Efficient, ecologically benign, aerobic oxidation of alcohols

Article abstract of DOI:10.1021/jo982239s

The oxidation of alcohols into aldehydes and ketones can be efficiently performed using catalytic amounts of CuCl·Phen and molecular oxygen or air. This novel, ecologically friendly procedure releases water as the only byproduct.

Full text of DOI:10.1021/jo982239s

J . Org. Chem. 1999, 64, 2433-2439
2433
Efficien t, Ecologica lly Ben ign , Aer obic Oxid a tion of Alcoh ols†
Istva´n. E. Marko´,*,‡ Paul R. Giles,‡ Masao Tsukazaki,‡ Isabelle Chelle´-Regnaut,‡
Arnaud Gautier,‡ Stephen M. Brown,§ and Christopher J . Urch
Universite´ Catholique de Louvain, De´partement de Chimie, Laboratoire de Chimie Organique,
Baˆtiment Lavoisier, Place Louis Pasteur 1, B-1348 Louvain-la-Neuve, Belgium
Received November 10, 1998
The oxidation of alcohols into aldehydes and ketones can be efficiently performed using catalytic
amounts of CuCl‚Phen and molecular oxygen or air. This novel, ecologically friendly procedure
releases water as the only byproduct.
The transformation of alcohols into aldehydes and
ketones is of paramount importance in organic chemistry,
both for laboratory-scale experiments and in the manu-
facturing processes.1 Unfortunately, the vast majority of
the common oxidants have to be used at least in stoichio-
metric amount. Moreover, they are usually hazardous or
toxic and generate large quantities of noxious byprod-
ucts.2 While many ecologically benign processes have
F igu r e 1.
been developed for the reduction of carbonyl derivatives,3
discovery of a novel and ecologically friendly, catalytic
aerobic protocol for the efficient oxidation of alcohols 1
into carbonyl derivatives 2 (Figure 1).6
In this paper, we wish to report full details on the
establishment of this useful catalytic process and develop
further its synthetic utility. In light of some of our
preliminary mechanistic studies, a plausible catalytic
cycle will also be discussed.
Our own work in the area of aerobic oxidations was
inspired by the exquisite research performed on the
structure and reactivity of the binuclear copper proteins,7
hemocyanin and tyrosinase, and by the seminal contribu-
tion of Rivie`re and J allabert.8
similar procedures have been far less investigated for the
oxidation of alcohols.4
Despite their obvious economical and ecological im-
portance, few catalytic systems are available for the
transformation of alcohols into aldehydes and ketones,
using molecular oxygen or air as the ultimate, stoichio-
metric oxidant.5 Moreover, most of the currently available
catalytic oxidation processes suffer from severe limita-
tions, being usually only effective with reactive alcohols,
such as benzylic and allylic ones, or requiring high
pressures, temperatures, and catalyst loading.
We have already described in preliminary form the
These two authors have shown that the simple copper
complex CuCl‚Phen (Phen ) 1,10-phenanthroline) pro-
* To whom correspondence should be addressed. Fax: 32-10-47 27
88. E-mail: marko@chor.ucl.ac.be.
† Dedicated with deep respect to Professor Theodore Cohen.
‡ Universite´ Catholique de Louvain.
(6) (a) Marko´, I. E.; Giles, P. R.; Tsukazaki, M.; Brown, S. M.; Urch
C. J . Science 1996, 274, 2044. (b) Marko´, I. E.; Giles, P. R.; Tsukazaki,
M.; Chelle´-Regnaut, I.; Urch C. J .; Brown, S. M. J . Am. Chem. Soc.
1997, 119, 12661. (c) Marko´, I. E.; Tsukazaki, M.; Giles, P. R.; Brown,
S. M.; Urch C. J . Angew. Chem., Int. Ed., Engl. 1997, 36, 2208. For an
independent report of the aerobic TPAP-catalyzed oxidation of alcohols,
see: Lenz, R.; Ley, S. V. J . Chem. Soc., Perkin Trans. 1 1997, 3291.
(7) For excellent reviews on the formation, isolation, and reactions
of dinuclear copper(II) peroxides, see: (a) Karlin, K. D.; Gultneh, Y.
Progr. Inorg. Chem. 1987, 35, 219-327. (b) Zuberbu¨hler, A. D. In
Copper Coordination Chemistry: Biochemical and Inorganic Perspec-
tives; Karlin, K. D., Zubieta, J ., Ed.; Adenine: Guilderland, New York,
1983. (c) Sakharov, A. M.; Skibida, I. P. Kinet. Catal. 1988, 29, 96-
102. (d) Tyleklar, Z.; J acobson, R. R.; Wei, N.; Murthy, N. N.; Zubieta,
J .; Karlin, K. D. J . Am. Chem. Soc. 1993, 115, 2677-2689. (e) Kitajima,
N.; Fujisawa, K.; Fujimoto, C.; Moro-oka, Y.; Hashimoto, S.; Kitagawa,
T.; Toriumi, K.; Tatsumi, K.; Nakamura, A. Ibid. 1992, 114, 1277-
1291. (f) Fox, S.; Nanthakumar, A.; Wikstrom, M.; Karlin, K. D.;
Blackburn, N. J . Ibid. 1996, 118, 24-34. (g) Solomon, E. I.; Sundaram,
U. M.; Machonkin, T. E. Chem. Rev. 1996, 96, 2563-2605.
(8) (a) J allabert, C.; Rivie`re, H. Tetrahedron Lett. 1977, 1215. (b)
J allabert, C.; Lapinte, C.; Rivie`re, H. J . Mol. Catal. 1980, 7, 127. (c)
J allabert, C.; Rivie`re, H. Tetrahedron 1980, 36, 1191. (d) J allabert,
C.; Lapinte, C.; Rivie`re, H. J . Mol. Catal. 1982, 14, 75. For other
pertinent studies on aerobic oxidation of alcohols using copper com-
plexes, see, for example: (a) Capdevielle, P.; Sparfel, D.; Baranne-
Lafont, J .; Cuong, N. K.; Maumy, M. J . Chem. Res., Synop. 1993, 10
and references therein. (b) Munakata, M.; Nishibayashi, S.; Sakamoto,
H. J . Chem. Soc., Chem. Commun. 1980, 219. (c) Bhaduri, S.; Sapre,
N. Y. J . Chem. Soc., Dalton Trans. 1981, 2585. (d) Semmelhack, M.
F.; Schmid, C. R.; Cortes, D. A.; Chon, C. S. J . Am. Chem. Soc. 1984,
106, 3374.
§ Zeneca Process Technology Department, Huddersfield Works, P.O.
Box A38, Leeds Road, Huddersfield HD2 1FF, U.K.
Zeneca Agrochemicals, J ealott’s Hill Research Station, Bracknell,
Berkshire RG42 6ET, U.K.
(1) For general reviews on oxidation reactions, see: (a) Larock, R.
C. In Comprehensive Organic Transformations; VCH Publishers Inc.:
New York, 1989; p 604. (b) Procter, G. In Comprehensive Organic
Synthesis; Ley, S. V., Ed.; Pergamon: Oxford, 1991; Vol. 7, p 305. (c)
Ley, S. V.; Madin, A. In Comprehensive Organic Synthesis; Trost, B.
M., Fleming, I., Eds.; Pergamon: Oxford, 1991; Vol. 7, p 251. (d) Lee,
T. V. In Comprehensive Organic Synthesis; Trost, B. M., Fleming, I.,
Eds.; Pergamon: Oxford, 1991; Vol. 7, p 291.
(2) Trahanovsky, W. S. In Oxidation in Organic Chemistry; Blom-
quist, A. T., Wasserman, H., Eds.; Academic Press: New York, 1978;
Part A-D.
(3) Noyori, R.; Hashigushi, S. Acc. Chem. Res. 1997, 30, 97 and
references therein.
(4) (a) Sheldon, R. A.; Kochi, J . K. In Metal-Catalyzed Oxidations
of Organic Compounds; Academic Press: New York, 1981. (b) Ley, S.
V.; Norman, J .; Griffith, W. P.; Marsden, S. P. Synthesis 1994, 639.
(c) Murahashi, S.-I.; Naota, T.; Oda, Y.; Hirai, N. Synlett 1995, 733.
(d) Krohn, K.; Vinke, I.; Adam, H. J . Org. Chem. 1996, 61, 1467 and
references therein.
(5) (a) Sheldon, R. A. In Dioxygen Activation and Homogeneous
Catalytic Oxidation; Simandi, L. L., Ed.; Elsevier: Amsterdam, 1991;
p 573. (b) J ames, B. R. In Dioxygen Activation and Homogeneous
Catalytic Oxidation; Simandi, L. L., Ed.; Elsevier: Amsterdam, 1991;
p 195. (c) Ba¨ckvall, J .-E.; Chowdhury, R. L.; Karlsson, U. J . Chem.
Soc., Chem. Commun. 1991, 473. (d) Iwahama, T.; Sakaguchi, S.;
Nishiyama, Y.; Ishii, Y. Tetrahedron Lett. 1995, 36, 6923. (e) Mandal,
A. K.; Iqbal, J . Tetrahedron 1997, 53, 7641 and references therein.
10.1021/jo982239s CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/18/1999

2434 J . Org. Chem., Vol. 64, No. 7, 1999
Marko´ et al.
F igu r e 2.
moted the aerobic oxidation of benzylic alcohols to the
corresponding aromatic aldehydes and ketones (Figure
2).
F igu r e 3.
Unfortunately, 2 equiv of the copper complex have to
be used to achieve good conversions and the system is
severely limited to benzylic substrates. Aliphatic alcohols
proved to be either unreactive or underwent competing
C-C bond cleavage.9
Our fascination for the Rivie`re and J allabert procedure
prompted us to reinvestigate this system and to modify
various parameters with the hope of achieving catalyst
turnover and establishing a useful and efficient aerobic
protocol for the oxidation of all classes of alcohols into
carbonyl derivatives.
Our initial experiments were performed on p-chlo-
robenzyl alcohol and employed 2 equiv of CuCl‚Phen. It
was rather disappointing to find that, besides NaOAc,
all the other bases tested were far less efficient than K2-
CO3.10 However, during the course of these optimization
studies, a dramatic influence of the solvent on the
reaction rate was uncovered. For example, a 3-4-fold
acceleration was obtained when toluene was substituted
for benzene. In contrast, replacing benzene by m- or
p-xylene resulted in a decrease in the rate of the reaction.
Although it is difficult to offer a rational explanation for
the profound effect displayed by minute changes in the
structure of the solvent, it is quite reasonable to assume
that the coordinating properties of these aromatic sol-
vents may alter significantly the stability and reactivity
of the copper complexes.11 Finally, it was also discovered
that molecular oxygen could be replaced by air, a more
readily available and inexpensive stoichiometric oxi-
dant.12
F igu r e 4.
toluene, reducing the quantity of the copper chloride‚Phen
complex did not impair the oxidation of the benzylic
alcohol. Although the reaction took longer to reach
completion, quantitative formation of p-chlorobenzalde-
hyde could be accomplished using as little as 0.05 equiv
of the catalyst.
But the real breakthrough was achieved when it was
decided to lower the amount of the catalyst (Figure 3).
Under the original Rivie`re and J allabert conditions (2
equiv of CuCl‚Phen; benzene) any attempt at decreasing
the concentration of the catalyst resulted in a disastrous
curtailment in the reaction conversion. However, in
Unfortunately, this initial catalytic system proved,
among other things, to be severely restricted to benzylic
alcohols. On the basis of previous work in the biochem-
istry of hemocyanins and tyrosinases,7 a reasonable
mechanism for this aerobic oxidation could be envisioned
in which the µ2-peroxide 6 occupies a cardinal position
(Figure 4). This intermediate 6 can be formed by two
different pathways: (1) either by the displacement of the
chloride ion in complex 3 by the alcohol nucleophile,13
(9) See, for example, ref 8a.
(10) Other bases tested include, e.g., Na2CO3, Li2CO3, Na2HPO4,
NaH2PO4, Al2O3, NaOAc, KOAc, KOH, and CuCO3. Only KOBut
appears to act as an efficient base in the catalytic oxidation process.
Its use is, however, limited at present to the oxidation of secondary
alcohols (Marko´, I. E.; Gautier, A.; Chelle´-Re´gnaut, I.; Mutokole K.
Unpublished results).
(11) Solomon, R. G.; Kochi, J . K. J . Am. Chem. Soc. 1973, 95, 3300.
(12) The use of air instead of oxygen results in a slower reaction
rate. The oxidation can be increased by passing the air through a
porous glass frit, which creates microbubbles. Under these conditions,
the speed of the catalytic oxidation of alcohols using air matches the
one employing oxygen.
(13) The preparation of copper(I) alkoxides and their reactivity
toward O2 has been reported in the literature. See, for example:
Capdevielle, P.; Audebert, P.; Maumy, M. Tetrahedron Lett. 1984, 25,
4397-4400.

Efficient Oxidation of Alcohols
J . Org. Chem., Vol. 64, No. 7, 1999 2435
Ta ble 1. Effect of th e Hyd r a zin e Ad d itives
Ta ble 2. Cop p er -Ca ta lyzed Aer obic Oxid a tion of
Alcoh ols Usin g DEAD-H2
conversion (%)
15 min 30 min
entry
additive
Me2NNH2
(MeO2CNH-)2
1
2
3
4
5
6
7
10
31
98
70
5
39
56
(EtO2CNH-)2 (DEAD-H2)
(iPrO2CNH-)2 (DIAD-H2)
(MeCONH-)2
>99
99
5
(PhCONH-)2
phthalhydrazide
<1
<1
<1
<1
followed by dimerization in the presence of O2, or (2) by
the initial formation of a chloro bis-copper peroxide 4
followed by the exchange of the chloride substituent for
the alcohol ligand. The loaded µ2-peroxide 6 can then
undergo homolytic cleavage of the labile O-O bond and
generate the reactive species 7. Intramolecular hydrogen
abstraction leads to the copper-bound carbonyl derivative
8 with concomitant reduction of CuII (or CuIII) to CuI.
Finally, ligand exchange with the starting alcohol and
release of H2O completes the catalytic cycle (Figure 4).
Such a simple mechanistic proposal accommodated the
observation that highly activated, benzylic alcohols were
good substrates due to the enhanced lability of their
R-hydrogen atoms. In contrast, aliphatic alcohols are far
less reactive toward H-radical abstraction, and accord-
ingly, poor conversions should ensue. However, it was
rather disturbing to note that allylic alcohols, such as
geraniol and nerol, displayed poor reactivity in this
system.
Furthermore, it was observed that the aerobic oxida-
tion of aliphatic alcohols invariably resulted in the rapid
formation of a green copper(II) salt, with concomitant
deactivation of the catalyst. This observation strongly
suggested that the regeneration of the active copper(I)
species was a serious predicament in the oxidation of
aliphatic alcohols. It was therefore decided to test the
effect of various reductants in this aerobic oxidation
reaction. Naturally, we turned to the hydrazine family
of reducing agents (Table 1).14
Remarkably, addition of hydrazine or N,N-dimethyl-
hydrazine (20 mol %) to the reaction mixture resulted in
a significant enhancement in the rate of the oxidation
reaction. The presence of electron-withdrawing groups
on the hydrazine led to an even more dramatic improve-
ment both in yield and reaction rate; the oxidation of 10
was virtually complete within 15 min using DEAD-H2
(Table 1, entry 3). Although the efficiency of the hydra-
zine additive depended to a small extent on steric
hindrance, it was largely affected by electronic factors.
For example, while a small methyl ester substituent
proved less efficient than the bulkier ethyl group, a more
sterically demanding isopropyl ester only reduced slightly
the rate of the reaction; complete conversion being
observed in 30 min (Table 1, entries 2-4). More impor-
a
The conversions were determined by 1H NMR spectroscopy
b
and/or by capillary GC analysis. Neral was not detected in this
reaction. c 20 mol % CuCl‚Phen was employed in this reaction.
tantly, if the ester substituent is replaced by an acyl func-
tion, such as acetyl or benzoyl, virtually no oxidation took
place, regardless of the s-cis or s-trans conformation of
the acyl group (Table 1, entries 5-7). Having found that
optimum conversions could be achieved using as little as
25 mol % of DEAD-H2, we then applied these conditions
to the oxidation of a range of representative alcohols.
Some pertinent results are collected in Table 2.
As can be seen from Table 2, both benzylic and allylic
alcohols underwent smooth and quantitative transforma-
tion into the corresponding aldehyde or ketone within
1-4 h. It is noteworthy that the catalyst tolerates sulfur
heterocycles.15 The stereochemical integrity of the C-C
double bond of the starting allylic alcohols is also retained
in the final products, with geraniol giving solely geranial
(Table 2, entry 5). Remarkably, trifluoromethyl alcohols
and R-ketols are excellent substrates, affording the
corresponding trifluoromethyl ketone and R-diketone
respectively, in high yield (Table 2, entries 3 and 6).
Interestingly, the corresponding azo dicarboxylate
(DEAD) could be substituted to the hydrazide derivative
(DEAD-H2) with equal efficiency.
Unfortunately, both primary and secondary aliphatic
alcohols proved to be poor substrates, and only modest
conversions could be achieved under these conditions,
even when a larger amount of the CuCl‚Phen catalyst
was employed (Table 2, entries 7 and 8).
A plausible mechanism, involving both the azo and
hydrazide derivatives, can be formulated as shown in
Figure 5.
(14) Stoichiometric amounts of substituted azo compounds have been
used to oxidize magnesium alkoxides to the corresponding carbonyl
compounds: Narasaka, K.; Morikawa, A.; Saigo, K.; Mukaiyama, T.
Bull. Chem. Soc. J pn. 1977, 50, 2773. The decomposition mechanism
of hydrazines in the presence of copper complexes has been reported:
(a) Erlenmeyer, H.; Flierl, C.; Sigel, H. J . Am. Chem. Soc. 1969, 91,
1065. (b) Zhong, Y.; Lim, P. K. J . Am. Chem. Soc. 1989, 111, 8398.
(15) Alcohols containing nitrogen heterocycles are also smoothly
oxidized to the corresponding carbonyl compounds. These heterocycles
include pyridine, imidazole, and triazole derivatives.

2436 J . Org. Chem., Vol. 64, No. 7, 1999
Marko´ et al.
F igu r e 5.
In the presence of DEAD-H2 (12) and base (K2CO3),
displacement of the chloride ligand by the hydrazide
nucleophile takes place, affording the hydrazino copper-
(I) complex 13.16 A rapid reaction with oxygen then
ensues, leading to the µ2-peroxo-bis-copper (II) derivative
14. It is believed that, upon heating, this complex
undergoes homolytic cleavage of the labile O-O peroxidic
bond resulting in the generation of the copper-alkoxy
radical 15. Intramolecular hydrogen atom abstraction
ensues, affording the captodatively stabilized17 nitrogen-
centered radical 16, which is nothing else than the azo-
substituted copper(I) hydroxyl species 17. This particular
sequence is thus responsible for the reduction of the
copper(II) species to the catalytically active copper(I)
complex. Ligand exchange with the alcohol and concomi-
tant release of H2O then results in the formation of the
ternary loaded catalyst 18. In this complex, both the
alcohol and azosubstituents are held together in close
proximity by coordination to the copper center. An
intramolecular hydride shift, akin to the Meerwein-
Pondorff-Verley-Oppenauer reaction,18 then takes place,
affording transiently a carbonyl-bound hydrazidocopper
derivative.19
the catalytic cycle. The replacement of DEAD-H2 (12) by
DEAD (19) can be easily understood when considering
this catalytic cycle. Indeed, several entries to the main
catalytic cycle are possible, either via the hydrazino
copper species 13 or via the direct formation of the
ternary loaded complex 18 from the azo derivative 19,
Phen.CuCl (3), and the alcohol 1. The key role played by
the hydrazine or azo compounds can also be readily
appreciated when considering this proposed mechanistic
rationnale. The hydrazide not only helps in reducing the
copper(II) salt to the copper(I) state, but by virtue of its
easy passage into the azo derivative, it also acts as a
hydride acceptor, allowing the efficient oxidation of the
alcohol into the carbonyl compound.
Moreover, we believe that the azo form helps in
stabilizing several of the reactive copper complexes
involved in this catalytic cycle such as the hydroxycopper
complex 17. Thus, we surmise that this novel catalytic,
aerobic oxidation procedure for alcohols into carbonyl
derivatives proceeds via a dehydrogenation mechanism
and relies on the effective role of hydrazine or azo
compounds as hydrogen shuttles and stabilizing ligands
for the various copper complexes.20
The aldehyde or ketone can now desorb, leading to the
initial copper(I) hydrazide complex 13, which re-enters
Further evidence for the occurrence of this dehydro-
genation mechanism can be gathered from the following
experiments. The hydrazidocopper complex 13 can be
independently prepared by reacting Phen.CuCl (3) with
the sodium salt of DEAD-H2. Addition of an alcohol in
the absence of O2 results in no oxidation to the corre-
sponding carbonyl compound. However, when oxygen is
admitted into the reaction medium, rapid and quantita-
tive conversion into the desired product is achieved.
Moreover, combining an alcohol with Phen.CuCl and
DEAD under anaerobic conditions led to the rapid
oxidation of the starting material and the simultaneous
generation of equimolar amounts of DEAD-H2. The
(16) The intermediacy of complex 13 in the aerobic oxidations was
supported by the following observations: (1) independently generated
complex 13 (CuCl. Phen/DBADH2/NaH) proved to be unreactive under
anaerobic conditions; (2) passing O2 through the reaction mixture
containing 13 and alcohol 1 restored the catalytic activity and good
yields of aldehyde 2 were again obtained.
(17) (a) Sustmann, R.; Mu¨ller, W.; Mignani, S.; Mere´nyi, R.; J an-
ousek, Z.; Viehe, H. G. New. J . Chem. 1989, 13, 557. (b) De Boeck, B.;
J anousek, Z.; Viehe, H. G. Tetrahedron 1995, 51, 13239-13246
(18) For general reviews on Oppenauer-type oxidations, see: (a) de
Graauw, C. F.; Peters, J . A.; Vandekkum, H.; Huskens, J . Synthesis
1994, 1007-1017. (b) Djerassi, C. Org. React. (N. Y.) 1951, 6, 207-
212. (c) Krohn, K.; Knauer, B.; Kupke, J .; Seebach, D.; Beck, A. K.;
Hayakawa, M. Synthesis 1996, 1341-1344.
(19) The use of stoichiometric amounts of dipiperidinyl azodicar-
boxamide to oxidise magnesium alkoxides to the corresponding car-
bonyl compounds has been described: Narasaka, K.; Morikawa, A.;
Saigo, K.; Mukaiyama, T. Bull. Chem. Soc. J pn. 1977, 50, 2773-2776.
No reaction is observed under our catalytic anaerobic conditions if
DBAD is replaced by the azodicarboxamide derivative.
(20) Another argument against the oxo-transfer mechanism in our
catalytic, aerobic, oxidation protocol is the lack of formation of
sulfoxides from sulfides, N-oxides from amines, and phosphine oxides
from phosphines. Alkenes also proved to be inert toward oxidation; no
epoxide formation could be detected under our reaction conditions.

Efficient Oxidation of Alcohols
J . Org. Chem., Vol. 64, No. 7, 1999 2437
F igu r e 7.
DEAD-H2 additive are intimately linked, and as the
hydrazide is removed from the reaction mixture, the
formation of the aldehyde simultaneously stops and its
concentration decreases after the hydrazide has been
totally consumed. A similar pattern is observed for the
related DEAD compound. In this case, however, an initial
and extremely rapid transformation of DEAD into DEAD-
H2 takes place. Only the hydrazide can be observed later
in the reaction medium. Closer examination of the
reaction byproducts using stoichiometric DEAD under
anaerobic conditions led to the isolation of unexpectedly
large quantity of the mixed carbonate 20 (Figure 7).
F igu r e 6.
proportion of aldehyde/ketone and hydrazine formed is
equivalent to the quantity of starting azo derivative.
Independent reaction of the alcohol with DEAD and K2-
CO3 in the absence of air and copper salts results in the
quantitative recovery of the starting alcohol.21 Although
we have not yet been able to obtain direct evidence for
some of the intermediates postulated in this catalytic
scheme, we believe that the above-mentionned experi-
ments lend credit to the involvement of complexes 13,
14, and 18 and strongly support our proposed mecha-
nism.
However, two major observations still need to be
accounted for: the lack of reactivity of aliphatic sub-
strates and the need for a 5-fold excess of DEAD or
DEAD-H2 over CuCl‚Phen to achieve quantitative oxida-
tions of benzylic and allylic alcohols.
Thus, the hydrazide or its azo analogue not only plays
a key role in the catalytic cycle as a hydride acceptor and
a reductant for the copper catalyst but also acts as an
acyl transfer reagent generating competitively the un-
desired mixed carbonate 20. This byproduct presumably
originates from the inter- or intramolecular nucleophilic
attack of the alcohol on either the copper hydrazide or
azo complexes 13 or 18, respectively, resulting ultimately
in the deactivation of the catalyst. To minimize this
undesired trans-acylation reaction, sterically demanding
azo derivatives were tested (Figure 7). While diisopropyl
azodicarboxylate (DIAD) produced a more favorable
aldehyde/carbonate ratio (Figure 7, entry 2), we were
gratified to find that the corresponding di-tert-butyl
azodicarboxylate (DBAD) led solely to the formation of
the desired oxidation product with no trace of the mixed
carbonate contaminant (Figure 7, Entry 3).
An interesting clue to these questions was provided
when the fate of the reagents and products involved in
the aerobic catalytic oxidation of undecanol to undecanal
was monitored (Figure 6).22
Whereas the decay of the alcohol follows the expected
kinetic course, the formation of the aldehyde shows an
abnormal behavior. In the early part of the reaction, the
aldehyde formation matches almost perfectly the disap-
pearance of the alcohol. However, after ca. 50% conver-
sion, it reaches a maximum and then slowly begins to
decrease. Clearly, a side reaction is consuming the
aldehyde product as soon as its concentration attains a
critical value. The fate of the DEAD-H2 additive is even
more interesting. In stark contrast to our expectations,
the concentration of DEAD-H2 does not remain constant
throughout the course of the reaction but gradually
decreases over time. The disappearance of DEAD-H2
corresponds exactly to the point of maximum aldehyde
formation. Thus, the destiny of the aldehyde and the
Under these optimized conditions, the aerobic oxidation
of alcohols can be efficiently achieved using as little as 5
mol % of the DBAD or DBAD-H2 additives (Table 3).
Using this improved and more ecologically friendly
protocol, both primary and secondary alcohols are now
smoothly oxidized to the corresponding carbonyl deriva-
tives in high yield (Table 3, entries 7-9).
In summary, we have discovered a simple and envi-
ronmentally friendly, catalytic aerobic protocol for the
efficient oxidation of a wide variety of alcohols into
aldehydes and ketones. This novel catalytic system uses
oxygen or air as the stoichiometric oxidant and releases
water as the sole byproduct. Although much remains to
be done to understand the intimate mechanistic details
of this catalytic oxidation procedure and to further
optimize the reaction conditions, we believe that a
genuine leap has been realized in the establishment of
mild and functionally tolerant, ecologically benign, cata-
lytic systems for the oxidation of alcohols into carbonyl
derivatives.
(21) The oxidation of alcohols using azodicarboxylates has been
previously reported (Yoneda, F.; Suzuki, K.; Nitta, Y. J . Org. Chem.
1967, 32, 727-729.). Control experiments were therefore performed
to establish the need for copper salts in our oxidation procedure. Thus,
under our reaction conditions, no aldehyde or ketone could be detected
in th e a bsen ce of the CuCl‚Phen catalyst, even if phenanthroline was
added as an activating base. Moreover, certain reactive alcohols were
oxidized partially by CuCl‚Phen in th e a bsen ce of the azo derivative
19, though only in moderate yields. These control experiments thus
clearly establish the key role of the copper ion in these oxidations.
(22) The oxidation reactions were monitored by GC (Permabond SE-
52-DF-0.25; 25 m × 0.25 mm i.d.) using tetradecane as the internal
standard.

2438 J . Org. Chem., Vol. 64, No. 7, 1999
Marko´ et al.
Ta ble 3. Cop p er -Ca ta lyzed Aer obic Oxid a tion of
Alcoh ols Usin g DBAD-H2
Acetop h en on e. The oxidation was performed as described
above (procedure A) using 250 mg (2 mmol) of 1-phenylethanol.
The title compound was isolated in 94% yield (226 mg): 1H
NMR (CDCl3, 300 MHz) δH 7.94 (2H, d, J ) 8 Hz), 7.56 (1H,
t, J ) 8 Hz), 7.45 (2H, t, J ) 8 Hz), 2.6 (3H, s); 13C NMR
(CDCl3, 75.5 MHz) δC 198.7, 137.7, 133.7, 129.2, 128.9, 27.2;
IR (film) 1706 cm-1
.
1,1,1-Tr iflu or om eth yla cetop h en on e. The oxidation was
performed as described above (procedure A) using 352 mg (2
mmol) of 1-phenyl-2,2,2-trifluoroethanol. The title compound
was isolated in 91% yield (318 mg): 1H NMR (CDCl3, 300 MHz)
δH 8.1 (2H, d, J ) 7.5 Hz), 7.7 (1H, t, J ) 7.5 Hz), 7.6 (2H, t,
J ) 7.5 Hz); 13C NMR (CDCl3, 75.5 MHz) δC 180.0 (q, J ) 35
Hz), 135.4, 130.0, 129.9, 129.0, 116.3 (q, J ) 292 Hz); IR (film)
1720 cm-1
.
Cin n a m a ld eh yd e. The oxidation was performed as de-
scribed above (procedure A) using 268 mg (2 mmol) of cinnamyl
alcohol. The title compound was isolated in 89% yield (236
mg): 1H NMR (CDCl3, 200 MHz) δH 9.7 (1H, d, J ) 7.7 Hz),
7.5 (5H, m), 6.7 (1H, dd, J 1 ) 16 Hz, J 2 )7.7 Hz), 6.6 (1H, dd,
J 1 ) 16 Hz, J 2 ) 7.7 Hz); 13C NMR (CDCl3, 50 MHz) δC 193.6,
152.7, 133.9, 131.2, 129.0, 128.5, 128.4; IR (film) 1682 cm-1
.
Ger a n ia l. The oxidation was performed as described above
(procedure A) using 308 mg (2 mmol) of geraniol. The title
compound was isolated in 69% yield (211 mg): 1H NMR
(CDCl3, 300 MHz) δH 10.0 (1H, d, J ) 8.1 Hz), 5.8 (1H, d, J )
8.1 Hz), 5.0 (1H, m), 2.2 (4H, bs), 2.1 (3H, s), 1.7 (3H, s), 1.6
(3H, s); 13C NMR (CDCl3, 75.5 MHz) δC 191.1, 163.6, 132.8,
127.3, 122.5, 40.5, 25.7, 25.6, 17.6, 17.5; IR (film) 1677 cm-1
.
Ben zyl. The oxidation was performed as described above
(procedure A) using 424 mg (2 mmol) of benzoin. The title
compound was isolated in 92% yield (388 mg): 1H NMR
(CDCl3, 300 MHz) δH 7.9 (4H, d, J ) 7.8 Hz), 7.7 (2H, t, J )
7.3 Hz), 7.5 (4H, t, J ) 7.4 Hz); 13C NMR (CDCl3, 75.5 MHz)
δC 195.2, 135.5, 133.7, 130.5, 129.7; IR (film) 1727 cm-1
.
a
b
All yields refer to pure, isolated compounds. Neral was not
4-ter t-Bu tylcycloh exa n on e. The oxidation was performed
as described above (procedure A) using 312 mg (2 mmol) of
4-tert-butyl-cyclohexanol. The title compound was isolated in
35% yield (108 mg): 1H NMR (CDCl3, 200 MHz) δH 2.4 (4H,
m), 2.1 (2H, m), 1.5 (3H, m), 0.9 (9H, s); 13C NMR (CDCl3, 50
detected in this reaction. c Geranial was not detected in this
d
experiment. 10 mol % CuCl‚Phen and 10 mol % DBAD were used
in this reaction. e 5 mol % DBAD was employed instead of DBAD-
H2.
MHz) δC 212.3, 46.7, 41.2, 32.4, 27.5, 27.3; IR (film) 1720 cm-1
.
Exp er im en ta l Section
2-Un d eca n on e. The oxidation was performed as described
above (procedure A) using 344 mg (2 mmol) of 2-undecanol.
The title compound was isolated in 39% yield (133 mg): 1H
NMR (CDCl3, 200 MHz) δH 2.4 (2H, t, J ) 7.5 Hz), 2.1 (3H, s),
1.56 (2H, m), 1.3 (12H, m), 0.88 (3H, t, J ) 6.8 Hz); 13C NMR
(CDCl3, 50 MHz) δC 208.9, 43.7, 31.8, 29.7, 29.4, 29.1, 23.8,
Gen er a l P r oced u r e for th e Aer obic Oxid a tion of Al-
coh ols Usin g DEADH2 (P r oced u r e A). P r ep a r a tion of
4-Ch lor oben za ld eh yd e. A 25 mL, two-necked flask was
fitted with a reflux condenser and an oxygen inlet. Toluene
(10 mL) was added followed sequentially by CuCl (9.9 mg;
0.1 mmol; 5 mol %) and phenanthroline (18 mg; 0.1 mmol;
5 mol %). The black complex that formed was stirred at
room temperature for 10 min. Diethylhydrazinodicarboxylate
(DEADH2; 88 mg; 0.5 mmol; 25 mol %) was added followed by
solid K2CO3 (552 mg; 4 mmol; 200 mol %) and the stirring
continued for another 5 min. 4-Chlorobenzyl alcohol (285 mg;
2 mmol) was added neat, and the solution was heated at 90
°C under a gentle stream of oxygen. The oxidation was followed
by removing aliquots and analyzing them by TLC, GC, or 1H
NMR spectroscopy. After 30 min, the reaction was found to
be complete, the mixture was cooled to room temperature and
filtered through a pad of Celigel (50% silica gel mixed with
50% Celite), and the solvent was evaporated in vacuo. The
crude product was further purified by chromatography on silica
gel using CH2Cl2 as the eluant. 4-Chlorobenzaldehyde was
obtained as a colorless liquid (278 mg; 99%): 1H NMR (CDCl3,
300 MHz) δH 9.98 (1H, s), 7.82 (2H, d, J ) 8.4 Hz), 7.5 (2H, d,
J ) 8.4 Hz); 13C NMR (CDCl3, 75.5 MHz) δC 191.3, 141.5, 135.4,
22.6, 13.9; IR (film) 1721 cm-1
.
Aer obic Oxid a tion Usin g Stoich iom etr ic Am ou n ts of
Dia lk yla zod ica r boxyla tes. A 25 mL, two-necked flask is
fitted with a reflux condenser and a oxygen inlet. Toluene (10
mL) was added followed sequentially by CuCl (9.9 mg; 0.1
mmol; 5 mol %) and phenanthroline (18 mg; 0.1 mmol; 5 mol
%). The black complex that formed was stirred at room
temperature for 10 min. The dialkylazodicarboxylate (DEAD,
DIAD or DBAD; 2 mmol; 100 mol %) was added followed by
solid K2CO3 (552 mg; 4 mmol; 200 mol %) and the stirring
continued for another 5 min. 4-Chlorobenzyl alcohol (285 mg;
2 mmol) was added neat, and the solution was heated at 90
°C under a gentle stream of oxygen. After 30 min, the reaction
was found to be complete, the mixture was cooled to room
temperature, filtered through a pad of Celigel (50% silica gel
mixed with 50% Celite), and the solvent was evaporated in
vacuo. The crude product was further purified by chromatog-
raphy on silica gel using CH2Cl2 as the eluant.
131.5, 130.0; IR (film) 1700 cm-1
.
Rea ction Usin g DEAD. 4-Chlorobenzaldehyde was iso-
lated in 30% yield (84 mg) together with 4-chlorobenzylethyl
carbonate (300 mg; 70%).
Th iop h en e-2-ca r boxa ld eh yd e. The oxidation was per-
formed as described above (procedure A) using 230 mg (2
mmol) of thiophene-2-methanol. The title compound was
isolated in 81% yield (183 mg): 1H NMR (CDCl3, 300 MHz)
δH 9.9 (1H, s), 7.7 (2H, m), 7.2 (1H, t, J ) 3.8 Hz); 13C NMR
(CDCl3, 75.5 MHz) δC 182.8, 143.9, 136.1, 135.0, 128.2; IR (film)
4-Ch lor oben zyleth yl ca r bon a te: 1H NMR (CDCl3, 300
MHz) δH 7.3 (4H, bs), 5.1 (2H, s), 4.2 (2H, q, J ) 7.5 Hz), 1.35
(3H, t, J ) 7.4 Hz); 13C NMR (CDCl3, 75.5 MHz) δC 154.9,
134.3, 133.9, 129.5, 128.6, 68.3, 64.1, 14.1; IR (film) 1745 cm-1
;
1673 cm-1
.
MS (EI) m/z (relative intensity) 214 (M+, 75).

Efficient Oxidation of Alcohols
J . Org. Chem., Vol. 64, No. 7, 1999 2439
Rea ction Usin g DIAD. 4-Chlorobenzaldehyde was isolated
in 47% yield (132 mg) together with 4-chlorobenzylisopropyl
carbonate (242 mg; 53%).
mmol) of pyridine-3-methanol. The title compound was isolated
in 81% yield (693 mg): 1H NMR (CDCl3, 300 MHz) δH 10.0
(1H, s), 8.99 (1H, d, J ) 2 Hz), 8.75 (1H, dd, J 1 ) 1.6 Hz, J 2
)
4-Ch lor oben zylisop r op yl ca r bon a te: 1H NMR (CDCl3,
300 MHz) δH 7.3 (4H, bs), 5.1 (2H, s), 4.9 (1H, spt, J ) 6.3
Hz), 1.3 (6H, d, J ) 6.3 Hz); 13C NMR (CDCl3, 75.5 MHz) δC
154.5, 134.3, 133.9, 129.6, 128.7, 72.3, 68.3, 21.7; IR (film) 1748
cm-1; MS (EI) m/z (relative intensity) 228 (M+, 44).
7.2 Hz), 8.09 (1H, dt, J 1 ) 1.7 Hz, J 2 ) 12 Hz), 7.4 (1H, dd, J 1
) 7.2 Hz, J 2 ) 12 Hz); 13C NMR (CDCl3, 50 MHz) δC 190.6,
154.5, 151.8, 135.7, 131.3, 123.9; IR (film) 1717 cm-1
.
4-Meth ylth ioben za ld eh yd e. The oxidation was performed
as described above (procedure B) using 1.23 g (8 mmol) of
4-methylthiobenzyl alcohol. The title compound was isolated
in 92% yield (1.12 g): 1H NMR (CDCl3, 200 MHz) δH 9.8 (1H,
s), 7.64 (2H, d, J ) 8.2 Hz), 7.2 (2H, d, J ) 8.0 Hz), 2.4 (3H,
s); 13C NMR (CDCl3, 50 MHz) δC 190.8, 147.6, 132.8, 129.7,
Rea ction w ith DBAD. 4-Chlorobenzaldehyde was isolated
in 99% yield (278 mg).
Gen er a l p r oced u r e for th e a er obic oxid a tion of a lco-
h ols u sin g DBADH2 (P r ocedu r e B). P r epar ation of 4-Ch lo-
r oben za ld eh yd e. A 100 mL, two-necked flask was fitted with
a reflux condenser and an oxygen inlet. Toluene (40 mL) was
added followed sequentially by CuCl (40 mg; 0.4 mmol; 5 mol
%) and phenanthroline (72 mg; 0.4 mmol; 5 mol %). The dark
complex that formed was stirred at room temperature for 10
min. Di-tert-butyllhydrazinodicarboxylate (DBADH2; 93 mg;
0.4 mmol; 5 mol %) was added followed by solid K2CO3 (2.208
g; 16 mmol; 200 mol %) and the stirring continued for another
5 min. 4-Chlorobenzyl alcohol (1.14 g; 8 mmol) was added neat,
and the solution was heated at 90 °C under a gentle stream
of oxygen. The progress of the reaction was monitored by TLC.
Upon completion of the reaction, the mixture was cooled to
room temperature and filtered through a pad of Celigel (50%
silica gel mixed with 50% Celite), and the solvent was
evaporated in vacuo. The crude product was further purified
by chromatography on silica gel using CH2Cl2 as the eluant.
4-Chlorobenzaldehyde was obtained as a colorless liquid (1.12
g; 99%): 1H NMR (CDCl3, 300 MHz) δH 9.98 (1H, s), 7.82 (2H,
d, J ) 8.4 Hz), 7.5 (2H, d, J ) 8.4 Hz); 13C NMR (CDCl3, 75.5
125.0, 14.4; IR (film) 1708 cm-1
.
Ner a l. The oxidation was performed as described above
(procedure B) using 1.23 g (8 mmol) of nerol. The title
compound was isolated in 73% yield (887 mg): 1H NMR
(CDCl3, 300 MHz) δH 9.9 (1H, d, J ) 8.3 Hz), 5.87 (1H, d, J )
8.2 Hz), 5.12 (1H, t, J ) 7.6 Hz), 2.6 (2H, t, J ) 8.0 Hz), 2.2
(2H, q, J ) 8.0 Hz), 2.0 (3H, s), 1.7 (3H, s), 1.6 (3H, s); 13C
NMR (CDCl3, 75.5 MHz) δC 190.5, 163.5, 133.5, 128.5, 122.2,
32.5, 26.9, 25.5, 17.6, 17.4; IR (film) 1675 cm-1
.
Deca n a l. The oxidation was performed as described above
(procedure B) using 1.26 g (8 mmol) of decanol. The title
compound was isolated in 65% yield (811 mg): 1H NMR
(CDCl3, 200 MHz) δH 9.8 (1H, t, J ) 1.8 Hz), 2.4 (2H, dt, J 1
)
1.8 Hz, J 2 ) 7.2 Hz), 1.6 (2H, t, J ) 7.3 Hz), 1.3 (12H, m), 0.88
(3H, t, J ) 6.1 Hz); 13C NMR (CDCl3, 50 MHz) δC 202.7, 43.8,
31.8, 29.4, 29.3, 29.14, 29.07, 22.6, 22.0, 14.0; IR (film) 1716
cm-1
.
Ack n ow led gm en t. Financial support of this work
was provided by Zeneca Limited, through the Zeneca
Strategic Research Fund. I.E.M. is grateful to Zeneca
for receipt of the Zeneca Fellowship (1994-1997).
MHz) δC 191.3, 141.5, 135.4, 131.5, 130.0; IR (film) 1700 cm-1
.
Table 2 summarizes the yields obtained for the various
substrates oxidized under these conditions.
P yr id in e-3-ca r b oxa ld eh yd e. The oxidation was per-
formed as described above (procedure B) using 872 mg (8
J O982239S