Dihydrogen reduction of carboxylic esters to alcohols under the catalysis of homogeneous ruthenium complexes: High efficiency and unprecedented chemoselectivity

Article abstract of DOI:10.1002/anie.200701015

(Chemical Equation Presented) The missing link: The presence of two N,P bridges in the ruthenium complexes used as catalysts for the title reaction is essential. Esters that contain isolated C=C bonds are reduced to the corresponding unsaturated alcohols in high yields with high chemoselectivity (see example).

Full text of DOI:10.1002/anie.200701015

Angewandte
Chemie
DOI: 10.1002/anie.200701015
Catalytic Hydrogenation
Dihydrogen Reduction of Carboxylic Esters to Alcohols under the
Catalysis of Homogeneous Ruthenium Complexes: High Efficiency
and Unprecedented Chemoselectivity
Lionel A. Saudan,* Christophe M. Saudan, Catherine Debieux, and Patrick Wyss
Dedicated to Prof. Valentin (Max) Rautenstrauch on the occasion of his 70th birthday
The reduction of carboxylic acid esters to alcohols is
commonly effected with a stoichiometric amount of a highly
reactive metal-hydride reagent (e.g. LiAlH4). The amount of
waste generated by this procedure would decrease strongly
through the use of H2 as the reducing agent. The homoge-
neous catalytic reduction of carboxylic esters to alcohols with
H2 is still a challenging process.[1] The use of mild conditions
(p(H2) < 10 bar, T< 1008C) in combination with an efficient
Noyori-type complexes 4 and 5 were inefficient. A character-
izing feature of complexes 1–3, and a feature which distin-
guishes them from complexes 4 and 5, is the presence of two
catalyst (TON > 2000, TOF > 1000 hÀ1
; TON = turnover
number, TOF = turnover frequency) has never been reported,
although promising results were described recently by
Milstein and co-workers with
a
ruthenium complex.[2]
Herein we report our contribution[3] to the search for a
highly efficient catalyst for the reduction of carboxylic esters
with H2 under mild conditions.
As ruthenium complexes with N,P ligands[4] are among the
most efficient catalysts for the mild hydrogenation of ketones
to alcohols,[5] we questioned their unreported use in the H2
reduction of esters to alcohols. In early experiments aimed at
the reduction of methyl benzoate to benzyl alcohol, we were
very pleased to find that complex 1[6] showed unexpectedly
high catalytic activity (Scheme 1), and we optimized the
conditions for this transformation (Table 1; see also the
Supporting Information).
The best yields were observed in ethereal solvents, and
THF was chosen as the most convenient. In contrast, almost
no reaction was observed in MeOH. A base was required to
transform the ruthenium complex 1 into an active catalyst in
situ. NaOMe (1–10 mol%) gave the best results, whereas
Et3N and DBU (1,8-diazabicyclo[5.4.0]undec-7-ene) were
inefficient. A hydrogen pressure of 50 bar and a temperature
of 1008C were sufficient for fast reactions with a low catalyst
loading.
Scheme 1. Reduction of methyl benzoate by H2 in the presence of
complexes 1–6.[11]
Table 1: Reduction of methyl benzoate by H2 in the presence of complex
1.[a]
We also examined the ruthenium complexes 2–6[7–10] in the
reduction of methyl benzoate to benzyl alcohol under the
previously optimized conditions (see Scheme 1). To our
surprise, complexes 2 and 3 were as active as 1, whereas
Entry
1
NaOMe
[mol%]
T
[8C]
H2
[bar]
t
[h]
Yield[b]
[%]
[mol%]
1
2
3
4
5
6
7
8
0.05
0.05
0.05
0.025
0.01
0.05
0.05
0.05
5
5
1
5
5
5
5
5
100
100
100
100
100
100
100
60
50
50
50
50
50
30
10
50
0.25
1
1
1
4
1
4
2
78
99 (97[c])
96
95
88
96
47
90
[*] Dr. L. A. Saudan, Dr. C. M. Saudan, C. Debieux, P. Wyss
Synthesis, Corporate R&D Division
Firmenich SA
P. O. Box 239, 1211 Geneva 8 (Switzerland)
Fax : (+41)22-780-3334
E-mail: lionel.saudan@firmenich.com
[a] Methyl benzoate: 20 mmol, THF: 10 mL. [b] The yield wasdetermined
by GC with n-tridecane as an internal standard. [c] Yield of the isolated
product.
Supporting information for thisarticle isavailable on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 7473 –7476
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7473

Communications
Table 3: Reduction of aliphatic esters by H2 in the presence of complex 1
or 2.[a]
amino–phosphino-bridged ligands. Moreover, complex 6,
which incorporates only one amino–phosphino-bridged
ligand, was inactive.
The most efficient complexes, 1 and 2, were then used to
examine the scope of this process.[12] Linear and branched
alkyl benzoates were also reduced efficiently (Table 2). The
Entry Ester
Alcohol
[Ru]
t
Yield
[h] [%][b]
Table 2: Reduction of alkyl benzoatesby H in the presence of complex
2
1.[a]
1
2
1
2
2.5 82
2.5 83
3
4
1
2
2.5 96
2.5 90
Entry
R
1 [mol%]
H2 [bar]
t [h]
Yield[b] [%]
5
6
1
2
2.5 88
2.5 89
1
2
3
4
5
6
7
Et
iPr
iPr
iPr
nBu
tBu
CH2Ph
0.05
0.05
0.01
0.05
0.05
0.05
0.05
50
50
50
10
50
50
50
1
1
4
4
1
1
1
99
99
99
99
99
99
99
7
8
1
2
2.5 94
2.5 87
9[c]
1
2
4
4
87
91
10[c]
[a] Alkyl benzoate: 20 mmol, THF: 10 mL. [b] The yield wasdetermined
by GC with n-tridecane asan internal standard.
11[c]
12[c]
1
2
4
4
93
86
use of isopropyl benzoate instead of methyl benzoate allowed
a decrease in either the catalyst loading (0.01 mol%; Table 2,
entry 3) or the H2 pressure (10 bar; Table 2, entry 4). This
improvement is probably due to the quasi-absence of MeOH
in the reaction mixture. The MeOH that forms during the
reduction of methyl benzoate by H2 may deactivate
[a] Ester: 20 mmol, THF: 14 mL. [b] Yield of the isolated product after
column chromatography. [c] THF wasreplaced by toluene, and NaOMe
by KOMe.
the catalyst through carbonylation of the metal.
[a]
=
Table 4: Reduction by H2 of esters with a C C bond.
Furthermore, the reduction of benzyl benzoate
(Table 2, entry 7) was particularly efficient (TON =
2000, TOF = 2000 hÀ1) relative to the equivalent
reaction with the ruthenium catalysts described by
Teunissen and Elsevier (TON = 2071, TOF = 129 hÀ1
at p(H2) = 85 bar, T= 1208C),[1b] and by Milstein and
co-workers (TON = 99, TOF = 14 hÀ1 at p(H2) =
5 bar, T= 1158C).[2] Aliphatic methyl esters (Table 3,
entries 1–8) and lactones (Table 3, entries 9–12) were
also reduced efficiently to provide the corresponding
alcohols and diols in high yields (82–96%).
Entry Ester
Major
alcohol
Product Yield
ratio[b]
[%][c]
1
98:2
90
2
99:1
93
3[d]
98.5:1.5 85
Next, the chemoselectivity of the process was
=
examined with esters that contained a C C bond
(Table 4). The reduction of methyl 3-cyclohexenecar-
boxylate (Table 4, entry 1) was investigated with
complexes 1–3. Although only moderate conversion
was observed with 1 (50%), good conversions were
observed with 2 (95%) and 3 (81%), and in all cases
4
>98:2 95
5
6
99:1
94
94
=
the C C bond was affected only minimally (unsatu-
35:65
rated/saturated product > 95:5). Complex 2, which
showed the highest reactivity and chemoselectivity,
was then used to investigate the scope and limitations
of the reaction with respect to the structure of the
unsaturated ester substrate.[12] As illustrated in
Table 4, the number of substituents on the alkene
group had a crucial effect on the chemoselectivity. An
ester and a lactone that incorporated an acyclic
7
12:88
87
[a] Ester: 20 mmol, THF: 10 mL. [b] The product ratio (unsaturated alcohol/
saturated alcohol) was determined by GC analysis of the crude reaction mixture.
[c] Yield of the isolated product as a mixture of unsaturated and saturated
alcoholsafter column chromatography or distillation. [d] THF wasreplaced by
toluene, and NaOMe by KOMe; reaction time: 3 h.
7474
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7473 –7476

Angewandte
Chemie
disubstituted alkene (Table 4, entries 2and 3) were reduced
with high chemoselectivity (> 98:2), and the desired unsatu-
rated alcohol and diol were obtained in high yields. Esters
with a cyclic or an acyclic trisubstituted alkene were also
reduced successfully under the same conditions (Table 4,
entries 4 and 5). In contrast, methyl 10-undecenoate, an ester
with a monosubstituted alkene functionality, was reduced
describes the different possible reactions that could occur
under the conditions of Table 4 with an ester that contains an
=
isolated C C bond. The H2 reduction of such an ester 7 to the
corresponding alcohol 10 could start with the hydrogenation
=
of the C O bond of 7 to give the hemiacetal 8. The
fragmentation of 8 into aldehyde 9 followed by hydrogena-
=
tion of the C O bond of 9 to give alcohol 10 could then
=
with competitive hydrogenation of the C C bond to give the
terminate the reaction. The unsaturated compounds 7–10
=
saturated alcohol as the major product (65:35; Table 4,
entry 6). When this reaction was stopped after 23 min, GC
analysis of the reaction mixture showed the following
composition: unsaturated ester (25%), saturated ester
(3%), unsaturated alcohol (47%), saturated alcohol (5%),
transesterified products (20%). This result indicates that the
could also undergo hydrogenation of the C C bond to give
the corresponding saturated products 11–14. In analogy to the
reduction of 7 to 10, the saturated ester 11 could be reduced to
alcohol 14. A comprehensive reaction scheme should also
include the possible transesterification of esters 7 and 11 with
alcohols 10 and 14.
=
hydrogenation of the C C bond is slower than the reduction
In the hydrogenation of ketones with diamine–diphos-
phine ruthenium complexes, the active catalyst is presumed to
be a trans-dihydride complex formed from the corresponding
dichloride complex upon reaction with an alkoxide base in the
presence of H2.[13] The transition state of this ketone reduction
is postulated to involve the assistance of a NH ligand and does
of the ester functionality. Therefore, the amount of unsatu-
rated alcohol obtained could be maximized by further
optimization of the reaction conditions. The extent of C C
bond hydrogenation was even higher in the reduction of an
a,b-unsaturated ester (Table 4, entry 7). When this reaction
was stopped after 15 min, GC analysis of the reaction mixture
showed the following composition: unsaturated ester (10%),
saturated ester (35%), unsaturated alcohol (35%), saturated
alcohol (10%), transesterified products (10%). In this case,
=
=
not involve the coordination of the C O bond to the metal
center.[5a,13] Similarly, the C O hydrogenation steps in
=
Scheme 2could imply a trans-dihy-
dride catalyst formed from complex
the hydrogenation of the activated C C bond appears to be as
1[4b] or 2,[5c] and the same type of
=
fast as the reduction of the ester.
outer-sphere proton–hydride trans-
At present, little is known about the mechanism of this
process and the nature of the active catalyst. Scheme 2
fer could occur via the transition
state TS.
The substitution of the NH2
groups in complex 1 with NMe2
groups[14] led to a complete shut-
down of the catalytic activity. This
effect was also reported for the
hydrogenation of ketones[5a] and
strengthens the hypothesis of an NH-assisted mechanism.
Moreover, the results in Scheme 1 show clearly that two
amino–phosphino-bridged ligands are needed. However, the
relationship between this structural arrangement and the high
catalytic activity has not yet been established.
=
In the H2 reduction of esters that contain an isolated C C
bond, the reduction of the ester group was rapid relative to
=
the hydrogenation of a di- or trisubstituted C C bond
(Table 4, entries 1–5). Even in the presence of a less sterically
=
hindered monosubstituted C C bond (Table 4, entry 6), good
chemoselectivity (90:10) was observed in the early stages of
=
the reaction. The observed hydrogenation of C C bonds in
these compounds probably occurs through coordination to
the metal center and is therefore sterically more demanding
=
than the proposed outer-sphere hydrogenation of the C O
bond. This difference may explain the faster hydrogenation of
=
=
the C O bond relative to that of the C C bond for almost all
compounds studied.
In summary, we have developed a reduction of aliphatic
and aromatic carboxylic esters to alcohols with H2 under the
highly efficient catalysis (TON ꢀ 2000, TOF = 800–2000 hÀ1)
of homogeneous ruthenium complexes with N,P ligands.
Under the relatively mild conditions required (p(H2) =
=
50 bar, T= 1008C), esters with a di- or trisubstituted C C
Scheme 2. The possible reaction pathways of an ester 7 with an
=
isolated C C bond under the hydrogenation conditions.
bond were reduced to the corresponding unsaturated alcohols
Angew. Chem. Int. Ed. 2007, 46, 7473 –7476
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7475

Communications
Saudan, P. Dupau, J.-J. Riedhauser, P. Wyss (Firmenich SA), WO
2006106483, WO 2006106484, 2006 (priority 05.04.2005).
[4] a) V. Rautenstrauch, R. Challand, R. Churlaud, R. H. Morris, K.
Abdur-Rashid, E. Brazi, H. Mimoun (Firmenich SA), WO
200222526, 2002 (priority 13.09.2000); b) K. Abdur-Rashid, R.
Guo, A. J. Lough, R. H. Morris, Adv. Synth. Catal. 2005, 347, 571;
c) F. Naud, C. Malan, F. Spindler, C. Rüggeberg, A. T. Schmidt,
H.-U. Blaser, Adv. Synth. Catal. 2006, 348, 47.
with high chemoselectivity (unsaturated/saturated product
> 98:2) in greater than 85% yield. The operative mechanism
has not yet been elucidated and is currently under inves-
tigation. Finally, this process constitutes a significant step
towards the discovery of a perfect alternative to waste-
generating ester reduction with stoichiometric metal-hydride
reagents.
[5] For leading references, see: a) R. Noyori, T. Ohkuma, Angew.
Chem. 2001, 113, 40; Angew. Chem. Int. Ed. 2001, 40, 40; b) V.
Rautenstrauch, X. Hoang-Cong, R. Churlaud, K. Abdur-Rashid,
R. H. Morris, Chem. Eur. J. 2003, 9, 4954; c) T. Li, R. Churlaud,
A. J. Lough, K. Abdur-Rashid, R. H. Morris, Organometallics
2004, 23, 6239; d) T. Ohkuma, C. A. Sandoval, R. Srinivasan, Q.
Lin, Y. Wei, K. Muꢀiz, R. Noyori, J. Am. Chem. Soc. 2005, 127,
8288; e) P. D. de Koning, M. Jackson, I. C. Lennon, Org. Process
Res. Dev. 2006, 10, 1054; f) H. Huang, T. Okuno, K. Tsuda, M.
Yoshimura, M. Kitamura, J. Am. Chem. Soc. 2006, 128, 8716.
[6] For the preparation of complex 1 and its use in the hydro-
genation of ketones, see reference [4a,b].
[7] For the preparation of complex 2 and its use in the hydro-
genation of ketones, see: a) J.-X. Gao, H.-L. Wan, W.-K. Wong,
M.-C. Tse, W.-T. Wong, Polyhedron 1996, 15, 12 41; b) V.
Rautenstrauch, R. Churlaud, R. H. Morris, K. Abdur-Rashid
(Firmenich SA), WO 200240155, 2002 (priority 17.11.2000).
[8] Complex 3 was prepared in the same way as complex 2; see
reference [7]. For the preparation of the ligand in complex 3,
see: T. D. Dubois, Inorg. Chem. 1972, 11, 718.
Experimental Section
Representative procedure: Complex 1 (6.7 mg, 0.01 mmol), solid
NaOMe (54.8 mg, 1.0 mmol), and THF (5 mL) were placed in a
stainless-steel 75-mL autoclave equipped with a magnetic stirring bar
under argon. A solution of methyl benzoate (2.73 g, 20 mmol) in THF
(5 mL) was added, and the autoclave was purged by three successive
cycles of pressurization/venting with H2 (20 bar), then pressurized
with H2 (50 bar), closed, and placed in a oil bath with a thermostat at
1008C for 1 h. The autoclave was then cooled in an ice/water bath and
vented. The reaction mixture was diluted with methyl tert-butyl ether
(MTBE; 50 mL) and washed successively with saturated aqueous
NaHCO3 (25 mL) and saturated aqueous NaCl (2 ” 25 mL). The
combined aqueous phases were extracted with MTBE (2” 25 mL),
and the combined organic phases were then dried (MgSO4), filtered,
and concentrated under vacuum. Kugelrohr distillation of the residue
gave the benzyl alcohol (2.1 g, 19.4 mmol, 97%) as a colorless liquid.
Received: March 7, 2007
Revised: June 11, 2007
Published online: August 14, 2007
[9] For the preparation of complex 4, see: H. Doucet, T. Ohkuma,
K. Murata, T. Yokozawa, M. Kozawa, E. Katayama, A. F.
England, T. Ikariya, R. Noyori, Angew. Chem. 1998, 110, 1792;
Angew. Chem. Int. Ed. 1998, 37, 1703.
[10] Complexes 5 and 6 were prepared by a method described for
similar complexes; see reference [9].
Keywords: homogeneouscatalysis· hydrogenation ·
.
N,P ligands· reduction · ruthenium
[11] Yields were determined by GC in the presence of n-tridecane as
an internal standard.
[12] The reaction conditions were not optimized for each substrate.
[13] a) C. A. Sandoval, T. Ohkuma, K. Muꢀiz, R. Noyori, J. Am.
Chem. Soc. 2003, 125, 13490; b) R. Abbel, K. Abdur-Rashid, M.
Faatz, A. Hadzovic, A. J. Lough, R. H. Morris, J. Am. Chem. Soc.
2005, 127, 1870; c) R. J. Hamilton, S. H. Bergens, J. Am. Chem.
Soc. 2006, 128, 13700.
[14] For the preparation of the analogue of complex 1 with NMe2
groups, see: J. Y. Shen, C. Slugovc, P. Wiede, K. Mereiter, R.
Schmid, K. Kirchner, Inorg. Chim. Acta 1998, 268, 69.
[1] a) T. Ohkuma, R. Noyori in Transition Metals for Organic
Synthesis (Eds.: M. Beller, C. Bolm), Wiley, New York, 1998,
pp. 25 – 69; b) H. T. Teunissen, C. J. Elsevier, Chem. Commun.
1998, 1367; c) M. L. Clark, M. B. Díaz-Valenzuela, A. M. Z.
Slawin, Organometallics 2007, 26, 16.
[2] J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, Angew. Chem.
2006, 118, 1131; Angew. Chem. Int. Ed. 2006, 45, 1113.
[3] For industrial reasons, the results of part of this research had to
be submitted for patent applications prior to publication: L.
7476 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7473 –7476