The asymmetric Meerwein - Schmidt - Ponndorf - Verley reduction of prochiral ketones with iPrOH catalyzed by Al catalysts

Article abstract of DOI:10.1002/1521-3773(20020315)41:6<1020::AID-ANIE1020>3.0.CO;2-S

Efficient and practical: An aluminum-based catalyst system for the asymmetric Meerwein - Schmidt - Ponndorf - Verley reduction of aromatic ketones is reported. By using iPrOH as the achiral hydride source and BINOL-type chiral ligands (see scheme), chiral secondary aromatic alcohols (up to 83% ee) can be obtained in good yields.

Full text of DOI:10.1002/1521-3773(20020315)41:6<1020::AID-ANIE1020>3.0.CO;2-S

COMMUNICATIONS
AlMe3 in a 1:1 ratio in toluene, and observed a white
precipitate after five minutes. With the addition of a prochiral
ketone (10-fold excess) and iPrOH (40-fold excess), the
reaction medium became homogeneous [Eq. (1)ꢀ. This simple
method was found to effect the catalytic reduction of
2-chloroacetophenone at room temperature to give (R)-a-
chloromethylbenzyl alcohol in 80% ee and 99% yield [see
Eq. (2); Table 1, entry 1ꢀ. Similarly, using (S)-(À)-BINOL as
The Asymmetric Meerwein ± Schmidt ±
Ponndorf ± Verley Reduction of Prochiral
Ketones with iPrOH Catalyzed by Al
Catalysts**
E. Joseph Campbell, Hongying Zhou, and
SonBinh T. Nguyen*
The Meerwein ± Schmidt ± Ponndorf ± Verley (MSPV) re-
duction of carbonyl substrates to alcohols was first reported
[
1±3ꢀ
O
OH
*
over seventy years ago.
It is highly chemoselective and
OH cat. (10 mol%)
O
(2)
utilizes iPrOH, an inexpensive hydride source.[ However,
4ꢀ
R +
R
+
since the late 1950s, the classical protocol for this reduction,
where Al(OiPr) was used as a reagent, has been largely
3
replaced by methods using boro- and aluminum hydrides.[
This decrease in the application of the classical MSPV
reduction is partly because greater than stoichiometric
5, 6ꢀ
Table 1. Asymmetric MSPV reduction.[aꢀ
Entry
R
2-Propanol
equivalents[bꢀ)
Product
Yield [%ꢀ ee [%ꢀ
(
amounts of Al(OiPr) are required to obtain satisfactory
3
[cꢀ
1
CH
2
Cl
4
4
4
4
5
4
5
4
15
4
4
5
4
4
99
99
99
30
80
32
35
20
46
54
58
80
95
41
43
80 (R)
80 (S)
yields of the alcohol in a reasonable time.[ Recently, we
demonstrated that the catalytic behavior of the MSPV
reduction is highly dependent upon the aggregation state of
the aluminum complex and if freshly made, low-aggregate
aluminum alkoxides are used as catalysts, high yields of the
alcohol can be achieved under mild reaction conditions.[
The asymmetric MSPV-type reduction has long been
4ꢀ
[dꢀ
83 (S)[dꢀ
2
3
2
CH Br
CH
50 (R)[
46 (R)
53 (S)[
35 (S)[
cꢀ
2
CH
3
[
cꢀ
1
1
dꢀ
dꢀ
4
5
6
CH
2
CH(CH
3 2
)
7ꢀ
[dꢀ
dꢀ
CH(CH
CH3
3
)
2
61 (S)
50 (S)[
[
8, 9ꢀ
[10ꢀ
30 (R)[cꢀ
prominent in synthetic methodology
and the Evans
28 (S)[
25 (R)[
dꢀ
and Noyori[
11ꢀ
groups have further improved its utility by
cꢀ
1
demonstrating the stereoselective reduction of prochiral
ketones using chiral variants of this reaction. These reaction
[cꢀ
7
8
CH
2
OCH
acetonaphthone[eꢀ
3
8 (R)
48 (S)[dꢀ
46 (S)[
dꢀ
variants utilize an achiral hydride source, iPrOH; however,
15
they employ chiral transition metal catalysts.[
10, 11ꢀ
To date, the
[
aꢀ Reaction conditions: BINOL (0.02 mmol), AlMe
3
(0.02 mmol), ketone
only examples of asymmetric MSPV reduction using alumi-
num-based catalysts require the use of stoichiometric
(0.20 mmol), and toluene (500 mL); room temperature, N2 for 16 h.
[bꢀ based on substrate. [cꢀ R isomer obtained from reactions with (R)-
(‡)-BINOL. [dꢀ S isomer obtained from reaction with (S)-(À)-BINOL.
amounts of chiral alcohols as the hydride source.[
7, 12, 13ꢀ
[
eꢀ acetonaphthone is the whole reactant, not an R group (not shown in
Herein, we report a practical, enantioselective, catalytic
MSPV reduction that utilizes iPrOH as the hydride source
Eq. (2)).
and is catalyzed by AlMe and enantiopure 2,2×-dihydroxy-
3
1
,1×-biphenyl [BINOL; see Eq. (1)ꢀ.
the ligand produced the (S)-isomer (80% ee, 99% yield). The
enantioselective reduction of 2-bromoacetophenone also
proceeds smoothly (Table 1, entry 2). The reduction of the
more electron-rich substrate propiophenone can also be
achieved in high yield (80%) and moderate ee value (46%)
by the addition of excess iPrOH (15-fold) (Table 1, entry 3).
In general, substrates that are more sterically hindered than
propiophenone lead to lower yields of the product (Table 1,
entries 4 and 5), while a substrate that is less hindered gives a
better yield but a lower ee value (Table 1, entry 6). As in the
case of propiophenone (see above), a higher yield of the
products can be obtained with increased iPrOH loading.
Although a-methoxyacetophenone gives an excellent yield of
the desired alcohol, the enantioselectivity is drastically
reduced (Table 1, entry 7), probably because of the disruption
of the tetradentate chiral intermediate (caused by the extra
coordination of the MeO group, see below). Acetonaphthone
On the basis of earlier precedents in aluminum aryloxide
chemistry,[
14±16ꢀ
we decided to combine (R)-(‡)-BINOL and
OH
OH
O
+
R"
OH
OH
*
O
(
10 mol%)
(1)
+
R'
AlMe3 (10 mol%)
toluene
R'
R"
[
*ꢀ Prof. Dr. S. T. Nguyen, Dr. E. J. Campbell, Dr. H. Zhou
Department of Chemistry
Northwestern University
2
145 Sheridan Road, Evanston, IL 60208-3113 (USA)
Fax : (‡1)847-467-5123
E-mail: stn@chem.northwestern.edu
[
**ꢀ Support from the Dupont Company and the Beckman, Dreyfus, and
Packard Foundations is gratefully acknowledged. S.T.N. is an Alfred P.
Sloan research fellow. E.J.C. acknowledges the GEM and IMGIP
fellowship programs for financial support. We thank Mr. Mark Staples
for helpful discussion.
(
see Table 1, entry 8), which possesses a larger aromatic group
than acetophenone, leads to higher enantioselectivity (Ta-
ble 1, entry 8).
Given that the ee value was greatly increased by the
presence of electron-withdrawing groups on the a-carbon of
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
1020
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1020 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 6

COMMUNICATIONS
the acetophenone moiety, we decided to investigate the effect
of inductive electronic parameters on the reaction shown in
Equation (1). As expected, decreased electron density on the
aromatic ring increases the yield of the desired alcohol.[
However, no significant change in ee value was observed
reduction of prochiral ketones can be accomplished using
simple aluminum-based catalysts and an achiral hydride
source. We were curious about the role of BINOL in this
reaction and set out to investigate the effect of the ligand
concentration and ligand binding mode, by exploring alumi-
num complexes of ligands 1 ± 3 (Scheme 1).
17, 18ꢀ
(
Table 2).
The addition of excess BINOL to the reaction in Equa-
tion (1), in which the substrate is acetophenone, does not
affect the ee value but is detrimental to the product yield. A
Table 2. Electronic effect on the MSPV reduction.[aꢀ
O
OH
*
OH
cat. (10 mol%)
O
1.5:1 ratio of BINOL:AlMe decreases the product yield by
3
(3)
+
+
half and a 2:1 ratio of BINOL:AlMe prevents the reaction
3
X
X
completely. The use of ligand 3 produces a very low yield of
the desired alcohol (Table 3). Other primary amines, such as
2-aminopyridine and 1-(1-naphthyl)ethylamine, were also
found to prevent the reaction completely. In comparison to
BINOL, ligand 1 maintains the same product yield, however
there is a marked decrease in the ee value (Table 3). Increasing
Entry
X
2-Propanol
equivalents[bꢀ)
Product
Yield [%ꢀ
(
ee [%ꢀ
30 (R)[
cꢀ
cꢀ
cꢀ
1
2
3
H
4
54
80
44
62
55
55
7
1
1
1
5
4
5
4
5
4
4
25 (R)[
[
CH
F
3
30 (R)
20 (S)
[
dꢀ
cꢀ
the 1:AlMe ratio to 2:1 diminishes the yield of the product but
3
30 (R)[
30 (R)[
cꢀ
cꢀ
cꢀ
does not further affect the ee value of the reaction. Interest-
30i n( gly however, with the neutral ligand 2, which can not
30d (e protonate before coordination to the aluminum catalyst, the
[
4
5
Cl
Br
0
0
R)
R)[
7
yield and ee value of the product are significantly diminished,
compared to 1 equivalent of BINOL.
[
(
[
aꢀ Reaction conditions: BINOL (0.02 mmol), AlMe
0.20 mmol), and toluene (500 mL); room temperature, N
bꢀ based on substrate. [cꢀ R isomer obtained from reactions with (R)-
3
(0.02 mmol), ketone
2
for 16 h.
(
‡)-BINOL. [dꢀ S isomer obtained from reactions with (S)-(À)-BINOL.
Table 3. Effect of ligand additive on the MSPV reduction.[aꢀ
O
OH
*
A critical characteristic of the MSPV reduction system in
OH
O
ligand, AlMe3
R
+
R
+
(4)
Equation (1) is the favorable equilibrium (both synthetically
and enantioselectively) exhibited by the BINOL ± Al catalyst
for the monoaromatic ketone and iPrOH reagent pairs.
Unlike previous aluminum-based asymmetric MSPV reduc-
R ˆ Me
Ligand Ligand:AlMe Yield ee
R ˆ CH
2
Cl
3
Ligand Ligand:AlMe Yield ee
3
tions, which utilized aromatic chiral alcohols as their hydride
[%ꢀ
[%ꢀ
[%ꢀ
[%ꢀ
sources,[
7, 12ꢀ
the enantiomeric purity of our product is main-
BINOL
BINOL
1:1
2:1
1:1
2:1
4:1
1:1
1:1
54
0
30
±
0
0
0
BINOL
BINOL
1:1
2:1
1:1
2:1
4:1
1:1
1:1
99
0
80
±
15
12
0
tained even after prolonged exposure to the catalyst system
see Supporting Information). After 24 h under catalytic
equilibrium conditions (10 mol% (R)-BINOL, based on
1
1
1
2
3
58
40
20
25
70
1
1
1
2
3
99
90
60
90
25
(
AlMe , 20 equivalents iPrOH, 20 equivalents acetone), the
3
0
10
0
enantiomeric excess of a sample of chirally pure (R)-1-
phenylethanol (98% ee) only decreased to 93%.
The results in Tables 1 and 2 are exciting because they
indicate, for the first time, that catalytic, asymmetric MSPV
[
(
aꢀ Reaction conditions: ligand 1 ± 3 (0.02 mmol), AlMe
0.20 mmol), iPrOH (0.80 mmol), and toluene (500 mL); room temperature, N
3
(0.02 mmol), ketone
2
for 16 h.
O
H
O
O
O
O
1
. AlMe3, Toluene
Ar
R
Al
R
OH
O
O
Al OiPr
OH 2. iPrOH (excess)
Ar
Proposed active catalyst
Proposed transition state for the
stereoselective MSPV reaction.
OMe
OH
OMe
OMe
NH2
NH2
1
2
3
Scheme 1. Proposed active catalyst and transition state for the stereoselective MSPV reaction catalyzed by [(binol)Al(OiPr)ꢀ. Ligands 1, 2, and 3 do not
allow for an optimal MSPV geometry (a tetrahedral, four coordinate transition state) around the aluminum center.
Angew. Chem. Int. Ed. 2002, 41, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1021 $ 17.50+.50/0
1021

COMMUNICATIONS
We reported that tetradentate ligands hinder hydride
transfer by aluminum catalysts.[7ꢀ In the classical MSPV
Scale-up Asymmetric MSPV Reduction: In the drybox, toluene (2 mL) was
added to a 20-mL vial, equipped with a magnetic stirrer bar, and containing
(
S)-(À)-BINOL (57.3 mg, 0.2 mmol). AlMe
3
(19.7 mL, 0.2 mmol) was added
reduction, the aluminum center is believed to be tetracoordi-
nate during the hydride transfer from one of the isopropoxide
ligands to the coordinated ketone substrate.[ We believe that
maximum ee value is achieved in our system when two of the
to the mixture by syringe and the reaction was stirred for 5 min, after which
time a white precipitate had formed. a-Bromoacetophenone (0.40 g,
2.0 mmol) and 2-propanol (610 mL, 8.0 mmol) were added and the vial
was sealed with a teflon-lined cap, taken out of the drybox, and stirred at
room temperature under nitrogen for 16 h. Flash column chromatography
using 230 ± 400 mesh silica gel (Merck, column dimensions ˆ 2.5 cm Â
4ꢀ
aluminum alkyl groups of AlMe are protonated by both
3
phenolic protons, to form a 1:1 metal:BINOL complex during
the initial catalyst formation stage. The third aluminum alkyl
bond is then protonated by iPrOH. We believe this mono-
meric [(binol)(iPrO)Alꢀ complex (Scheme 1) is the active
chiral catalyst for this asymmetric reduction. Indeed, the use
of an isolated sample of [(binol)AlMe(thf)ꢀ, the THF-
solvated precursor of [(binol)(iPrO)Alꢀ, leads to similar
catalytic activity and selectivity in the reduction of a-
bromoacetophenone (95% yield and 79% ee; see Table 1,
entry 2; see Supporting Information).
1
2 cm) and CH
2 2
Cl as eluent gave a pure sample of a-(bromomethyl)benzyl
alcohol, in 93% yield and 79% ee.
Received: December 7, 2001 [Z18351ꢀ
[
1ꢀ H. Meerwein, R. Schmidt, Justus Liebigs Ann. Chem. 1925, 39, 221 ±
38.
[2ꢀ W. Ponndorf, Angew. Chem. 1926, 39, 138 ± 146.
2
[
[
3ꢀ M. Verley, Bull. Soc. Chim. Fr. 1925, 37, 871 ± 874.
4ꢀ C. F. de Graauw, J. A. Peters, H. van Bekkum, J. Huskens, Synthesis
A preferred requirement for the aforementioned model is
that the active catalyst center is tetrahedral during the hydride
1
994, 1007± 1017.
[
5ꢀ A. Hajos, Complex Hydrides and Related Reducing Agents in Organic
Synthesis, Elsevier, Amsterdam, 1979.
À
transfer. Accounting for the necessary iPrO ion and
[
[
6ꢀ B. Knauer, K. Krohn, Liebigs Ann. 1995, 677 ± 683.
7ꢀ E. J. Campbell, H. Zhou, S. T. Nguyen, Org. Lett. 2001, 3, 2391 ±
the carbonyl substrate, this model leaves only two open
coordination sites for any ligand to occupy. We believe that
any catalytic system in which the ligands coordinate to the
metal center so as to occupy more than two coordination sites
will reduce the reaction rate. The diminished activity observed
when excess bidentate ligand is used (see above) supports this
theory.
Increased loading of iPrOH shifts the equilibrium to a
higher yield for the desired product, but also decreases the
ee value (Tables 1, entries 3 ± 6). This effect may be explained
by a decrease in the effective bidentate coordination of
BINOL by competing iPrOH coordination, thereby decreas-
ing the ee value. This is a minimal factor, however, because
changing from four to over 15 equivalents of iPrOH de-
creased the ee value at most by 5 ± 15%, while increasing the
yield by over 50% for propiophenone (Table 1, entry 3).
In summary, we have observed for the first time an
asymmetric MSPV reduction that utilizes an achiral hydride
source and a chiral aluminum alkoxide catalyst. A high yield
was achieved using a variety of ketone substrates and a good
ee value was obtained when the carbonyl a-carbon was
sufficiently electron withdrawing.
2
393.
8ꢀ R. B. Woodward, N. L. Wendler, F. J. Brutschy, J. Am. Chem. Soc.
945, 67, 1425 ± 1429.
[9ꢀ W. v. E. Doering, T. C. Aschner, J. Am. Chem. Soc. 1953, 75, 393 ±
97.
[
1
3
[
[
10ꢀ D. A. Evans, S. G. Nelson, M. R. Gagne, A. R. Muci, J. Am. Chem.
Soc. 1993, 115, 9800 ± 9801.
11ꢀ K.-J. Haack, S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew.
Chem. 1997, 109, 297± 300; Angew. Chem. Int. Ed. Engl. 1997, 36, 285 ±
2
88.
[
12ꢀ T. Ooi, T. Miura, K. Maruoka, Angew. Chem. 1998, 110, 2524 ± 2526;
Angew. Chem. Int. Ed. 1998, 37, 2347± 2349.
[
13ꢀ M. Fujita, Y. Takarada, T. Sugimura, A. Tai, Chem. Commun. 1997,
1
631 ± 1632.
[14ꢀ I. P. Fu, B. J. Uang, Tetrahedron Asymmetry 2001, 12, 45 ± 48.
[
15ꢀ K. Ishihara, J. Kobayashi, K. Inanaga, H. Yamamoto, Synlett 2001,
94 ± 396.
16ꢀ T. Ooi, H. Ichikawa, K. Maruoka, Angew. Chem. 2001, 113, 3722 ±
724; Angew. Chem. Int. Ed. 2001, 40, 3610 ± 3612.
[17ꢀ H. Adkins, F. W. Cox, J. Am. Chem. Soc. 1938, 60, 1151 ± 1159.
3
[
3
[
18ꢀ H. Adkins, R. M. Elofson, A. G. Rossow, C. C. Robinson, J. Am.
Chem. Soc. 1949, 71, 3622 ± 3629.
Experimental Section
Available in the Supporting Information: General experimental procedure,
quantitative analysis (including chiral gas chromatography (GC) traces),
plots of conversion and ee value versus time for the reduction of a-
bromoacetophenone, and the characterization of the proposed active chiral
aluminum alkoxide catalyst (9 pages).
General MSPV Reduction Procedure: All reactions were carried out under
a dry nitrogen atmosphere unless otherwise noted. Toluene (500 mL) and
enantiopure BINOL (5.8 mg, 0.02 mmol) were added to a 4-mL vial,
equipped with a magnetic stirrer bar. AlMe (1.9 mL, 0.02 mmol) was added
3
to the mixture by syringe and the reaction was stirred for 5 min, after which
time a white precipitate had formed. The carbonyl substrate (10 equiv.) and
2
-propanol (40 or 150 equiv.) were added, and the vial was sealed with a
teflon-lined silicone septum. The reaction was stirred at room temperature
under nitrogen for 16 h. Aliquots (20 mL) were passed through a plug of
neutral aluminum oxide (activated, Brockmann activity 1, ꢀ150 mesh) and
analyzed by GC to determine selectivity and conversion data (see the
Supporting Information).
1022
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1022 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 6