Unexpected role of anionic ligands in the ruthenium-catalyzed base-free selective hydrogenation of aldehydes

Article abstract of DOI:10.1002/anie.201304912

Bigger and better: The replacement of anionic chloride ligands in Noyori-type [(diamine)(diphosphine)RuCl₂] catalysts with bulky carboxylate ligands enabled the efficient selective hydrogenation of a variety of aldehydes under base-free conditions (see scheme). Turnover numbers of up to 100 000 were reached in the presence of a bulky carboxylic acid co-catalyst. This type of catalytic system probably operates through an inner-sphere mechanism. Copyright

Full text of DOI:10.1002/anie.201304912

Angewandte
Chemie
DOI: 10.1002/anie.201304912
Catalytic Hydrogenation
Unexpected Role of Anionic Ligands in the Ruthenium-Catalyzed
Base-Free Selective Hydrogenation of Aldehydes
Philippe Dupau,* Lucia Bonomo, and Laurent Kermorvan
Among alumino- and borohydrides commonly used in
organic synthesis,[1] NaBH4 has been the reagent of choice
for the selective reduction of ketones and aldehydes in the
presence of alkenes.[2] Nevertheless, the discovery in 1995 by
Noyori and co-workers[3] of selective hydrogenation with very
low amounts of [(diphosphine)(diamine)RuCl2] catalysts
afforded a highly sustainable alternative to the use of
stoichiometric amounts of this highly waste generating
reducing agent. Although this methodology is nowadays
employed extensively for the reduction of ketones[4] owing to
its widely recognized exceptional efficiency,[5] only few
examples have been described in the case of aldehydes,[3,6]
possibly because reported systems generally require the use of
a strong base as a cocatalyst. As aldehydes are far more
sensitive to such conditions than ketones, self-aldol side
reactions could readily occur.
led to some different combinations of the P2N2 ligand
system[12] and to the further replacement of a phosphorus
atom with a sulfur atom.[13] As a logical next step, we
therefore investigated the modification of the nature of
anionic ligands in these systems. We also replaced chloride
ligands in the original Noyori system with carboxylates, which
had previously been reported to play an important role in
hydrogen activation.[14] Some ruthenium complexes of type 3
(Scheme 1) have previously been mentioned in the context of
the hydrogenation of ketones in the presence of a base;[15]
however, we now report[16] the use of non-hydride complexes
of this type as highly efficient catalysts for the challenging
base-free selective hydrogenation of aldehydes in the pres-
ence of alkenes.
Starting from readily available [(cod)Ru(OCOR)2]-type
precursors (cod = 1,5-cyclooctadiene),[17] we synthesized
[(diamine)(diphosphine)Ru(OCOR)2] complexes 3, gener-
ally as a cis/trans mixture of biscarboxylate isomers, through
sequential ligand-coordination steps, as described in the
Supporting Information. We tested a very large variety of
complexes of type 3 without isomer separation in the selective
hydrogenation of aldehydes[16] and report herein the use of
a limited number of these catalysts with the substrates shown
in Scheme 2.
For a hydrogenation reaction under base-free conditions,
some nonhalogenated hydride ruthenium catalysts of type
1 and 2 (Scheme 1) were later developed by Noyori and co-
Scheme 1. Different types of ruthenium catalysts for base-free hydro-
genation.
workers[7] and Morris and co-workers.[8] These systems proved
to be intrinsically less active,[9] but they also unfortunately
appeared to be much more sensitive to various reaction
parameters, including substrate quality: an issue that is
generally much more important in the case of aldehydes
than for ketones. Recently, the use of other metals, such as
rhodium and iron, enabled Breit and co-workers[10] and Beller
and co-workers[11] to circumvent this problem and access
unsaturated primary alcohols in high yields through the
hydrogenation of aldehydes.
Scheme 2. Substrates tested in this study.
The influence of the nature of neutral ligands in systems of
the type reported by Noyori and co-workers have been widely
examined in the case of ketone reduction, and these studies
We compared various biscarboxylate complexes with
hydride and borohydride complexes as catalysts in the
hydrogenation of aldehyde A1 in iPrOH under H2 (50 bar)
at a ruthenium loading of 0.01 mol% under base-free neutral
conditions (Table 1). Whereas the acetate and trifluoroace-
tate complexes 3a and 3b afforded only moderate results
comparable to those observed with Noyori-type systems 1a
[*] Dr. P. Dupau, Dr. L. Bonomo, L. Kermorvan
Corporate R&D Division, Firmenich SA
Route de la Plaine 125, 1283 La Plaine (GE) (Switzerland)
E-mail: philippe.dupau@firmenich.com
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304912.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
Table 1: Influence of the ruthenium catalyst on the hydrogenation of
aldehyde A1.[a]
obtained with complete conversion and chemoselectivity and
isolated in high yield (ꢀ 98%).
Entry Ruthenium complex[b]
T
t
Conv. Yield
The catalytic results reported in Table 3 support the
general efficiency of the biscarboxylate complexes of type 3 in
the selective base-free hydrogenation of aldehydes; nearly no
heavy by-products were formed under these conditions. The
catalyst [(en)(dppe)Ru(OCOtBu)2] (3 f) proved to be highly
selective (ꢀ 98%) in the presence of tetra- and trisubstituted,
E, and gem-disubstituted alkenes as well as epoxides, which
remained intact, and aldehydes A1–7 were reduced with high
efficiency to the desired alcohols at 1008C under hydrogen
(10–50 bar) with TONs of 40000–100000. Even the sterically
hindered aldehyde A4 could be hydrogenated at a ruthenium
loading of just 0.002 mol% to afford the desired alcohol with
nearly no alkene hydrogenation or isomerization nor epime-
rization at the a position (99.5% selectivity). In the case of
substrates containing terminal or Z-disubstituted alkenes,
such as A8 and A9, we applied slightly different conditions to
maintain high chemoselectivity (90 and 95%, respectively).
Because of the absence of an acidic cocatalyst (to minimize
alkene isomerization) and the lower reaction temperature,
a higher catalyst loading was required (0.01 mol%). The
reactions also needed to be stopped right upon completion to
avoid a noticeable decrease in selectivity. The complex
[(en)(dppe)Ru(OCO(1-adamantyl))2] (3g), which displayed
[8C] [h][c] [%][d] [%][e]
1
2
trans-[Ru(H)(BH4)(S-binap)(S,S-dpen)] 30 24
(1a)
trans-[Ru(H)(BH4)(S-binap)(S,S-dpen)] 100 24
(1a)
12
15
12
14
3
4
5
6
7
8
9
trans-[Ru(H)(BH4)(dppe)(en)] (1b)[f]
[(en)(dppe)Ru(OCOMe)2] (3a)
[(en)(dppe)Ru(OCOCF3)2](3b)
cis-[(en)(dppe)Ru(OCOtBu)2] (3c)
trans-[(en)(dppe)Ru(OCOtBu)2] (3d)
[(en)(dppe)Ru(OCOPh)2] (3e)
[(en)(dppe)Ru(OCOPh)2] (3e)
100 24
100 24
100 24
14
10
8
13
10
7
100 8.5 100 >99
100
100
8
6
100 >99
100 >99
0
30 24
[a] Reaction conditions: aldehyde A1 (0.05 mol), catalyst (0.01 mol%),
iPrOH (13 mL), H2 (50 bar: maintained throughout the reaction).
[b] binap=2,2’-bis(diphenylphosphanyl)-1,1’-binaphthyl, dpen=1,2-
diphenyl-1,2-diaminoethane, en=ethylenediamine, dppe=1,2-bis(di-
phenylphosphanyl)ethane. [c] The time required for the reaction to reach
completion was determined by H2 consumption. [d] Conversion was
measured by GC. [e] After solvent removal and bulb-to-bulb distillation,
the reaction yield was determined on the basis of product purity by GC.
[f] Complex 1b was synthesized according to the procedure described by
Noyori and co-workers.[7]
and 1b, the pivalate and benzoate derivatives 3c–
e (isomers 3c and 3d were separated by crystalliza-
tion) displayed much higher catalytic activity, with
complete conversion in 6–8.5 h. The desired unsatu-
rated primary alcohol was then isolated in nearly
quantitative yield: we did not detect the formation of
heavy by-products or the hydrogenation of either of
the two alkene groups in the substrate. Although
previously reported ruthenium base-free systems
had been reported to be active at room temper-
ature,[7,8] a higher reaction temperature was gener-
ally required when the biscarboxylate complexes
were used as catalysts.
In attempts to further improve these results, we
found that the reaction is usually most efficient in
apolar aprotic solvents, such as heptane. We then
discovered that the catalytic efficiency could be
further increased significantly by carrying out the
hydrogenation in the presence of catalytic amounts
of carboxylic acids. Turnover numbers (TONs) up to
Table 2: Influence of the carboxylic acid additive on the hydrogenation of aldehyde
A1.[a]
Entry
Carboxylic acid
t
Conv.
[%][c]
Yield
[%][d]
Remarks[e]
[h][b]
1
2
3
4
5
6
7
8
9
none
48
48
24
24
20
20
12
8
2
n.d.
98
3-methylbutanoic acid
2,4,6-trimethylbenzoic acid
isobutyric acid
3,3-dimethylbutanoic acid
1-adamantanecarboxylic acid
pivalic acid
100
100
100
100
100
100
100
100
25% at 6.5 h
65% at 6.5 h
70% at 6.5 h
75% at 6.5 h
75% at 6.5 h
87% at 6.5 h
95% at 6.5 h
>99
>99
>99
>99
>99
>99
>99
benzoic acid
2-naphthoic acid
6.5
[a] Reaction conditions: aldehyde A1 (0.1 mol), catalyst 3a (0.0025 mol%),
carboxylic acid (2.5 mol%), heptane (30 mL), 1008C, H2 (50 bar: maintained
throughout the reaction). [b] The time required for the reaction to reach completion
was determined by H2 consumption. [c] Conversion was measured by GC. [d] After
solvent removal and bulb-to-bulb distillation, the reaction yield was determined on
the basis of product purity by GC; n.d.=not determined. [e] Partial conversion was
40000 and turnover frequencies (TOFs) up to determined on the basis of H2 consumption.
6153 hÀ1 were then observed at complete conversion
in the hydrogenation of the same aldehyde A1 even
with the poorly active bisacetate complex 3a (Table 2).
Whereas the catalytic activity generally increased with the
steric hindrance of the aliphatic carboxylic acid cocatalyst, in
analogy with the trend observed for the carboxylate ligand,
we observed the reverse tendency between 1-adamantane-
carboxylic acid and pivalic acid (Table 2, entries 6 and 7). In
the case of aromatic cocatalysts, highly bulky 2,4,6-trimethyl-
benzoic acid also proved to be inferior to both benzoic acid
and 2-naphthoic acid, which were the most effective cocata-
lysts tested at a 2.5 mol% loading (< Table 2). Under these
mildly acidic conditions, the unsaturated primary alcohol was
higher catalytic activity under such conditions, was used in
these two cases. With a,b-unsaturated aldehyde substrates,
such as A10–12, the catalyst [(en)(Xantphos)Ru(OCOtBu)2]
(3h) was generally used to promote high chemoselectivity (ꢀ
96%) and thus avoid further reduction of the allylic alcohol
formed. Although substrates of this type are often less
sensitive to basic conditions, the desired products were
obtained with complete conversion at catalyst loadings of
0.0025–0.005 mol% and isolated in higher yields than those
observed for reactions carried out under classical basic
conditions.[6,18]
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Table 3: Hydrogenation of aldehydes A1–12 (see Scheme 2).[a]
Entry
Aldehyde
Catalyst
S/Ru[b]
Acid
S/acid[c]
Solvent
P
T
[8C]
t
Sel.
[bar][d]
[h][e]
[%][f]
1
2
3
4
5
6
7
8
9
A1
A2
A3
A4
A5
A6
A7
A8
3 f
3 f
3 f
3 f
3 f
3 f
3 f
3g
3g
3h
3h
3h
50000
100000
40000
50000
50000
40000
40000
10000
10000
20000
40000
30000
2-naphthoic acid
2-naphthoic acid
benzoic acid
2-naphthoic acid
benzoic acid
pivalic acid
benzoic acid
–
80
40
45
80
30
–
–
50
50
10
50
30
50
50
10
20
50
50
30
100
100
100
100
100
100
100
70
8
24
12
10
8
20
8
20
16
8
100
100
100
99.5
98
98.5
100
90
toluene
–
pentane
CH2Cl2
–
pentane
heptane
CH2Cl2
–
60
100
A9
–
80
95
10
11
12
A10
A11
A12
2-naphthoic acid
benzoic acid
benzoic acid
120
100
133
100
100
100
99.5
98.5
96
7
5
–
[a] Reaction conditions: aldehyde (0.5–1 mol), [(en)(dppe)Ru(OCOtBu)2] (3 f), [(en)(dppe)Ru(OCO(1-adamantyl))2] (3g), or [(en)(Xantphos)Ru-
(OCOtBu)2] (3h), carboxylic acid cocatalyst, solvent (1 wt equiv), heat, H2 pressure. Upon reaction completion and after solvent removal (when
required), the crude product was flash distilled under a high vacuum in the presence of a ballast, and the desired alcohol was obtained with residues
comprising less than 1 wt%. [b] Ratio of the substrate to the ruthenium catalyst. [c] Ratio of the substrate to the carboxylic acid cocatalyst. [d] The H2
pressure given was maintained throughout the reaction. [e] The time required for the reaction to reach completion was determined by H2
consumption. [f] The selectivity of the reaction was determined by GC. Xantphos=4,5-bis(diphenylphosphanyl)-9,9-dimethylxanthene.
On the basis of the catalytic data described herein and
a recent experimental–theoretical study reported by Crab-
tree, Eisenstein, and co-workers,[19] we propose that ruthe-
nium catalysts of type 3 could operate through an inner-
sphere mechanism (Scheme 3) even though they display high
nation of Int6 would finally regenerate Int1 and provide the
desired primary alcohol.
Such a reaction mechanism is consistent with the highly
positive effect that we generally observed when the hydro-
genation was carried out in the presence of carboxylic acids.
Nevertheless, competition between a classical acid–base
reaction and H2 deprotonation may lead to some optimal
value for the cocatalyst loading that could potentially
correspond to the variations applied in the reactions in
Table 3. As in the case of H2 deprotonation, too much steric
hindrance could be detrimental to the efficiency of the
protonation of Int6, as observed when 1-adamantanecarbox-
ylic acid and 2,4,6-benzoic acid were used as cocatalysts
(Table 2). Finally, not only could the absence of steric
hindrance potentially disfavor the ionization reaction in
step a, but if an esterification reaction of acetic acid with
the primary alcohol product is responsible for catalyst
deactivation in the case of complex 3a, the pKb value of
trifluoroacetate (13.7) could explain the poor results also
observed for derivative 3b under neutral conditions (Table 1).
In conclusion, we have developed some (diamine)(di-
phosphine)ruthenium-type complexes as unprecedented
highly efficient catalysts for the hydrogenation of aldehydes
under base-free conditions by using sterically hindered
carboxylates as anionic ligands. These relatively air stable
and robust catalysts showed high selectivity for aldehyde
hydrogenation in the presence of alkene and epoxide
functionalities and were successfully used for the hydro-
genation of a large variety of aldehydes. Their better
performance in apolar aprotic solvents and the impressive
increase in catalytic activity in the presence of catalytic
amounts of carboxylic acids strongly suggest that these
catalysts operate by a different reaction mechanism to the
outer-sphere mechanism generally used to describe catalysis
by the Noyori-type system. We are currently further inves-
tigating the reaction mechanism along with the tolerance
towards other functional groups of the base-free hydro-
Scheme 3. Proposed mechanism for the hydrogenation reaction.
selectivity in the presence of alkene functionalities. Biscar-
boxylate complexes of type 3 would first be activated by H2 to
produce a free acid along with a (hydride)(carboxylate)ruthe-
nium intermediate Int1.[20] In step a, Int1 would then undergo
some thermal ionization of the carboxylate ligand with the
formation of the cationic 16-electron species Int2. Such
a reaction could be highly favored by steric congestion caused
by bulky carboxylate ligands. After aldehyde coordination to
À
give Int3 (step b) and insertion into the Ru H bond to give
Int4 (step c),[21] coordinated H2 in Int5 (formed in step d) may
À [22]
be deprotonated by RCO2
to afford neutral Int6 along
with the corresponding carboxylic acid (step e). The proto-
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[12] a) V. Rautenstrauch, R. Challand, R. Churlaud, R. H. Morris, K.
Abdur-Rashid, E. Brazi, H. Mimoun (Firmenich SA), WO
2002022526, 2002 (priority 13.09.2000); b) V. Rautenstrauch, R.
Churlaud, R. H. Morris, K. Abdur-Rashid K (Firmenich SA),
WO 2002040155, 2002 (priority 17.11.2000); c) D. Spasyuk, D. G.
Gusev, Organometallics 2012, 31, 5239 – 5242.
[13] a) C. Saudan, F. Santoro, L. Saudan (Firmenich SA), WO
2012084810, 2012 (priority 21.12.2010); b) D. Spasyuk, S. Smith,
D. G. Gusev, Angew. Chem. 2013, 125, 2598 – 2602; Angew.
Chem. Int. Ed. 2013, 52, 2538 – 2542.
[14] a) Y. Musashi, S. Sakaki, J. Am. Chem. Soc. 2002, 124, 7588 –
7603; b) M.-Y. Ngai, A. Barchuk, M. J. Krische, J. Am. Chem.
Soc. 2007, 129, 280 – 281.
[15] M. J. Burk, W. Hems, A. Zanotti-Gerosa (Chirotech Technology
Ltd), WO 200174829, 2001 (priority 30.03.2000).
genation of aldehydes in the presence of these
[(diamine)(diphosphine)Ru(OCOR)2] catalysts.
Received: June 7, 2013
Revised: July 29, 2013
Published online: && &&, &&&&
Keywords: aldehydes · base-free conditions ·
.
homogeneous catalysis · hydrogenation · ruthenium
[1] J. Seyden-Penne, Reductions by the Alumino- and Borohydrides
in Organic Synthesis, VCH, New York, 1991.
[2] a) H. C. Brown, H. I. Schlesinger, I. Sheft, D. M. Ritter, J. Am.
Chem. Soc. 1953, 75, 192 – 195; b) H. I. Schlesinger, H. C. Brown,
A. E. Finholt, J. Am. Chem. Soc. 1953, 75, 205 – 209; c) H. C.
Brown, S. Krishnamurthy, Tetrahedron 1979, 35, 567 – 607.
[3] T. Ohkuma, H. Ooka, T. Ikariya, R. Noyori, J. Am. Chem. Soc.
1995, 117, 10417 – 10418.
[4] C. Hedberg in Modern Reduction Methods (Ed.: P. G. Ander-
sson, I. J. Munslow), 2008, pp. 109 – 134.
[5] R. Noyori, Angew. Chem. 2002, 114, 2108 – 2123; Angew. Chem.
Int. Ed. 2002, 41, 2008 – 2022.
[16] For intellectual-property reasons, the described results had to be
submitted for patent application prior to publication: P. Dupau,
L. Bonomo, L. Kermorvan (Firmenich SA), WO 2013050297,
2013 (priority 06.10.2011).
[17] L. Bonomo, P. Dupau, S. Bonnaudet (Firmenich SA), WO
2011145032, 2011 (priority 21.05.2010).
[18] P. Dupau, Top. Organomet. Chem. 2012, 42, 47 – 64 (Eds.: M.
Beller, H. U. Blaser).
[19] A. Nova, D. Balcells, N. D. Schley, G. E. Dobereiner, R. H.
Crabtree, O. Eisenstein, Organometallics 2010, 29, 6548 – 6655.
[20] For the synthesis of a ruthenium complex [(tmen)(PPh3)2Ru(H)-
(OCOH)] (tmen = NH2CMe2CMe2NH2) similar to the postu-
lated intermediate Int1 according to an alternative route, see: K.
Abdur-Rashid, S. E. Clapham, A. Hadzovic, J. N. Harvey, A. J.
Lough, R. H. Morris, J. Am. Chem. Soc. 2002, 124, 15104 – 15118.
[21] The activation of aldehydes upon coordination to an electron-
poor ruthenium center was described previously on the basis of
X-ray crystal structures: A. Sorkau, C. Wagner, D. Steinborn, E.
Poetsch, A. E. F. Wꢀchtler, Trans. Met. Chem. 2005, 30, 691 – 695.
[22] For some cationic ruthenium complexes, pKa values as low as
around 6 were reported for coordinated hydrogen: a) G. Jia,
R. H. Morris, J. Am. Chem. Soc. 1991, 113, 875 – 883; b) G. Jia,
A. J. Lough, R. H. Morris, Organometallics 1992, 11, 161 – 171.
[6] A. Takashi, H. Makoto, I. Hisao, A. Akira, Y. Tetsuro, Y. Takeshi
(Takasago), EP 829463, 1998 (priority 17.09.1996).
[7] T. Ohkuma, M. Koizumi, K. MuÇiz, G. Hilt, C. Kabuto, R.
Noyori, J. Am. Chem. Soc. 2002, 124, 6508 – 6509.
[8] S. E. Clapham, R. Guo, M. Zimmer-De Iuliis, N. Rasool, A.
Lough, R. H. Morris, Organometallics 2006, 25, 5477 – 5486.
[9] C. A. Sandoval, T. Ohkuma, K. Muniz, R. Noyori, J. Am. Chem.
Soc. 2003, 125, 13490 – 13503.
[10] L. Diab, T. Mejkal, J. Geier, B. Breit, Angew. Chem. 2009, 121,
8166 – 8170; Angew. Chem. Int. Ed. 2009, 48, 8022 – 8026.
[11] a) S. Fleischer, S. Zhou, K. Junge, M. Beller, Angew. Chem. 2013,
125, 5224 – 5228; Angew. Chem. Int. Ed. 2013, 52, 5120 – 5124;
b) G. Wienhçfer, F. A. Westerhaus, K. Junge, R. Ludwig, M.
Beller, Chem. Eur. J. 2013, 19, 7701 – 7707.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Catalytic Hydrogenation
P. Dupau,* L. Bonomo,
L. Kermorvan
&&&& — &&&&
Unexpected Role of Anionic Ligands in
the Ruthenium-Catalyzed Base-Free
Selective Hydrogenation of Aldehydes
Bigger and better: The replacement of
anionic chloride ligands in Noyori-type
[(diamine)(diphosphine)RuCl2] catalysts
with bulky carboxylate ligands enabled
the efficient selective hydrogenation of
a variety of aldehydes under base-free
conditions (see scheme). Turnover num-
bers of up to 100000 were reached in the
presence of a bulky carboxylic acid coca-
talyst. This type of catalytic system prob-
ably operates through an inner-sphere
mechanism.
Angew. Chem. Int. Ed. 2013, 52, 1 – 5
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!