Baeyer-Villiger oxidation in compressed CO2

Article abstract of DOI:10.1039/b202727e

The aerobic Baeyer-Villiger oxidation of a wide range of ketones, both cyclic and acyclic to the corresponding esters or lactones can be efficiently carried out in compressed carbon dioxide in the presence of an aldehyde as co-reductant.

Full text of DOI:10.1039/b202727e

Baeyer–Villiger oxidation in compressed CO2†
Carsten Bolm,*a Chiara Palazzi,a Giancarlo Franciòc and Walter Leitner*bc
a Institut für Organische Chemie der RWTH Aachen, Professor Pirlet-Str. 1, D-52056, Aachen, Germany.
E-mail: Carsten.Bolm@oc.rwth-aachen.de; Fax: +49 (0)241 8092391
b Institut für Technische Chemie und Makromolekulare Chemie der RWTH Aachen, Worringer Weg 1,
D-52056, Aachen, Germany. E-mail: leitner@itmc.rwth-aachen.de; Fax: +49 (0)241 8022177
c
Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, D-45470, Mülheim an der Ruhr,
Germany
Received (in Cambridge, UK) 19th March 2002, Accepted 10th June 2002
First published as an Advance Article on the web 24th June 2002
The aerobic Baeyer–Villiger oxidation of a wide range of
ketones, both cyclic and acyclic to the corresponding esters
or lactones can be efficiently carried out in compressed
carbon dioxide in the presence of an aldehyde as co-
reductant
compressed carbon dioxide as the reaction medium. The
oxidation occurs without the need for a catalyst and converts the
starting ketones to their corresponding lactones efficiently and
with high selectivity. Conversion in CO2 is considerably higher
than in conventional organic solvents under comparable
conditions.
Selective oxidation of organic compounds is an important area
of synthetic chemistry, the Baeyer–Villiger reaction being one
of the most prominent examples.1 In this context, the use of
molecular oxygen as primary oxidant represents a particularly
attractive approach and at the same time a scientific challenge.2
The Baeyer–Villiger oxidation is generally carried out as a
solution phase process with certain boundary conditions for the
choice of the solvent. The solvent should possess high solubility
for oxygen and organic substrates, it should be stable towards
oxidation to avoid contamination by solvent degradation
products, and it should be non-inflammable for safety reasons.
Therefore, chlorinated solvents are generally preferred solvents
for Baeyer–Villiger oxidation despite the environmental and
toxicological hazards connected with their use.
Recently, compressed CO2 (liquefied or supercritical) has
emerged as an alternative reaction medium in organic synthe-
sis.3 It is environmentally friendly, non-toxic, and highly
miscible with many gases, thus circumventing potential prob-
lems associated with limited mass-transfer in conventional gas–
liquid reaction systems. Especially for oxidation reactions,
safety considerations are undoubtedly the most convincing
argument for the use of compressed CO2 as reaction medium. In
fact, it is a priori non-inflammable and non-oxidisable, and acts
as an efficient diluting agent raising the explosion limit so that
even relatively high substrate and oxygen concentrations can
safely be employed. Intrigued by these obvious advantages,
attempts have been made to employ compressed CO2 for
synthetic processes using oxygen in the presence of homoge-
neous4a–d and heterogeneous4e–g catalysts, or with aldehydes as
co-reductants.5
The study was initiated by screening various oxidants on their
capability to selectively convert cyclohexanone to caprolactone
in liquid or supercritical CO2.‡ Hydrogen peroxide (30% in
water) remained ineffective as oxidant even when hexa-
fluoropropan-2-ol,8 a catalytic amount of Ph2Se2,9 or dimethyl
carbonate10 were added as potential activators. In contrast, the
molecular oxygen–aldehyde combination led to high conver-
sion even in the absence of any additives, affording the desired
cyclic esters with good to excellent yields. Addition of catalytic
amounts of Cu(OAc)2 or Sc(OTf)3 to the reaction mixture did
not improve the conversion further. Among various commer-
cially available aldehydes, benzaldehyde gave the best result
leading to 75% conversion of the ketone under standard
conditions (Scheme 1). Replacing the aldehydes with the
corresponding acetals11 did not result in the formation of
caprolactone.
Scheme 1
The influence of the reaction conditions was investigated in
more detail using cyclohexanone (A) and 3-n-octylcyclobuta-
none (B) as substrates and pivalaldehyde as the co-reductant
(Table 1).
Oxidation processes based on molecular oxygen as primary
oxidant can be achieved with aldehydes as oxygen transfer
agents and sacrificial co-reductants. Various reactions of this
type occur under very mild reaction conditions in common
organic solvents in the presence of metal catalysts (Mukaiyama
conditions).6 Certain aldehyde-based oxidations, including the
Baeyer–Villiger reaction,7 are also known to proceed in the
absence of a catalyst, but the use of chlorinated solvents is
essential for synthetic efficiency. Recently, some of us reported
that highly efficient catalyst-free epoxidation of olefins5a and
oxidation of alkanes5b can be achieved in compressed CO2. The
stainless steel of the reactor wall is responsible for the
generation of an active oxidative species, most likely through its
ability to promote the formation of acylperoxy radicals.
In the present communication we report on the use of
oxygen–aldehyde mixtures for Baeyer–Villiger reactions using
The amount of aldehyde had a major impact on the reaction
course and high conversion required an aldehyde–substrate
ratio of 3+1 (entry 2, 5). In contrast, the partial oxygen pressure
Table 1 Baeyer–Villiger reaction of cycloalkanones in compressed CO2
under different reaction conditions using pivalaldehyde as co-reductant
Aldehyde/
eq.
Entry Substratea
CO2/g
pO2/bar T/°C Conv. (%)
1
2
3
4
5
6
7
8
A
A
A
A
A
B
B
B
3
3
3
3
1.5
3
3
7.3
7.4
7.0
15.1
8.3
9.0
5
20
30
20
20
20
20
20
rt
rt
rt
rt
rt
rt
60
60
71
74
70
39
41
100
42
91
18.6
17.2
10
† Safety warning: The use of compressed gases and especially oxygen in
the presence of organic substrates requires appropriate safety precautions
and must be carried out using suitable equipment only.
a Substrate A: cyclohexanone, substrate B
b Determined by GC using n-decane as internal standard.
= 3-n-octylcyclobutanone.
1588
CHEM. COMMUN., 2002, 1588–1589
This journal is © The Royal Society of Chemistry 2002

did not affect the oxidation significantly in the range of 5–30 bar
(entries 1–3). The noticeable effect of temperature reflects
partly an intrinsic response of the reaction rates and partly the
phase behaviour of the reaction medium. The rate of the
aldehyde auto-oxidation increases more strongly with increas-
ing temperature as compared to the Baeyer–Villiger process.
Consequently, a larger excess of aldehyde is required to achieve
similar ketone conversion at higher temperatures (entry 7, 8). At
the same time, the temperature increase results in transition
from a biphasic (liquid and gaseous CO2) to a monophasic
(supercritical) reaction mixture as confirmed by visual inspec-
tion. The resulting dilution of substrates over the whole reactor
volume makes the oxidation also less effective as demonstrated
by dilution with liquid CO2 at room temperature (entry 3, 4).
Similar observations were made in Baeyer–Villiger reactions of
3-benzylcyclobutanone and bicycloctanone.
The Baeyer–Villiger oxidation of A was run in conventional
organic solvents under conditions comparable to those of Table
1, entry 1 (3 equiv. of aldehyde, rt, atmospheric pressure of O2,
24 h). In the absence of steel, a modest conversion of 23% was
achieved in CH2Cl2 only. In the presence of steel shavings,
conversion were 44% (toluene), 25% (ethyl acetate) and 27%
(CH2Cl2). These data demonstrate that CO2 is a “greener”
replacement for CH2Cl2 providing significant advantages for
the efficiency of the oxidation process.
and cyclohexanone react smoothly under these conditions
giving d- and e-lactone, in 70% and 75% conversion, re-
spectively (entries 4 and 5). Larger rings, such as cyclohepta-
none and cyclooctanone, did not undergo ring expansion.5b The
bicyclic ketones, bicyclo[4.2.0]octan-7-one and bicyclo-
[2.2.1]heptan-2-one, were transformed readily into hexahy-
drobenzofuran-2-one (entry 6) and 2-oxabicyclo[3.2.1]octan-
3-one (entry 7), respectively, following the expected
regioselectivity. The acyclic substrates 3-phenylbutan-2-one
and p-methoxybenzophenone underwent oxidation too (entries
8 and 9), the higher conversion being obtained with the latter
substrate. Also in these cases, the oxygen inserted exclusively
into the C–C-bond between the carbonyl and the most
substituted carbon atom.
In conclusion, we have demonstrated that Baeyer–Villiger
reactions can be performed very efficiently in compressed CO2
at room temperature using oxygen as primary oxidant and an
aldehyde as co-reductant. No additional catalyst is required and
good to excellent yields were achieved for a wide range of
cyclic (up to C6) and acyclic ketones. These results emphasise
the large potential of compressed CO2 as a benign and safe
(‘green’) reaction medium for oxidation processes in fine
chemical synthesis.
Notes and references
An initial assessment of the scope of the reaction under a
standard set of reaction conditions demonstrated that a wide
range of ketones, both cyclic and acyclic can be oxidised
efficiently (Table 2).
Various 3-substituted cyclobutanones are readily converted
to their corresponding lactones (entries 1–3). Cyclopentanone
‡ The reactions were performed in a stainless steel high pressure reactor (25
mL) equipped with thick walled glass windows. In a typical experiment, the
ketone (1 mmol), the aldehyde (3 mmol) and a weighed amount of CO2 (9
g) were introduced into the reactor. The reactor was then charged with O2
from a storage vessel, whereby the partial pressure was controlled by the use
of calibration curves.5 After 18 h, the autoclave was carefully vented
through a cryo-trap (acetone–dry ice, 250 °C) to collect the product. The
autoclave and the trap were washed with MTBE and the combined sample
was directly submitted to GC and GC-MS analysis (n-decane was used as
internal standard).
a
Table 2 Baeyer–Villiger oxidation in compressed CO2
Entry Substrate
1
Co-reductant
Pivalaldehyde
Product
Conv. (%)b
100
1 (a) For comprehensive reviews concerning Baeyer–Villiger oxidations,
see: G. R. Krow, Org. React., 1993, 43, 251; (b) M. Renz and B.
Meunier, Eur. J. Org. Chem., 1999, 4, 737; (c) C. Bolm, in Advances in
Catalytic Processes, ed. M. P. Doyle, JAI Press, Greenwich, 1997, Vol.
2, p. 43; (d) C. Bolm, O. Beckmann and T. K. K. Luong in Transition
Metals for Organic Synthesis, ed. M. Beller and C. Bolm, Wiley-VCH,
Weinheim, 1998, p. 213.
2
Benzaldehyde
82
2 (a) R. A. Sheldon and J. M. Kochi, in Metal-Catalyzed Oxidation of
Organic Compounds, Academic Press, New York, 1981; (b) K. A.
Jørgensen, Chem Rev., 1989, 89, 431.
3 Chemical Synthesis Using Supercritical Fluids, ed. P. G. Jessop and W.
Leitner, Wiley-VCH, 1999, p. 478.
4 (a) X. W. Wu, Y. Oshima and S. Koda, Chem. Lett., 1997, 1045; (b) E.
R. Birbaum, R. M. Le Lacheur, A. C. Horton and W. Tumas, J. Mol.
Catal. A: Chem., 1999, 139, 11; (c) D. Hâncu and E. J. Beckman, Green
Chemistry, 2001, 3, 80; (d) M. Wei, G. T. Musie, D. H. Busch and B.
Subramaniam, J. Am. Chem. Soc., 2002, 124, 2513; (e) A. Martin and B.
Kerler, Chem. Ing. Tech., 2000, 72, 382; (f) G. Jenzer, T. Mallat and A.
Baiker, Catal. Lett., 2001, 73, 5; (g) A. M. Steele, J. Zhu and S. C.
Tsang, Catal. Lett., 2001, 73, 9.
3
4
5
Pivalaldehyde
Pivalaldehyde
Benzaldehyde
75
70
75
6
7
Benzaldehyde
Benzaldehyde
64
92
5 (a) F. Loeker and W. Leitner, Chem. Eur. J., 2000, 6, 2011; (b) N.
Theyssen and W. Leitner, Chem. Commun., 2002, 410.
6 T. Mukaiyama and T. Yamada, Bull. Chem. Soc. Jpn., 1995, 68, 17 and
references therein.
8
9
Pivalaldehyde
Benzaldehyde
48
79
7 (a) K. Kaneda, S. Ueno, T. Imanaka, E. Shimotsuma, Y. Nishiyama and
Y. Ishii, J. Org. Chem., 1994, 59, 2915; (b) X. Li, F. Wang, H. Zhang,
C. Wang and G. Song, Synth. Commun., 1996, 26, 1613.
8 K. Neimann and R. Neumann, Org. Lett., 2000, 2, 2861.
9 G. J. Brink, J. M. Vis, I. W. C. E. Arends and R. A. Sheldon, J. Org.
Chem., 2001, 66, 2429.
10 M. Rüsch gen. Klaas and S. Warwel, Org. Lett., 1999, 7, 1025.
11 K. Yoruzu, T. Takai, T. Yamada and T. Mukaiyama, Bull. Chem. Soc.
Jpn., 1994, 67, 2195.
a For the reaction conditions see note‡. b Determined by GC using n-decane
as internal standard.
CHEM. COMMUN., 2002, 1588–1589
1589