Functional group-selective poisoning of molecular catalysts: A ruthenium cluster-catalysed highly amide-selective silane reduction that does not affect ketones or esters

Article abstract of DOI:10.1039/b711743d

The addition of amines eliminates the catalytic activity of a triruthenium cluster in the hydrosilane reduction of ketones and esters without affecting the rate of reduction of amides; selective reduction of the amide group in amido ketones and amido esters is accomplished. The Royal Society of Chemistry.

Full text of DOI:10.1039/b711743d

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Functional group-selective poisoning of molecular catalysts: a
ruthenium cluster-catalysed highly amide-selective silane reduction that
does not affect ketones or esters{
Hidehiro Sasakuma,a Yukihiro Motoyamaab and Hideo Nagashima*ab
Received (in Cambridge, UK) 1st August 2007, Accepted 29th August 2007
First published as an Advance Article on the web 14th September 2007
DOI: 10.1039/b711743d
such as triethylamine, diisopropylethylamine or diethylaniline, to
the catalyst solution inhibited the reduction of some substrates.{
To our astonishment, detailed investigations of this amine effect
showed that inhibition of the catalytic activity was highly selective
of functional group and occurred in the reaction of ketones and
esters, but not in the reduction of amides. This discovery provides
a clear answer to the unsolved question of how the less-reactive
amide group is selectively reduced while the ketone or ester group
remains intact.
The addition of amines eliminates the catalytic activity of a
triruthenium cluster in the hydrosilane reduction of ketones
and esters without affecting the rate of reduction of amides;
selective reduction of the amide group in amido ketones and
amido esters is accomplished.
The chemoselective transformation of multi-functionalized sub-
strates by virtue of catalysis is a challenge to be developed in
organic synthesis.1 The use of ‘‘catalyst poisoning’’ is one of the
solutions, in which the addition of a catalyst inhibitor eliminates
some of the properties of the original catalyst, leading to the
accomplishment of ‘‘highly selective reactions’’. As seen in the
famous Lindlar catalyst, which makes possible the selective
hydrogenation of triple bonds to double bonds by the addition
of quinoline or lead tetraacetate to Pd/CaCO3,2a catalyst poisoning
generally decreases the reaction rate, imposing other drawbacks,
such as the difficulty of applying catalysts to the reaction of less-
reactive functional groups.2 The hydride reduction of carbonyl
compounds is commonly performed using metal–hydride systems,3
and the selective reduction of one carbonyl group in a molecule
without affecting the others has long been desired. In reactions
using typical reducing reagents, such as LiAlH4 or NaBH4, the
reactivity of carbonyl compounds is known to decrease in the
order ketone . ester . amide. Therefore, very little selective
reduction of amides containing esters and ketones has been
As representative examples, reaction profiles for the individual
reduction of 2-decanone, methyl decanoate and N,N-dimethyl-
decanamide with PhMe2SiH, catalyzed by 1 (1 mol%), are shown
in Fig. 1 A, whereas those of acetophenone, methyl benzoate and
N,N-dimethylbenzamide are depicted in Fig. 1 C. In the absence of
Et3N, the reduction of all the substrates took place smoothly. At
room temperature, the t1/5 values of 2-decanone, methyl decanoate
and N,N-dimethyldecanamide were 1.4, 0.35 and 0.20 h, whereas
those of acetophenone and N,N-dimethylbenzamide were 6.7 6
1022 and 2.0 6 1022 h, respectively. The reaction of methyl
benzoate was somewhat slower than the other substrates; the t1/5
value was 0.1 h at 50 uC. In sharp contrast, addition of Et3N
(1 equiv. with respect to the substrate) eliminated the catalytic
activity of 1 in the reduction of ketones and esters, as shown in
Fig. 1 B and D; under the same conditions as above, no reduction
product was detected in the reactions of methyl decanoate, methyl
benzoate and acetophenone, whereas less than 2% 2-decanone was
converted to the corresponding silyl ether after 5 h. Interestingly,
no retardation was observed in the reduction of amides in the
presence of Et3N (t1/5 = 0.17 h for N,N-dimethyl decanamide and
1.9 6 1022 h for N,N-dimethyl benzamide). Thus, the catalytic
activity towards ketones and esters was eliminated completely,
whilst the reaction rates of the amides were unchanged.{ In other
words, functional group-selective poisoning of 1 can be achieved
by the addition of appropriate amines.
4
achieved. Notable exceptions are reductions with B2H6 and that
with the Ph2SiH2/RhH(CO)(PPh3)3 system,5 in which the reactivity
of the amide group is sometimes higher than that of the ester
moiety; the amidic group in amides containing ester groups is
reduced preferentially to the aminoester. In contrast, to the best of
our knowledge, there are few examples of reductions, in which the
reactivity of amides are higher than ketones, such that the amides
of keto acids are normally reduced to the amides of hydroxyacids.
We have recently found that a triruthenium carbonyl cluster,
(m3,g2;g3;g5-acenaphthylene)Ru3(CO)7 (1), efficiently catalyses the
reduction of carbonyl compounds with trialkylsilanes at room
temperature within a few hours (Scheme 1).6 In the course of our
study, we were aware that the addition of amines (0.01–1 equiv.),
These results clearly suggest the possibility that selective
reduction of the amide group can be achieved with unchanged
ester and ketone moieties present in the same molecule. This was
aGraduate School of Engineering Sciences, Kyushu University, Kasuga,
Fukuoka 816-8580, Japan. E-mail: nagasima@cm.kyushu-u.ac.jp;
Fax: +81 92-583-7819; Tel: +81 92-583-7819
bInstitute for Materials Chemistry and Engineering, Kyushu University,
Kasuga, Fukuoka 816-8580, Japan
{ Electronic supplementary information (ESI) available: Detailed experi-
mental procedures, and characterization data of both the substrates and
the products. See DOI: 10.1039/b711743d
Scheme 1 Silane reduction of carbonyl compounds with ruthenium
cluster complex 1.
4916 | Chem. Commun., 2007, 4916–4918
This journal is ß The Royal Society of Chemistry 2007

View Article Online
nicely realized: The amide-selective reduction of the molecules
listed in Table 1 was successfully accomplished by three methods.§
As reported previously,6b Method A is reduction in a benzene or
toluene solution. Method B improves the reaction rate by prior
activation of the catalyst. Method C is a further improvement of
Method B using the minimum amount of solvent, leading to
great enhancement of the rate. All reactions proceeded at
room temperature, and the starting materials were consumed in
0.5–20 h. The products were the corresponding amino ketones (4)
or amino esters (5). The only detectable by-products were the
amino alcohols resulting from further reduction of the primary
products (amino ketones 4 and amino esters 5) with excess
amounts of hydrosilane.
As shown in Table 1, judicious choice of Method A, B or C and
the amount of silane led to reduction of the amide group in the
molecules to give the desired amines in good yields. The ratio of
the desired amine to the amino alcohol by-product became 100 : 0
when the reaction conditions were controlled. The addition of
Et3N was not always necessary because reduction of the amide
group produced an amino moiety, which generated catalytic
poisoning similar to Et3N. In the case of the reduction of both the
aromatic and aliphatic amido esters (3a–c), amide-selective
reduction was achieved by all three methods in the absence of
Et3N, and the corresponding amino esters 5a–c were isolated by
alumina column chromatography in 81–96% yields (Table 1,
entries 7–11). In contrast, the concomitant reduction of ketones is
relatively rapid in the reduction of amido ketones (2a–c), which
caused a decrease in selectivity. However, this was improved by the
addition of Et3N (1 equiv.). The use of appropriate amounts of
Fig. 1 The reaction profiles of eqns 1–3. Reactions were carried out
using carbonyl compounds (0.2 mmol), PhMe2SiH (0.6 mmol) and 1
(0.002 mmol) in the absence (A and C) or presence (B and D) of Et3N
(0.2 mmol; 1 equiv. with respect to the carbonyl compound) at room
temperature. The reaction of methyl benzoate was performed at 50 uC;
eqn 1 (6), eqn 2 (#), eqn 3 ($).
Table 1 Amide-selective reduction with PhMe2SiH catalysed by 1a
Entry
Substrate
Methodb (equiv. Si–H)
Time/h
Product
Yield (%)c
Selectivity (%)d
1
2
3
4
A (3.0)
B (3.0)
C (3.0)
C (2.5)
20
8
1
81 (92)
81 (93)
72 (83)
91 (.99)
92
93
83
.99
1
5e
C (2.5)
1.5
94 (.99)
.99
.99
6f
B (3.5)
18
88 (.99)
7f
8f
9f
A (3.0)
B (3.0)
C (3.0)
6
1.5
0.5
88 (.99)
91 (.99)
87 (.99)
.99
.99
.99
10f
C (3.0)
0.5
3.5
96 (.99)
.99
11f
B (3.0)
81 (.99)
.99
a
All reactions were carried out using 1 mmol of substrate, 2.5–3.5 mmol of PhMe2SiH, 1.0 mmol of Et3N and 0.01 mmol of 1 at room
temperature. See footnote and ESI. Isolated yield. The yield in parentheses is that determined by 1H NMR. The ratio (%) of the desired
b
c
d
amino ketones or amino esters in the crude product (determined by 1H NMR in the presence of an internal standard). In entries 1–3, the by-
product was the corresponding amino alcohol. 0.5 mmol of substrate and 0.005 mmol of 1 were used. In the absence of Et3N.
e
f
This journal is ß The Royal Society of Chemistry 2007
Chem. Commun., 2007, 4916–4918 | 4917

View Article Online
consider that catalyst poisoning often kills catalytic activity
towards all functional groups. We conclude that the observed
effect of amines is new and important as a rare example of
the functional group-selective poisoning of homogeneous
catalysts. Further studies on this new poisoning system, in
particular, the mechanisms involved" and their application to
the selective reduction of other functional groups, are now being
conducted.
This work was supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture, Sports, Science
and Technology, Japan.
Scheme 2 Separation of amides contaminated with ketones and esters
by a 1-catalyzed, amide-selective reduction with PhMe2SiH.
PhMe2SiH is sometimes important to raise the selectivity; for
example, the reduction of N-benzyl-N-methyl-6-oxo-heptanamide
(2a) with PhMe2SiH (3 equiv. with respect to 2a), in the presence
of Et3N, showed 83–93% selectivity by all three methods.
However, the use of 2.5 equiv. of silane by Method C suppressed
the reduction of 4a to amino alcohol, giving the desired amino
ketone 4a as a single product (Table 1, entries 1–3 vs. 4). The
reactivity of the aromatic substrate was relatively low compared to
aliphatic examples, and the use of a larger amount of PhMe2SiH
(3.5 equiv.) and the application of a longer reaction time (18 h)
afforded product 4c in a satisfactory yield with 100% selectivity
(Table 1, entry 6).
Notes and references
{ Pyridine and triphenylamine (1 equiv. with respect to carbonyl)
completely inhibit the reduction of esters; however, they are not very
effective for the inhibition of ketone reductions, which are slowly reduced
(10–40% conversion of ketones in 5 h).
§ Method A: A solution of the substrate (1 mmol) and PhMe2SiH
(3.0 mmol) in 2 mL of toluene (or benzene) was added to a flask charged
with 1 (0.01 mmol), and the mixture was stirred at room temperature.
Method B: To a solution of 1 (0.01 mmol) in THP or 1,4-dioxane
(180 mL) was added PhMe2SiH (3.0 mmol), and the mixture stirred for
30 min at room temperature. A solution of the substrate (1 mmol) in 2 mL
of toluene (or benzene) was added to the catalyst solution, and the mixture
was stirred at room temperature.
Method C: To a solution of 1 (0.01 mmol) in THP or 1,4-dioxane
(180 mL) was added PhMe2SiH (2.5 or 3.0 mmol), and the mixture was
stirred for 30 min at room temperature. The substrate (1 mmol: neat) was
added to the catalyst solution, and the mixture was stirred at room
temperature.
" The poisoning was observed in the presence of small amounts of amines
(ca. 10 equiv. with respect to the catalyst). At the present stage, we consider
that the chemical properties of the catalyst species may be being changed by
the coordination of amines to the silyl-ruthenium intermediates formed by
the reaction of 1 with PhMe2SiH.
The present amide-selective reaction is also applicable to the
facile separation of amides from a mixture of amides and ketones,
or of amides and esters. When the catalytic reduction of these
mixtures were carried out in the presence of appropriate amounts
of Et3N and PhMe2SiH, only the amides were converted to the
corresponding amines, with the ketones and esters remaining
intact. A subsequent acidic work-up gave aqueous phases
containing an ammonium salt of the formed amine, while
unreacted ketones and esters were recovered from the organic
phase (Scheme 2). In a typical example, a 5 : 1 molar ratio mixture
of 2-heptanone and N,N-dimethyldecanamide (total 30 mmol) was
subjected to a reduction with PhMe2SiH (15 mmol) in the
presence of 1 (0.05 mmol) and Et3N (5 mmol) to afford
N,N-dimethyldecanamine (0.8 g; 79%) and unreacted 2-heptanone
(2.0 g; 70%), respectively. Similarly, reduction of a 5 : 1 mixture of
ethyl hexanoate and N,N-dimethyldecanamide (total 30 mmol)
resulted in the separation of the unreacted ester (2.46 g; 68%) and
the amine (0.82 g; 71%).{
1 (a) H.-U. Blaser, C. Malan, B. Pugin, F. Spindler, H. Steiner and
M. Studer, Adv. Synth. Catal., 2003, 345, 103; (b) Modern Oxidation
Methods, ed. J.-E. Backvall, Wiley-VCH, Weinheim, 2004, p. 193; (c)
P. Maki-Arvela, J. Hajek, T. Salmi and D. Y. Murzin, Appl. Catal., A,
2005, 292, 1.
2 (a) H. Lindlar and R. Dubuis, Organic Synthesis, John Wiley & Sons,
New York, 1973, collective vol. 5, pp. 880; (b) Other deactivated catalysts
have been reported, see: H. Sajiki, K. Hattori and K. Hirota, J. Org.
Chem., 1998, 63, 7990 and references cited therein.
3 (a) J. Seyden-Penne, Reductions by the Alumino- and Borohydrides in
Organic Synthesis, Wiley, New York, 2nd edn, 1997; (b) H. C. Brown and
S. Krishnamurthy, Tetrahedron, 1979, 35, 567; (c) E. Burkhardt and
K. Matos, Chem. Rev., 2006, 106, 2617.
4 (a) M. J. Kornet, P. A. Thio and S. I. Tan, J. Org. Chem., 1968, 33, 3637;
(b) P. Dietrerich and D. W. Young, Tetrahedron Lett., 1993, 34, 5455; (c)
The NaBH4/I2 system is also useful for this selective reduction: P. Haldar
and J. K. Ray, Tetrahedron Lett., 2003, 44, 8229.
5 (a) R. Kuwano, M. Takahashi and Y. Ito, Tetrahedron Lett., 1998, 39,
1017; (b) G. Gerona-Navarro, M. A. Bonache, M. Alias, M. J. P. de
Vega, M. T. Garcia-Lopez, P. Lopez, C. Cativiela and R. Gonzalez,
Tetrahedron Lett., 2004, 45, 2193.
6 (a) H. Nagashima, A. Suzuki, T. Iura, K. Ryu and K. Matsubara,
Organometallics, 2000, 19, 3579; (b) K. Matsubara, T. Iura, T. Maki and
H. Nagashima, J. Org. Chem., 2002, 67, 4985; (c) Y. Motoyama,
C. Itonaga, T. Ishida, M. Takasaki and H. Nagashima, Org. Synth.,
2005, 82, 188; (d) Y. Motoyama, K. Mitsui, T. Ishida and H. Nagashima,
J. Am. Chem. Soc., 2005, 127, 13150.
The catalytic transformation of multi-functionalized organic
molecules with high chemoselectivity has long been desired by
organic chemists. The silane reduction presented in this
Communication is a clear solution to this issue, containing
the first examples of the selective reduction of amides that
leaves the ketone functionality unchanged. This reduction also
offers a general solution to the selective reduction of amides
without ester groups being affected. The key components that
allow this are appropriate amines, which, as additives, eliminate
catalytic reactivity towards ketones and esters without retarding
the reduction rate of amides. This is noteworthy when you
4918 | Chem. Commun., 2007, 4916–4918
This journal is ß The Royal Society of Chemistry 2007