Highly sustainable catalytic dehydrogenation of alcohols with evolution of hydrogen gas

Article abstract of DOI:10.1016/S0040-4039(02)02842-3

The catalytic dehydrogenation of alcohols into aldehydes and ketones in the absence of H-acceptors was studied with several transition metal catalysts in order to develop a large-scale procedure. Applying Ru(OCOCF₃)₂(CO)(PPh₃)₂, the so called Robinson catalyst, several secondary alcohols could be dehydrogenated with high selectivity into the corresponding ketones in relatively short reaction times. Highly effective atom utilization could be realized avoiding solvents and giving hydrogen gas as the sole by-product. However, in contrast to Robinson's work the catalytic dehydrogenation of primary alcohols appeared to be problematic due to decarbonylation with concomitant catalyst deactivation and aldol condensation under the strong acid or basic conditions applied.

Full text of DOI:10.1016/S0040-4039(02)02842-3

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 1507–1509
Highly sustainable catalytic dehydrogenation of alcohols with
evolution of hydrogen gas
G. B. W. L. Ligthart, R. H. Meijer, M. P. J. Donners, J. Meuldijk, J. A. J. M. Vekemans and
L. A. Hulshof*
Eindhoven University of Technology, Laboratory of Macromolecular and Organic Chemistry, PO Box 513, 5600 MB Eindhoven,
The Netherlands
Received 25 November 2002; revised 4 December 2002; accepted 4 December 2002
Abstract—The catalytic dehydrogenation of alcohols into aldehydes and ketones in the absence of H-acceptors was studied with
several transition metal catalysts in order to develop a large-scale procedure. Applying Ru(OCOCF3)2(CO)(PPh3)2, the so called
Robinson catalyst, several secondary alcohols could be dehydrogenated with high selectivity into the corresponding ketones in
relatively short reaction times. Highly effective atom utilization could be realized avoiding solvents and giving hydrogen gas as the
sole by-product. However, in contrast to Robinson’s work the catalytic dehydrogenation of primary alcohols appeared to be
problematic due to decarbonylation with concomitant catalyst deactivation and aldol condensation under the strong acid or basic
conditions applied. © 2003 Elsevier Science Ltd. All rights reserved.
The development of catalytic oxidations under mild
reaction conditions is highly appealing. The fine chemi-
cal industry invests strong effort in the development of
sustainable technologies, based on environmental
acceptability. The criteria include high atom efficiency,
formation of little inorganic waste and selective synthe-
sis of the desired products. Usually, salts represent the
main waste.1,2 Moreover, the reactions are often per-
formed in environmentally unfriendly solvents, typically
chlorinated hydrocarbons.3 Here we wish to report our
initial results with an environmentally benign catalytic
oxidation methodology using transition metal cata-
lysts.4 Homogeneous systems capable of catalyzing
these reactions are rare. Inspired by the work of
Robinson et al.5 in 1975, the feasibility of the dehydro-
genation of primary and secondary alcohols was stud-
ied with respect to avoiding H-acceptors and
from dehydrogenation of the primary alcohols ethanol,
n-propanol and n-butanol was disputed.7
Compared to classical oxidations, the catalytic oxida-
tion accompanied by elimination of hydrogen gas will
enhance the atom efficiency from ꢀ50% up to ꢀ98%.
The catalytic dehydrogenation of secondary alcohols
described in this paper can schematically be represented
by Eq. (1).
The applied Ru(OCOCF3)2(CO)(PPh3)2 catalyst could
be prepared in situ from commercially available chemi-
cals. as reported by Robinson et al.4–6 In addition to
their work with primary alcohols such as ethanol, n-
propanol to n-hexanol and benzyl alcohol and sec-
ondary alcohols such as isopropanol and cyclohexanol,
we found that Ru(OCOCF3)2(CO)(PPh3)2 also catalyzes
selective hydrogen elimination from long chain
aliphatic secondary alcohols. Ketones were produced in
high yields and in relatively short reaction times at
130°C using 1 mol% of catalyst with respect to the
substrate. In the absence of solvents only 12–18 equiv.
(with respect to the catalyst) of CF3CO2H were
required. All reaction mixtures, started as a bright
yellow solution, turned dark red after a few minutes
developing
a
large-scale synthesis using their
Ru(OCOCF3)2(CO)(PPh3)2 catalyst. Although applica-
tion of the catalyst is known to produce hydrogen gas,
to our surprise no systematic and quantitative study
with respect to the fate of the alcohols was presented,4–7
and to the best of our knowledge such information was
not reported elsewhere. Instead, the specific activity of
the Robinson catalyst in hydrogen production resulting
(1)
* Corresponding author.
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)02842-3

1508
G. B. W. L. Ligthart et al. / Tetrahedron Letters 44 (2003) 1507–1509
Four acids of varying strength were investigated in the
dehydrogenation of 2-decanol and 2-octanol using
RuH2(CO)(PPh3)3 as the catalyst precursor. The experi-
ments were performed in the presence of 1 mol%
RuH2(CO)(PPh3)3, with different stoichiometries of acid
compared to the catalyst but always at 130°C in the
absence of solvent. All experiments were stopped after 5
h. The results of these experiments are collected in Table
2.
and changed again to bright yellow at the end of the
reaction. The catalyst could be recycled three times
without significant loss in catalytic activity.
Furthermore, no loss in catalytic activity was observed
when the catalyst came in contact with water. The
turnover number could well exceed 200 in the case of the
oxidation of 2-octanol.8 The results are depicted in Table
1. Very low concentrations of H2 in boiling alcohol as
a result of immediate removal ensure that the reverse
reaction, i.e. hydrogenation of the ketone or aldehyde,
does not play a significant role.
Due to the high volatility of CF3CO2H (bp 72°C)—it
evaporates from the reaction mixture—acids with higher
boiling points were investigated. Small amounts of p-
toluenesulfonic acid (p-TsOH) as relatively strong acid
and varying amounts of trichloroacetic acid and acetic
acid as weaker acids were tested as well. With an
increasing amount of acid the dehydrogenation rate
increases but the selectivity to the corresponding ketone
decreases; presumably due to acid-catalyzed aldol con-
densation of the ketone. In contrast, the selectivity and
rate of dehydrogenation improve with increasing
amounts of the less strong acid CF3CO2H indicating a
plateau value for acid amounts larger than 12 equiv.
The dehydrogenation of secondary alcohols in the
absence of a H-acceptor to produce ketones can be
rationalized with the catalytic cycle for the production of
hydrogen gas from simple alcohols like ethanol or
2,3-butanediol as proposed by Dobson and Robinson.
These authors stated that the catalysis is promoted by the
addition of small amounts of free acid (CF3CO2H) and
is inhibited by aldehyde or ketone and large amounts of
acid.6c However, in our hands, in situ preparation from
1 mol% RuH2(CO)(PPh3)3 and 12–18 molar equiv. (with
respect to the catalyst) of CF3CO2H yielded the highest
selectivities in the dehydrogenation of secondary alco-
hols. To obtain mechanistic information, a ruthenium
complex formed during the dehydrogenation of 2-
octanol with 18 molar equiv. of CF3CO2H was isolated
after distillation of the reaction mixture. The mass (m/z)
of 877 obtained from MALDI-TOF measurements was
assigned to Ru(OCOCF3)2(CO)(PPh3)2 thus proving the
statement of Robinson et al.6
When even less strong acids are used, hydrogen gas is
released from the metal center at a lower rate and
consequently the overall dehydrogenation is slower. This
observation can be rationalized by a more limited ability
to protonate the metal hydride compared to CF3CO2H
and p-TsOH.
In contrast to the claims of Dobson and Robinson, low
conversions and selectivities were observed when primary
Table 1. Ru(OCOCF3)2(CO)(PPh3)2-catalyzed hydrogen elimination from secondary alcohols at 130°C9
Entry
Substrate
Time (h)
Conv. (%)
Yield (%)
Sel. (%)
1
2
3
2-Octanol
2-Decanol
1-Phenylethanol
1-Phenylethanola
4-tert-Butylcyclohexanol
Menthol
5
5
2
5
5
84
86
94
54
53
73
85
81
82
59
49
34
51
80
96
95
63
91
64
70
94
4
5
6
25
5
Cyclohexanol
a 24% v/v of 1-phenylethanol in toluene, 5 molar equiv. (with respect to the catalyst) of CF3CO2H, 110°C.
Table 2. Effect of properties and amount of acid on the catalytic dehydrogenation of 2-decanol or 2-octanola
Acid
pKa in water
−6.5
Eq. acid
Conv. (%)
Yield (%)
Sel. (%)
p-CH3C6H4SO3Hb
1
2
4
0
2
77
97
100
58
52
79
86
84
40
62
55
64
44
24
33
54
82
81
33
49
71
66
44
41
63
68
95
96
83
79
CF3CO2Hc
0.23
6
12
18
5
CCl3CO2Hb
CH3CO2Hb
0.65
4.76
12
a Conditions: neat, 130°C, 1 mol% RuH2(CO)(PPh3)3, 5 h.
b 2-Octanol was used as substrate; some dehydration to 2-octene was observed.
c 2-Decanol was used.

G. B. W. L. Ligthart et al. / Tetrahedron Letters 44 (2003) 1507–1509
1509
References
alcohols were dehydrogenated. The aldehydes proba-
bly decarbonylate and produce CO that poisons the
catalyst as was demonstrated by Geoffroy and Pieran-
tozzi.10 Furthermore, the strong acidic conditions
employed are known to catalyze aldol condensation
and dehydration. Milder reaction conditions have to
be found to render the oxidation of primary alcohols
successful. These milder reaction conditions can prob-
ably be achieved by varying the nature of either the
phosphine or carboxylato ligands.
1. Murahashi, S.-I.; Takaya, H. Acc. Chem. Res. 2000, 33, 225.
2. Hoelderich, W. F. Catal. Today 2000, 62, 115.
3. Sheldon, R. A. J. Mol. Catal. A. 1996, 107, 75.
4. Robinson, S. D.; Uttley, M. F. J. Chem. Soc., Dalton Trans.
1973, 1912.
5. Dobson, A.; Robinson, S. D. J. Organomet. Chem. 1975,
87, C52.
6. (a) Dobson, A.; Robinson, S. D.; Uttley, M. F. J. Chem.
Soc., Dalton Trans. 1975, 370; (b) Dobson, A.; Robinson,
S. D. Inorg. Chem. 1977, 16, 1321; (c) Dobson, A.;
Robinson, S. D. Inorg. Chem. 1977, 16, 137.
Conclusions
7. (a) Rybak, W. K.; Ziolkowski, J. J. J. Mol. Catal. 1981,
11, 365; (b) Jung, C. W.; Garrou, P. E. Organometallics
1982, 1, 658.
An environmentally benign dehydrogenation method-
ology
using
transition
metal
catalysts
like
Ru(OCOCF3)2(CO)(PPh3)2 has been developed. The
criteria include high atom efficiency as well as forma-
tion of little inorganic and organic waste. Several sec-
ondary alcohols have been selectively dehydrogenated
into the corresponding ketones in relatively short
reaction times and in the absence of solvents. Fur-
thermore, the catalysts used in these reactions could
be prepared in situ from commercially available
chemicals. The catalyst could be recycled three times
without significant loss in catalytic activity. As the
best acid for the dehydrogenation CF3CO2H has been
selected.
8. Based on once repeated use of catalyst.
9. All catalytic oxidation experiments were performed in a dry,
oxygen-free argon atmosphere. A typical experiment con-
sisted of the following. An oven-dry 40 ml Radley carousel
reaction tube was flushed with argon before it was charged
with RuH2(CO)(PPh3)3 (91.8 mg, 0.10 mmol) and
CF3CO2H (205 mg, 1.8 mmol). A mixture of alcohol (10.00
mmol) and internal standard (1,3,5-tri-tert-butylbenzene,
81 mg, 0.33 mmol) was added to the in situ prepared
catalyst. The reaction tube was placed in a 12 tube Radley
reaction carousel and the mixture was heated to 130°C and
stirred for several hours. Small aliquots of reaction mixture
weretakenforGCanalysis. Theconversionsandyieldswere
determined with GLC. GLC analyses were performed using
a Zebron ZB-35 column on a Perkin Elmer Autosystem.
Conversions and yields were determined by using 1,3,5-tri-
tert-butylbenzene as internal standard. GC/MS measure-
ments were obtained with a Shimadzu GC/MS-QP5000
using a Zebron ZB-35 column. MALDI-TOF spectra were
obtained on a PerSeptive Biosystems Voyager DE PRO
spectrometer using a-cyano-4-hydroxycinnamic acid as a
matrix. The products were characterized (GLC) by com-
parison with authentic samples (Table 2). Recycling of the
catalystwasestablishedbyevaporatingthereactionmixture
followed by addition of fresh alcohol.
For secondary alcohols the intrinsic catalytic activity
will be studied further by fine-tuning the ligands of
the Robinson catalyst.
Acknowledgements
We gratefully acknowledge the financial support by
the Ministry of Economic Affairs within the frame-
work of the Innovation Oriented Research Program
on Catalysis (IOP-Catalysis). We thank Ir. J. van
Buijtenen for his assistance in various parts of this
study.
10. (a) Pierantozzi, R.; Geoffroy, G. L. Inorg. Chem. 1980, 19,
1821; (b) Geoffroy, G. L.; Pierantozzi, R. J. Am. Chem.
Soc. 1976, 98, 8054.