Hydrogenation of hindered ketones catalyzed by a silica-supported compact phosphine-Rh system

Article abstract of DOI:10.1021/ol801996r

(Chemical Equation Presented) A heterogeneous mono(phosphine)-Rh catalyst system silica-SMAP-Rh(OMe)(cod), where silica-SMAP stands for a caged, compact trialkylphosphine (SMAP) supported on silica gel, showed broad applicability toward the hydrogenation of hindered ketones. Doubly α-branched ketones such as diisopropyl ketone was hydrogenated under nearly atmospheric conditions. Di-tert-butyl ketone could be hydrogenated under more forcing conditions.

Full text of DOI:10.1021/ol801996r

ORGANIC
LETTERS
2008
Vol. 10, No. 20
4697-4700
Hydrogenation of Hindered Ketones
Catalyzed by a Silica-Supported
Compact Phosphine-Rh System
Soichiro Kawamorita,† Go Hamasaka,† Hirohisa Ohmiya,† Kenji Hara,‡
Atsushi Fukuoka,‡ and Masaya Sawamura*,†
Department of Chemistry, Faculty of Science, Hokkaido UniVersity, Sapporo 060-0810,
Japan, and Catalysis Research Center, Hokkaido UniVersity, Sapporo 001-0021, Japan
sawamura@sci.hokudai.ac.jp
Received August 29, 2008
ABSTRACT
A heterogeneous mono(phosphine)-Rh catalyst system silica-SMAP-Rh(OMe)(cod), where silica-SMAP stands for a caged, compact
trialkylphosphine (SMAP) supported on silica gel, showed broad applicability toward the hydrogenation of hindered ketones. Doubly r-branched
ketones such as diisopropyl ketone was hydrogenated under nearly atmospheric conditions. Di-tert-butyl ketone could be hydrogenated under
more forcing conditions.
Hydrogenation of ketones is feasible with various types of
catalysts under transfer hydrogenation conditions that make
use of secondary alcohols as a hydrogen source,1 but catalysts
effective for that of simple, nonchelating ketones with
molecular hydrogen is not so common.2 In particular, the
hydrogenation of hindered ketones is extremely difficult.3,4
In the course of our studies on the application of caged,
compact trialkylphosphine Ph-SMAP5 toward transition
metal catalysis, we found that its rhodium complex displays
exceptionally high catalytic activity for the hydrogenation
of simple, noncoordinating, and sterically demanding
ketones.5c On the other hand, in our previous study, the silica-
immobilized SMAP (silica-SMAP, 1) was used for the
selective formation of a mono(phosphine)-rhodium species.
The immobilized complex was identified to be a precursor
that generates a highly active catalyst for the hydrosilylation
of ketones with triorganosilanes especially for the reaction
† Department of Chemistry.
‡ Catalysis Research Center.
(1) Klomp, D.; Hanefeld, U.; Peters, J. A. In Handbook of Homogeneous
Hydrogenation; de Vries, J. G.; Elsevier, C. J., Ed.; Wiley-VCH: Winheim,
Germany, 2007; Vol. 2, pp 585-630.
(2) (a) Blum, Y.; Czarkle, D.; Rahamim, Y.; Shvo, Y. Organometallics
1985, 4, 1459–1461. For review on homogeneous hydrogenation of carbonyl
compounds, see: (b) Clarke, M. L.; Roff, G. J. In Handbook of Homogeneous
Hydrogenation; de Vries, J. G.; Elsevier, C. J., Ed.; Wiley-VCH: Winheim,
Germany, 2007; Vol. 1, pp 413-454.
(3) Diisopropyl ketone (3a) could not be reduced with Raney nickel or
with nickel and some promoters, but was finally reduced with an equal
weight of catalyst promoted with chloroplatinic acid and sodium hydroxide.
See: (a) Blance, R. B.; Gibson, D. T. J. Chem. Soc. 1954, 2487–2489. See
also: (b) Freiferder, M. In Catalytic Hydrogenation in Organic Synthesis.
Procedures and Commentary; Wiley: New York, 1978; Chapter 9, pp
78-89. Hydrogenation of 3a with Shvo’s Ru catalyst required a high
hydrogenation pressure (34 atm) and a high reaction temperature (145 °C,
3 h) to gain the TON of 1760 (ref 2a).
(4) For the vapor phase transfer hydrogenation using isopropyl alcohol
as a hydrogen donor over a calcinated Mg-Al hydrotalcites catalyst at 200
°C, see: Jyothi, T. M.; Raja, T.; Rao, B. S. J. Mol. Catal., A 2001, 168,
187–191
.
(5) SMAP: silicon-constrained monodentate trialkylphosphine. See: (a)
Ochida, A.; Hara, K.; Ito, H.; Sawamura, M. Org. Lett. 2003, 5, 2671–
2674. (b) Ochida, A.; Ito, S.; Miyahara, T.; Ito, H.; Sawamura, M. Chem.
Lett. 2006, 35, 294–295. (c) Ochida, A.; Hamasaka, G.; Yamauchi, Y.;
Kawamorita, S.; Oshima, N.; Hara, K.; Ohmiya, H.; Sawamura,
Organometallics, in press.
10.1021/ol801996r CCC: $40.75
Published on Web 09/24/2008
2008 American Chemical Society

Table 1. Hydrogenation of Diisopropyl Ketone (3a) Catalyzed by Rh-Phosphine Systemsa
catalyst
precursor
b
entry
Rh/3a
solvent
H2 (atm)
temp (°C)
time (h)
yieldc (%)
1
Silica-SMAP-Rh(OMe)(cod) (2a)
0.005
0.005
0.001
0.005
0.005
0.005
0.005
0.005
0.005
0.005
0.01
0.01
0.01
0.01
0.01
0.01
0.01
hexane
hexane
hexane
MeOH
i-PrOH
THF
CH2Cl2
MeOH
MeOH
MeOH
THF
THF
THF
THF
THF
THF
THF
5
5
10
5
5
5
5
5
5
5
23
23
50
23
23
23
23
23
23
10
3
6
100
100
98
57
95
45
6
13
32
0
0
46
1
4
0
2
2a/Et3Nd
3e
4
2a/Et3N
2a
2a
2a
10
10
10
10
10
10
10
20
20
20
20
20
20
20
5
6
7
8
2a
silica-SMAP-RhCl(cod) (2b)
silica-SMAP (1)/[Rh(cod)2]BF4
2b/Na[B{3,5-(CF3)2-C6H3}4] (1:1)
9
10
11f
12f
13f
14f
15f
16f
17f
23
23
g
Ph-SMAP/{Rh(OMe)(cod)}2
5
g
Ph-SMAP/{Rh(OMe)(cod)}2
30
30
30
30
30
30
100
100
100
100
100
100
g
Me3P/{Rh(OMe)(cod)}2
Bu3P/{Rh(OMe)(cod)}2
g
g
(t-Bu3)P/{Rh(OMe)(cod)}2
g
Ph3P/{Rh(OMe)(cod)}2
0
0
h
dppe/{Rh(OMe)(cod)}2
c
a 3a, 1.0 mmol; solvent, 1.0 mL unless otherwise noted. b Initial hydrogen pressure. Determined by H NMR spectroscopy using 1,1,2,2-tetrachloro-
1
ethane as an internal standard. d Et3N; 0.1 mol %. e 3a, 5.0 mmol; Et3N, 0.02 mol %; hexane 1 mL. f Data taken from ref 5c. g Rh/P; 1:1. h Rh/P; 1:2.
of sterically demanding substrates.6,7 Accordingly, we de-
cided to examine silica-SMAP for ligand performance in the
Rh-catalyzed hydrogenation of hindered ketones and found
that anchoring SMAP on the silica support caused remarkable
increase of catalytic activity.
Results of the hydrogenation of diisopropyl ketone (3a)
are summarized in Table 1. The hydrogenation of 3a
catalyzed by 0.5 mol % of silica-SMAP-Rh(OMe)(cod) [2a,
0.06-0.07 mmol g-1 (for Rh)] (S/C ) 200), which was
prepared from [{Rh(OMe)(cod)}2] and silica-SMAP (1)
(Scheme 1)8 proceeded smoothly at rt with 5 atm of initial
Alcoholic solvents such as MeOH and i-PrOH as well as
THF can be used but the reactions were significantly slower
than in hexane (entries 4-6). The reaction was almost
inhibited with CH2Cl2 solvent (entry 7). The chloro complex
silica-SMAP-RhCl(cod) (2b)6b is less reactive than methoxo
complex 2a (entries 1 vs 8). The catalyst system that was
prepared in situ from 1 and cationic complex [Rh(cod)2]BF4
was slightly less active than the methoxo complex (2a) (entry
9). The addition of Na[B{3,5-(CF3)2-C6H3}4] to the chloror-
hodium catalyst system (2b) caused complete loss of the
activity (entry 10).
With the silica-SMAP-Rh(OMe)(cod) (2a)/Et3N/hexane
system, the catalyst loading could be reduced to 0.1 mol %
(S/C ) 1000) by a slight increase of initial hydrogen pressure
(10 atm), reaction temperature (50 °C), and concentration
(Table 1, entry 3).
Scheme 1. Synthesis of Silica-SMAP and Its Rh Complex
The bond between the P atom of 1 and the Rh atom is
fairly robust. Leaching of Rh after the reaction (Table 1,
entries 1-10) was not visually observed, and the silica-
SMAP-Rh catalysts could be removed by simple filtration.10
(6) (a) Hamasaka, G.; Ochida, A.; Hara, K.; Sawamura, M. Angew.
Chem., Int. Ed. 2007, 46, 5381–5383. (b) Hamasaka, G.; Kawamorita, S.;
Ochida, A.; Akiyama, R.; Hara, K.; Fukuoka, A.; Asakura, K.; Chun, W. J.;
Ohmiya, H.; Sawamura, M. Submitted for publication.
(7) For a related study with a self-assembled monolayer of SMAP on
gold surface, see: Hara, K.; Akiyama, R.; Takakusagi, S.; Uosaki, K.;
Yoshino, T.; Kagi, H.; Sawamura, M. Angew. Chem., Int. Ed. 2008, 47,
5627–5630.
hydrogen pressure in hexane, reaching quantitative conver-
sion to the corresponding alcohol in 10 h (Table 1, entry 1).
Addition of a subcatalytic amount (0.1 mol %) of Et3N
caused significant rate enhancement; the reaction time was
shortened to 3 h (entry 2).9
(8) Silica-SMAP (1) was prepared with CARiACT Q-10 silica gel (Fuji
Silysia Chemical Ltd.). Rh complex 2a (yellow powder) was prepared
according to the reported procedure for a preparation of the corresponding
chloro complex (2b) (see ref 6).
(9) A mechanism for the rate enhancement with Et3N is unclear.
4698
Org. Lett., Vol. 10, No. 20, 2008

Rh, Rh/P 1:1) with 5 atm of initial hydrogenation pressure
at rt (entry 11).
Table 2. Hydrogenation of Nonfunctionalized Ketones
Catalyzed by the Silica-SMAP-Rh System
Next, we applied the Rh-silica SMAP catalytic system
(silica-SMAP-Rh(OMe)(cod) (2a)/Et3N in hexane) to the
hydrogenation of various ketones with simple hydrocarbon
substituents. Results obtained with 0.5 mol % of catalyst
loading and 5 atm of initial hydrogen pressure at rt are
summarized in Table 2. The reaction of less-hindered ketones
such as cyclohexanone (3b), diethyl ketone (3c), and dipropyl
ketone (3d) required shorter reaction times (1-1.5 hs) for
completion as compared with the reaction of diisopropyl
ketone (3a) (Table 2, entries 1-3). Acyclic, tert-butyl methyl
ketone (3e) and caged ketone 2-adamantanone (3f) were
quantitatively converted to the corresponding alcohols in 10
and 8 h, respectively (entries 4 and 5). Hindered substituted
cyclohexanone 2-(tert-butyl)cyclohexanone (3g) and bicyclic
ketone fenchone (3h) were also hydrogenated smoothly (12
h, 100% and 16 h, 93%, respectively) albeit with low
stereoselectivities (cis/trans, 55:45 and endo/exo, 68:32,
respectively) (entries 6 and 7).
Extremely congested di-tert-butyl ketone (3i) did not
undergo hydrogenation under the nearly ambient conditions
(Table 2, entry 8), but complete conversion to di-tert-
butylmethanol was accomplished under more forcing condi-
tions (0.5 mol % Rh, 30 atm of initial hydrogen pressure,
100 °C, 48 h) (Scheme 2).
Scheme 2
.
Hydrogenation of Di-tert-butyl Ketone (3i) Catalyzed
by the Silica-SMAP-Rh System
a Determined by 1H NMR spectroscopy using 1,1,2,2-tetrachloroethane
as an internal standard.
Results of the hydrogenation of functionalized ketones
catalyzed by 2a/Et3N system are summarized in Figure 1.
While functional groups such as ester, methoxy group, and
the tertiary amino group in 3j-m caused a slight retarding
effect for hydrogenation, bulky steroidal ketone 3n with a
secondary hydroxy group was hydrogenated smoothly with
good diastereoselectivity (R/ꢀ; 92:8). Unfortunately, how-
ever, no reaction was observed with pyridyl ketones 3o-q
and sulfide-functionalized ketone 3r at least under the nearly
ambient conditions. Overall, it seems that strongly coordinat-
ing substrates inhibit the hydrogenation catalyzed by the
silica-SMAP-Rh system, contrasting with a general trend
in metal-catalyzed hydrogenations that substrate chelation
is favored for faster reaction.
As reported previously, Ph-SMAP is a specifically effec-
tive ligand among soluble phosphine ligands for the hydro-
genation of sterically hindered ketones.5c Namely, the
hydrogenation of diisopropyl ketone (3a) proceeded in the
presence of a catalyst prepared in situ from [{Rh(OMe)-
(cod)}2] and Ph-SMAP (1 mol % Rh, Rh/P 1:1) under 30
atm of hydrogen pressure at 100 °C in THF, showing 46%
conversion after 20 h (Table 1, entry 12). In contrast, the
hydrogenation of 3a showed only a trace of conversion when
Me3P and Bu3P were employed as a ligand (Rh/P 1:1) under
the otherwise identical conditions (entries 13 and 14), and
did not take place at all with (t-Bu)3P, PPh3 (Rh/P 1:1) and
dppe (Rh/P 1:2) (entries 15-17). Nevertheless, the ligand
effect of the soluble phosphine Ph-SMAP is much smaller
than that of the supported phosphine silica-SMAP (1); the
effect of the anchoring of SMAP on the solid support is
critical for the hydrogenation activity at nearly ambient
conditions. Indeed, no hydrogenation of 3a occurred with
the [{Rh(OMe)(cod)}2]/Ph-SMAP catalyst system (1 mol %
The high activity of the silica-SMAP-Rh systems toward
the hindered ketones would be attributable to the sparse
(10) Reusability of the silica-SMAP-Rh system has not been examined
in the present study because of the expected difficulty in handling highly
active catalytic species under the hydrogenation conditions in the laboratory
experiments. Instead, the silica-SMAP-Rh catalyst system that was
generated from chloro complex 2b displayed excellent reusability in the
hydrosilylation of ketones (see, ref 6).
Org. Lett., Vol. 10, No. 20, 2008
4699

upon hydrogenation prior to entering a catalytic cycle. Strong
coordination of ketonic substrates through chelation and
coordination of multiple substrates may prevent the formation
of catalytically active species.
In summary, our studies identified the heterogeneous
catalyst system composed of silica-supported, caged, compact
phosphine silica-SMAP and rhodium(I)-alkoxo complex
silica-SMAP-Rh(OMe)(cod) to be a useful catalyst precur-
sor that initiates the hydrogenation of sterically hindered
ketones with unprecedented high efficiency. Studies to
expand applicability of the present method are underway.
Acknowledgment. This work was supported by the
PRESTO program (JST), Grant-in-Aid for Scientific Re-
search (B) (No. 18350047) (JSPS) and Grant-in-Aid for
Scientific Research on Priority Area (No. 20037003, “Chem-
istry of Concerto Catalysis”) (MEXT). G.H. was supported
by MEXT program, Initiatives for Attractive Education in
Graduate Schools (T-type Chemists with Lofty Ambition)
and GCOE (Catalysis as the Basis for Innovation in Materials
Science).
Figure 1. Results (yields and selectivities) of the hydrogenation of
functionalized ketones catalyzed by 2a. Conditions: 3, 1.0 mmol;
2a, 0.5 mol % Rh; Et3N, 0.1 mol %; hexane, 1.0 mL; initial
hydrogen pressure, 5 atm; 23 °C; 10 h unless otherwise noted in
the parentheses.
Supporting Information Available: Experimental pro-
cedures and NMR spectra for the hydrogenation products.
This material is available free of charge via the Internet at
http://pubs.acs.org.
nature of the Rh catalytic center coordinated with only a
single phosphine ligand with extreme compactness, which
is favorable for the access of the bulky substrates; the
catalytic center should lose the 1,5-cyclooctadiene ligand
OL801996R
4700
Org. Lett., Vol. 10, No. 20, 2008