Asymmetric hydrogenation of tert-alkyl ketones: DMSO effect in unification of stereoisomeric ruthenium complexes

Article abstract of DOI:10.1002/anie.201304408

Face off: The ruthenium complexes of a new axially chiral PNNligand (L) are highly efficient in the presence of dimethylsulfoxide (DMSO) for hydrogenation of both functionalized and unfunctionalized tert-alkyl ketones. DMSO is thought to narrow down the many possible complex stereoisomers into a single facial L/Ru complex, thus enhancing the reactivity, selectivity, and productivity. Copyright

Full text of DOI:10.1002/anie.201304408

Angewandte
Chemie
Communications
Synthetic Methods
T. Yamamura, H. Nakatsuka, S. Tanaka,
M. Kitamura*
&&&& — &&&&
Asymmetric Hydrogenation of tert-Alkyl
Ketones: DMSO Effect in Unification of
Stereoisomeric Ruthenium Complexes
Face off: The ruthenium complexes of
a new axially chiral PNN ligand (L) are
highly efficient in the presence of dime-
thylsulfoxide (DMSO) for hydrogenation
of both functionalized and unfunctional-
ized tert-alkyl ketones. DMSO is thought
to narrow down the many possible com-
plex stereoisomers into a single facial L/
Ru complex, thus enhancing the reactiv-
ity, selectivity, and productivity.
Angew. Chem. Int. Ed. 2013, 52, 1 – 4
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
DOI: 10.1002/anie.201304408
Synthetic Methods
Asymmetric Hydrogenation of tert-Alkyl Ketones: DMSO Effect in
Unification of Stereoisomeric Ruthenium Complexes**
Tomoya Yamamura, Hiroshi Nakatsuka, Shinji Tanaka, and Masato Kitamura*
Noyori and co-workers revolutionized asymmetric hydro-
genation of functionalized ketones in 1987 through the
invention of binap/Ru(OCOCH3)2/HCl (binap = 2,2’-bis(di-
phenylphosphanyl)-1,1’-binaphthyl),[1] thereby establishing
the leading concept of a soft transition metal/hard Brønsted
acid combined catalyst or an intermolecular donor–acceptor
bifunctional catalyst (Intermol DACat).[2] Subsequent devel-
opment of the intramolecular version (Intramol DACat),
a binap/Ru/diamine ternary catalyst, expanded the substrate
scope to aromatic and sterically bulky unfunctionalized
ketones.[3] Complementary use of the two binap/Ru methods
covers almost all types of ketonic substrates except for
functionalized tert-alkyl ketones.[4] The ternary catalyst is
thought to lose its Intramol DACat ability when the diamine
moiety is replaced with a chelatable functionalized ketone,
thereby limiting substrate generality to simple ketones.
With this drawback in mind, we designed the ligand 2’-
(diphenylphosphino)-N-(pyridine-2-ylmethyl)-[1,1’-binaph-
thalen]-2-amine (binan-Py-PPh2; 1), in which one of the PPh2
as the standard reaction. This reaction is catalyzed neither by
the binap/Ru/HCl system nor by the binap/Ru/diamine
ternary system. The results are shown in Table 1. The starting
conditions of 1m 2a, 2 mm (R)-4, 10 mm tBuOK, 100 atm H2,
Table 1: Asymmetric hydrogenation of methyl 4,4-dimethyl-3-oxopenta-
noate (2a) using (R)-4.[a]
Entry (R)-4 [mm] tBuOK [mm] DMSO [mm] Yield [%][b] S/R[c]
1
2
3
2
2
2
1
1
1
0.5
–
10
10
10
10
20
20
30
10
3[d]
100
22
74
85:15
99:1
99:1
99:1
99:1
1:99
98:2
98:2
1400[e]
1400[e]
1400[e]
1400[e]
1400[e]
1400[e]
>99
>99
>99
>99
>99
1
4[f]
5
6[g]
7[h]
8[i]
[a] All of the reactions were carried out in CH3OH at RT for 24 h under
100 atm H2 atmosphere unless otherwise specified. [b] Determined by
1H NMR analysis. [c] Determined by HPLC analysis. [d] Complete
removal of DMSO from 4 was impossible. [e] CH3OH/DMSO=9:1.
[f] 140 atm H2. [g] (S)-4 was used. [h] 508C. [i] [{RuCl2(cod)}n]=[(R)-
1]=2 mm. cod=cyclo-1,5-octadiene, DMSO=dimethylsulfoxide.
CH3OH, RT (25–288C) for 24 h afforded 3a with an S/R
enantiomeric ratio (e.r.) of 85:15 in 22% yield (entry 1).
Addition of DMSO (100 mm) to this system increased the
reactivity by about threefold to give (S)-3a with 99:1 e.r.
(entry 2), and a further increase in the DMSO concentration
to 1400 mm led to quantitative conversion of 2a into (S)-3a
with 99:1 e.r. (entry 3). The concentration of the catalyst (R)-
4 could be reduced to 1 mm [substrate/catalyst (S/C) = 1000]
either at 140 atm of H2 (entry 4) or with the concentration of
base doubled (20 mm tBuOK; entry 5). The enantiomer of the
catalyst [(S)-4] gave (R)-3a (entry 6). A further decrease in
(R)-4 concentration to 0.5 mm (S/C = 2000) required a tBuOK
concentration of 30 mm for completion of the reaction within
24 hours (entry 7), and with the b-keto ester substrate 2a, an
S/C ratio of 5000 was the limit.[8] The [{RuCl2(cod)}n]/(R)-
1 system was not as efficient as (R)-4 even in the presence of
DMSO (1400 mm; entry 8). Addition of P(OCH3)3, CO,
P(CH3)3, N(C2H5)3, or pyridine (each 100 mm) instead of
DMSO stopped the reaction.[8] CH3OH was the solvent of
choice, and the reactivity in C2H5OH decreased at least by
one order, although the high enantioselectivity was main-
tained. Little reaction occurred in iPrOH, tBuOH, DMSO,
THF, CH2Cl2, or toluene,[8] and tBuOK was the best choice
because tBuOLi hardly produced 3a.[8]
groups of binap is replaced with 2-PyCH2NH, which is an
[3c,5]
=
excellent group for C O hydrogenation.
The non-pincer-
type ligand, characterized by axial chirality, flexibility, and
a linearly arranged Psp3Nsp2Nsp3 system,[6] was prepared in 78%
=
total yield by Staudinger-type reaction/hydrolysis/P O reduc-
tion starting from a known compound, 2’-(diphenylphos-
phino)-[1,1’-binaphthalen]-2-yl trifluoromethanesulfonate.[7,8]
To evaluate the utility of the ruthenium complex of (R)-1,
[Ru((R)-1)(dmso)3](BF4)2 [(R)-4], asymmetric hydrogenation
of the C3-tBu-substituted b-keto ester 2a to 3a was selected
[*] T. Yamamura, H. Nakatsuka, Prof. Dr. M. Kitamura
Graduate School of Pharmaceutical Sciences
Graduate School of Science, Nagoya University
Chikusa, Nagoya 464-8601 (Japan)
E-mail: kitamura@os.rcms.nagoya-u.ac.jp
S. Tanaka, Prof. Dr. M. Kitamura
Research Center for Materials Science, Nagoya University (Japan)
[**] This work was supported by a Grant-in-Aid for Scientific Research
(No. 25E07B212 and 23005914) from the Ministry of Education,
Culture, Sports, Science and Technology (Japan).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304408.
2
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 4
These are not the final page numbers!

Angewandte
Chemie
The scope and limitations of the present asymmetric
hydrogenation are shown in Table 2. The alkoxycarbonyl
group of 2a can be replaced with an aminocarbonyl,
dialkoxyphosphonyl or dimethylamino group (entries 1–4).
Introduction of an alkoxycarbonyl or hydroxymethyl group
into the tert-alkyl part is acceptable (entries 5 and 6). Double
hydrogenation of the 1,3-diketone 2g afforded the corre-
sponding chiral diol 3g with a high e.r. value by using a meso
trick, in which the minor enantiomer of the first hydro-
genation is scavenged to the meso product at the second stage
(entry 7).[1b] Not only functionalized ketones but also unfunc-
tionalized simple tert-alkyl ketones can be used as substrates.
Pinacolone (2h) was quantitatively hydrogenated in the
presence of 0.01 mol% of (R)-4 (S/C = 10000) in the same
CH3OH/DMSO solvent system to give (S)-3h with 98:2 e.r.
(entry 8). Various primary alkyl tert-butyl ketones (2i–k) can
be hydrogenated to the secondary alcohols with 99:1 e.r.
(entries 9–11). Adamantyl and methyl groups are also tol-
erated by (R)-4 (entry 12). With a,a-dimethyl cyclic ketones,
the enantioselectivity tends to decrease (entries 13–15), and
a mesityl group also acts as a bulky substituent, thus giving
(S)-1-mesitylethanol
[(S)-2,4,6-(CH3)3C6H2CH(OH)CH3]
with 83:17 e.r. at 508C in 95% yield upon isolation.[8]
Acetophenone and tert-butyl phenyl ketone were not the
substrate of choice.[8]
As can be seen from the stereochemical outcomes shown
in Table 2, in all cases using the ligand (R)-1, a hydrogen
molecule is formally delivered from the upper side, when the
ketone structures are drawn with the larger tert-alkyl group on
the right and with the smaller substituent on the left. We
assume that the present hydrogenation proceeds by a Noyori
mechanism (Figure 1).[3b,5a,9,10] Here, the polarized H-Ru···N-
Table 2: Asymmetric hydrogenation of tert-alkyl ketones catalyzed by the
(R)-4.[a]
Entry
Substrate
Product
Yield [%][b]
(S/R)[c]
1[d]
2
2a: R1 =COOCH3, R2 =CH3
2b: R1 =CON(CH3)2, R2 =CH3
2c: R1 =PO(OCH3)2, R2 =CH3
2d: R1 =N(CH3)2, R2 =CH3
2e: R1 =H, R2 =COOCH3
2 f: R1 =H, R2 =CH2OH
3a
3b
3c
3d
3e
3 f
98 (99:1)
97 (92:8)
97 (2:98)
91 (1:99)
97 (96:4)
97 (89:11)
3
4[e]
5[f]
6
7[g]
2g
3g
95 (99:1)[h]
8[i]
9
10
11
2h: R=H
3h
3i
3j
99[j] (98:2)
97 (99:1)
99 (99:1)
95 (99:1)
Figure 1. Proposed transition states.
2i: R=CH2CH3
2j: R=n-C7H15
2k: R=CH2C6H5
3k
H four-atom system in 5, generated from 4 probably via
a ruthenium amide species, and subsequent heterolytic
=
cleavage of H2 captures a C O substrate to give TSS or TSR.
The Intramol DACat ability allows quick and simultaneous
12[f]
2l
3l
99 (99:1)
=
delivery of hydrogen atoms to Nsp3 and Ru to the C O double
bond to give the ruthenium amide species by liberation of the
ꢀ
alcoholic product. The hydrogenolysis of the Ru N bond
regenerates 5. The catalyst (R)-4 was quantitatively generated
by treatment with the structurally unambiguous fac-[Ru((R)-
13
14
2m: n=1
2m: n=2
3m
3n
97 (76:24)
94 (89:11)
[11]
1)(C6H6)](BF4)2
in DMSO at 508C for 48 hours. The
1H NMR analysis of (R)-4 in [D6]DMSO indicates that
1) the three Psp3Nsp2Nsp3 coordinating atoms are arranged in
a facial manner (7.0% NOE between C(8)H of the naphtha-
lene ring and NCHRHSPy), and 2) the conformation of
DMSO trans to Nsp3 is fixed by a hydrogen bond between
15
2o
3o
97 (5:95)
[a] Conditions unless otherwise specified: [Substrate]=1m; [(R)-4)]=
1 mm; [tBuOK]=10 mm; [DMSO]=1400 mm; CH3OH; 100 atm H2; RT;
24 h. [b] Yield of isolated product. In all cases, the conversion was greater
than 99%. [c] Determined by GC or HPLC analysis.[8] [d] [tBuOK]=
20 mm. [e] 508C. [f] 48 h. [g] 72 h. [h] dl/meso 91:9, determined by
1H NMR analysis.[8] [i] [(R)-4]=0.01 mm. [j] Determined by GC analysis
using mesitylene as an internal standard.[8]
=
S O and C(6)H of the pyridine moiety to make O-S···Ru-
2
=
sp N C6 in the plane (C(6)H of (R)-4: d 9.74 versus C(6)H of
fac-[Ru((R)-1)(C6H6)](BF4)2: d = 9.17).[8] Treatment of (R)-4
with 2 mol amounts of tBuOK in 5:1 DMSO/CH3OH (10 mm)
at 258C gives a hydride methoxide complex,[12] which may act
as a repository for the reactive ruthenium amide or may react
Angew. Chem. Int. Ed. 2013, 52, 1 – 4
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
directly with H2 to give 5.[8] In the transition states TSS and
TSR, TSR suffers from steric repulsion between the methyl
group of a coordinated DMSO and the tert-alkyl substituent
Kumobayashi, S. Akutagawa, T. Ohta, H. Takaya, R. Noyori, J.
Am. Chem. Soc. 1988, 110, 629 – 631.
[2] a) M. Kitamura, H. Nakatsuka, Chem. Commun. 2011, 47, 842 –
846; b) R. Noyori, M. Kitamura, T. Ohkuma, Proc. Natl. Acad.
Sci. USA 2004, 101, 5356 – 5362. For the original reaction
establishing DACat concept, see: c) M. Kitamura, S. Suga, K.
Kawai, R. Noyori, J. Am. Chem. Soc. 1986, 108, 6071 – 6072;
d) R. Noyori, M. Kitamura, Angew. Chem. 1991, 103, 34 – 55;
Angew. Chem. Int. Ed. Engl. 1991, 30, 49 – 69.
[3] a) T. Ohkuma, H. Ooka, S. Hashiguchi, T. Ikariya, R. Noyori, J.
Am. Chem. Soc. 1995, 117, 2675 – 2676; b) R. Noyori, T.
Ohkuma, Angew. Chem. 2001, 113, 40 – 75; Angew. Chem. Int.
Ed. 2001, 40, 40 – 73; c) T. Ohkuma, C. A. Sandoval, R.
Srinivasan, Q. Lin, Y. Wei, K. MuÇiz, R. Noyori, J. Am. Chem.
Soc. 2005, 127, 8288 – 8289.
[4] a) M. Kitamura, R. Noyori in Ruthenium in Organic Synthesis
(Ed.: S.-I. Murahashi), Wiley-VCH, Weinheim, 2005, pp. 3 – 41;
b) T. Ohkuma, M. Kitamura, R. Noyori in Catalytic Asymmetric
Synthesis, 2nd ed. (Ed.: I. Ojima), Wiley-VCH, New York, 2000,
pp. 1 – 110. Replacement of binap with a modified biphemp
ligand in the ruthenium-catalyzed hydrogenation is reported to
reduce a sterically bulky b-keto ester: c) Y. Sun, X. Wan, M.
Guo, D. Wang, X. Dong, Y. Pan, Z. Zhang, Tetrahedron:
Asymmetry 2004, 15, 2185 – 2188.
[5] a) H. Huang, T. Okuno, K. Tsuda, M. Yoshimura, M. Kitamura,
J. Am. Chem. Soc. 2006, 128, 8716 – 8717. For the iridium-
catalyzed hydrogenation of aromatic ketones, see: b) J.-H. Xie,
X.-Y. Liu, X.-H. Yang, J.-B. Xie, L.-X. Wang, Q.-L. Zhou,
Angew. Chem. 2012, 124, 205 – 207; Angew. Chem. Int. Ed. 2012,
51, 201 – 203; c) J.-H. Xie, X.-Y. Liu, J.-B. Xie, L.-X. Wang, Q.-L.
Zhou, Angew. Chem. 2011, 123, 7467 – 7470; Angew. Chem. Int.
Ed. 2011, 50, 7329 – 7332.
=
on C O, thus leading to the S product (R < tBu) via TSS. This
scheme explains the general rule of enantioselection in this
asymmetric catalysis (Table 2). Elucidation of the detailed
mechanism is ongoing project in our group.
In summary, we have realized the highly enantioselective
hydrogenation of both functionalized and unfunctionalized
tert-alkyl ketones by use of a chiral ruthenium(II) complex of
the axially chiral PNN ligand binan-Py-PPh2, thereby solving
the problem of the asymmetric hydrogenation of functional-
ized tert-alkyl ketones for the first time. The DMSO effect[13]
is important to attain high performance. Combining the
flexible Psp3Nsp2Nsp3 tridentate ligand with DMSO unifies the
many possible stereoisomeric octahedral ruthenium com-
plexes to one PNN facial and Nsp3/DMSO trans species.
=
Furthermore, a PyC(6)H···O S(CH3)2 hydrogen bond fixes
=
the conformation of DMSO on ruthenium to control the C O
enantiosurface selection of the ketone substrates. These
steric, electronic, and orbital factors are synergistically
effected to endow the binan-Py-PPh2/Ru complex with the
highly enantioselective Intramol DACat ability. This success
hints at the development of even higher performance
molecular catalysts.
Experimental Section
[6] Chiral Ru/PNNP-catalyzed transfer hydrogenation and hydro-
genation: a) J.-X. Gao, T. Ikariya, R. Noyori, Organometallics
1996, 15, 1087 – 1089; b) V. Rautenstrauch, X. Hoang-Cong, R.
Churlaud, K. Abdur-Rashid, R. H. Morris, Chem. Eur. J. 2003, 9,
4954 – 4967.
[7] P. N. M. Botman, O. David, A. Amore, J. Dinkelaar, M. T. Vlaar,
K. Goubitz, J. Fraanje, H. Schenk, H. Hiemstra, J. H. van Maar-
seveen, Angew. Chem. 2004, 116, 3553 – 3555; Angew. Chem. Int.
Ed. 2004, 43, 3471 – 3473.
The chemicals for hydrogenation were manipulated in a glove box.
The catalyst precursor [Ru((R)-1)(dmso)3](BF4)2 (10 mm in DMSO,
6.32 mL, 63.2 mmol), CH3OH (35.0 mL), tBuOK (100 mm in CH3OH,
12.6 mL, 1.26 mmol), and methyl 4,4-dimethyl-3-oxopentanoate (2a)
(10.0 g, 9.35 mL, 63.2 mmol) were added to an inner glass tube
containing a Teflon-coated magnetic stirring bar. The tube was then
placed in a 250 mL stainless high-pressure autoclave. After closing the
autoclave, the hydrogenation apparatus was taken out from the glove
box and connected to an H2 cylinder. After three displacements of air
in the introductory pipe by H2 gas, a pressure of 10 atm of H2 gas was
introduced and carefully released. This fill–release cycle was replaced
three times; the autoclave was then pressurized to 100 atm with H2.
The solution was vigorously stirred for 24 h at RT. After careful
release of the H2 gas, the resulting yellow-to-reddish solution was
concentrated under a reduced pressure (ca. 260 mmHg) to give
a crude reaction mixture containing DMSO (25.0 g). This was passed
through a short silica gel column (6 cmfꢀ 20 cm; 250 g; eluent: 2:1 n-
hexane/Et2O). The filtrate was concentrated to give (S)-methyl 3-
hydroxy-4,4-dimethylpentanoate (3a) (9.92 g, 98% yield) with a 99:1
[8] For details, see the Supporting Information.
[9] a) C. A. Sandoval, T. Ohkuma, K. MuÇiz, R. Noyori, J. Am.
Chem. Soc. 2003, 125, 13490 – 13503; b) R. J. Hamilton, C. G.
Leong, G. Bigam, M. Miskolzie, S. H. Bergens, J. Am. Chem. Soc.
2005, 127, 4152 – 4153; c) A. Hadzovic, D. Song, C. M.
MacLaughlin, R. H. Morris, Organometallics 2007, 26, 5987 –
5999.
[10] The mechanism proposed by Milstein and co-workers cannot be
completely excluded. With the N-methyl derivative of (R)-1, the
4-related complex could not be obtained. The N-methyl ligand in
combination with [{RuCl2(cod)}n] under the standard conditions
gave no product: a) J. Zhang, G. Leitus, Y. Ben-David, D.
Milstein, J. Am. Chem. Soc. 2005, 127, 10840 – 10841; b) J. Zhang,
G. Leitus, Y. Ben-David, D. Milstein, Angew. Chem. 2006, 118,
1131 – 1133; Angew. Chem. Int. Ed. 2006, 45, 1113 – 1115.
[11] The structure was determined by X-ray crystallographic analy-
sis.[8]
25
S/R e.r. as a colorless oil (½aꢁD ¼ꢀ35.7 (c = 2.7, C2H5OH)).
Received: May 22, 2013
Published online: && &&, &&&&
Keywords: hydrogenation · ketones · ligand design · ruthenium ·
.
synthetic methods
[12] The NMR analysis implies a PNN facial, Nsp3/DMSO trans, and
Psp3/RuH cis structure.[8]
[13] Review for DMSO complexes: M. Calligaris, Coord. Chem. Rev.
2004, 248, 351 – 375; See also for Ru-DMSO complexes, M.
Toyama, S. Iwamatsu, K.-i. Inoue, N. Nagao, Bull. Chem. Soc.
Jpn. 2010, 83, 1518 – 1527.
[1] a) R. Noyori, T. Ohkuma, M. Kitamura, H. Takaya, N. Sayo, H.
Kumobayashi, S. Akutagawa, J. Am. Chem. Soc. 1987, 109,
5856 – 5858; b) M. Kitamura, T. Ohkuma, S. Inoue, N. Sayo, H.
4
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 1 – 4
These are not the final page numbers!