Kwon, Polymer, 1997, 38, 317; (c) A. Riecke,
13 T. Seki, J.-D. Grunwaldt and A. Baiker, Chem. Commun., 2007,
3562.
14 U. Schuchardt, D. Cardoso, R. Sercheli, R. Pereira, R. S. da Cruz,
8
´
M. C. Guerreiro, D. Mandelli, E. V. Spinace and E. L. Pires, Appl.
Catal., A, 2001, 211, 1.
1
640 | New J. Chem., 2009, 33, 1637–1640
This journal is ꢀc The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2009