Enantio- and regioselective iridium-catalyzed allylic hydroxylation

Article abstract of DOI:10.1021/ja109953v

The first enantioselective allylic hydroxylation to prepare branched allylic alcohols directly is described. Bicarbonate was used as nucleophile in conjunction with new single component Ir-catalysts, which are stable to air and water. Excellent regio- and

Full text of DOI:10.1021/ja109953v

COMMUNICATION
pubs.acs.org/JACS
Enantio- and Regioselective Iridium-Catalyzed Allylic Hydroxylation
Martin G€artner, Steffen Mader, Kai Seehafer, and G€unter Helmchen*
Organisch-Chemisches Institut der Ruprecht-Karls-Universit€at Heidelberg, Im Neuenheimer Feld 270, D-69120 Heidelberg, Germany
S Supporting Information
b
Scheme 1. Ir-Catalyzed Allylic Substitution
ABSTRACT: The first enantioselective allylic hydroxylation
to prepare branched allylic alcohols directly is described.
Bicarbonate was used as nucleophile in conjunction with new
single component Ir-catalysts, which are stable to air and
water. Excellent regio- and enantioselectivities have been
achieved with a representative set of substrates.
he transition-metal catalyzed allylic substitution is one of
the most studied approaches for the synthesis of chiral allyl
T
compounds. For many years this field was dominated by Pd-
catalyzed reactions1 of symmetrically disubstituted substrates,
which are often cumbersome to prepare. More recently, enantio-
selective reactions with readily available monosubstituted sub-
strates (Scheme 1) have come into focus. With these, particularly
good results were obtained for allylic alkylations and aminations
upon use of iridium catalysts.2 Less developed are reactions with
O-nucleophiles, because standard reaction conditions working
well for C- and N-nucleophiles are usually not suitable. Never-
theless, procedures have been worked out for enantioselective
allylic reactions with phenolates and alkoxides as well as carboxy-
lates using Ir-,3 Pd-,4 Rh-,5 or Ru-catalysts.6
Scheme 2. Allylic Alcohol As Side Product of an Intramolec-
ular Allylic Aminationa
a TBD = 1,5,7-triazabicyclo[4.4.0]dec-5-ene.
Despite the recent progress, the direct allylic hydroxylation
(Nu=OH), yielding synthetically highly valuable allylic alcohols7
and arguably the most important reaction with O-nucleophiles,
has defied all attempts. Carreira et al. have perhaps most intensely
pursued this topic.8 Their catalyst was generated in situ from
[Ir(cod)Cl]2 (cod = cyclooctadiene), a chiral phosphoramidite
ligand and a base for C-H activation; this in situ catalyst is noto-
riously sensitive to air and moisture. As substitute for the elusive
reaction the Carreira team successfully used a reaction with a
silanolate as nucleophile yielding branched allylic silyl ethers.
In view of the prior art, we are pleased to report the first highly
enantioselective allylic hydroxylation of linear carbonates to
give branched allylic alcohols directly. Our starting point was
the cyclization reaction described in Scheme 2, which shows
strong substrate control. Upon double asymmetric induction, the
matched combination (L*=ent-L2) was fast and gave the pyrro-
lidine B with excellent diastereoselectivity (>98:2) in favor of the
trans-isomer, while the mismatched combination (L* = L2) was
slow, and the diastereomerically pure alcohol C was isolated in up
to 16% yield as a side product. This compound could be formed
by a direct reaction of an intermediary (π-allyl)Ir complex with
H2O or OH-. Another possibility is a reaction with HCO3- anion
Scheme 3. Iridium-Catalyzed Allylic Hydroxylation
kinetic resolution (DYCAT) of racemic allylic carbonates.9
Accordingly, a route according to eq 1 appeared most likely.
In order to follow the lead provided by Scheme 2, hydroxyla-
tion experiments were carried out with the carbonate 1a, which is
-
(eq 1), resulting from the reaction of water with H3COCO2
formed from the leaving group. Gais et al. have shown that
bicarbonate anion is a good nucleophile in Pd-catalyzed dynamic
Received: November 5, 2010
Published: January 28, 2011
r
2011 American Chemical Society
2072
dx.doi.org/10.1021/ja109953v J. Am. Chem. Soc. 2011, 133, 2072–2075
|

Journal of the American Chemical Society
COMMUNICATION
Table 1. Screening According to Scheme 3 using Carbonate 1a As Substratea
entry
catalyst
pronucleophile
H2O f
solvent
temp (°C)
time (h)
2/3b
yield (%)c
ee (%)d
remarks
1
2
in situ/ent-L2e
THF
THF
THF
THF
THF
THF
THF
50
50
50
50
50
50
50
rt
84
20
19
24
0.5
16
17
96
2
n.d.
n.d.
n.d.
-
28
29g
54
-
89
96
97
-
many products
many products
many products
no conversion
many products
clean reaction
many products
clean reaction
clean reaction
clean reaction
clean reaction
clean reaction
clean reaction
clean reaction
clean reaction
in situ/L2e
in situ/L2e
C1
KHCO3
3
KHCO3, 18-crown-6
H2O f
4
5
C1
KHCO3, 18-crown-6
H2O f
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
94:6
98:2
95:5
96:4
78:22
51
65g
38g
67g
84
66
86
89
92
87
48
93
55
80
89
87
93
94
94
94
94
76
6
C3
7
C3
KHCO3, 18-crown-6
KHCO3
8
C3
CH2Cl2/H2O 10:1
CH2Cl2/H2O 10:1
CH2Cl2/H2O 10:1
DMF/H2O 10:1
DMF
9
C3
KHCO3, 18-crown-6
KHCO3, NBu4ClO4
KHCO3
rt
10
11
12
13h
14
15
C3
42
rt
7.5
2
C3
C3
KHCO3
rt
2
C3
KHCO3
DMF/H2O 10:1
DMF/H2O 10:1
DMF/H2O 10:1
rt
1
C2
KHCO3
rt
1
C1
KHCO3
50
21
a Conditions: argon atmosphere, carbonate 1a (1 equiv), pronucleophile (1.7 equiv), catalyst (4 mol %), organic solvent (4 mL/mmol of 1a),
occasionally water (0.4 mL/mmol of 1a). b Determined by 1H NMR of the crude product; n.d. = not determined. c Isolated yield of branched product.
d Determined by HPLC on a chiral column. e In situ preparation: argon atmosphere, [Ir(cod)Cl]2 (2 mol %), L2 (4 mol %), TBD (8 mol %), THF, rt,
5 min. f 2 equiv were used. g Incomplete conversion. h Reaction under air.
prone to give low degrees of selectivity10 and therefore is a sig-
nificant substrate (Scheme 3).
First, a catalyst prepared in situ from ligand L2 was employed
(Table 1, entries 1-3). With water and KHCO3 as pronucleo-
philes (THF, 50 °C), the yield of the hydroxylation product 2a
was poor, probably because of catalyst deactivation caused by
reversibility of the cyclometallation or insolubility of the salt
in THF.11 Considerable improvement with respect to rate was
achieved upon addition of 18-crown-6, however, numerous side
products were also formed. On the other hand, the enantio-
selectivity of the reaction with bicarbonate was high.
Figure 1. (π-Allyl)Ir-complexes, prepared from 1e according to eq 2,
used in this work.
Next, the isolated allyl complex C1 (Figure 1) derived from
L2, cod and carbonate 1e was probed. Complexes of this type
have furnished excellent results in allylic substitutions.12,13a Their
CH2Cl2. The reaction was very slow under these conditions
(entry 8). However, phase transfer in the biphasic mixture with
preparation has become very easy with a direct one-step proce-
dure recently developed by us (eq 2).13b However, the reaction
18-crown-6 was beneficial for the rate, the reaction time was
2 h at room temperature (entry 9), but enantioselectivity was
low; with NBu4ClO4 as additive (entry 10) the situation was
reversed.
The results at that stage indicated that the dbcot complexes are
very robust catalysts tolerating even water as solvent. As a
of 1a with H2O as pronucleophile gave no conversion when C1
was used as catalyst (entry 4). With KHCO3/18-crown-6 as
pronucleophile, hydroxylation product 2a was obtained with
high enantiomeric excess though in moderate yield (entry 5).
It appeared likely that the complex C1 is too sensitive to
withstand an aqueous reaction medium.11 Better stability would
consequence, it was possible to screen a wide variety of solvents
(see Supporting Information). Excellent results were obtained
be expected for corresponding complexes of dbcot (dbcot =
with EtOAc, CH3CN, and acetone, while Et2O, toluene, DMSO,
dibenzocyclooctatetraene) (Scheme 1), which displays stronger
binding to Ir than cod.14 Our preparative procedure (eq 2)
and iPrOH were not satisfactory. Finally, we found DMF/H2O
10:1 particularly well suited, because 18-crown-6 was not re-
quired as additive (entry 11), although it does accelerate the
reaction (see Supporting Information). An excellent regioselec-
tivity of typically 96:4 was found. Control experiments gave
further significant results: (a) Water is not a necessary additive in
case of DMF as solvent (entry 12); (b) Carbonates other than
KHCO3 can be used as pronucleophiles (see Supporting In-
formation); (c) It is possible to carry out reactions without inert
gas atmosphere (entry 13); (d) The complex C2, derived from
the commercially available ligand L1, also gave rise to excellent
regio- and enantioselectivity (entry 14); and (e) Even under the
optimal reaction conditions, the complex C1 derived from cod
failed to give a useful result (entry 15).
worked as well with dbcot as with cod as coligand, and the com-
plexes C2 and C3 were readily prepared.15 These compounds
indeed were found to be stable toward air and water.
½IrðdieneÞClꢀ2 þ L þ 1 þ AgOTf f C þ AgCl þ CO2
þ CH3OH
ð2Þ
An initial allylic substitution with water as pronucleophile and
C3 as catalyst gave low enantioselectivity and was slow (entry 6);
with KHCO3/18-crown-6 improved selectivity was obtained
(entry 7), but there were many unidentified side products.
Next we tried the conditions of Gais et al. mentioned above,
i.e., KHCO3 as pronucleophile in a biphasic mixture of water and
2073
dx.doi.org/10.1021/ja109953v |J. Am. Chem. Soc. 2011, 133, 2072–2075

Journal of the American Chemical Society
COMMUNICATION
Table 2. Substrate Scope of the Reaction According to
Golgi apparatus.16 The tetrazole 2g (Scheme 3) is a precursor for
the corresponding side chain of our route16 to these analogs. This
example demonstrates absence of interference by the potentially
coordinating tetrazole moiety (Table 2, entry 12).
Scheme 3a
catalyst
time
(h)
yield
(%)b
entry R (substrate)
(mol %)
2/3c ee (%)d
In another approach to the same target (Scheme 4), the
carbonate 4 was prepared from natural brefeldin A and subjected
to the hydroxylation reaction (Scheme 4). In this case, the
products are diastereomers rather than enantiomers. Substrate
control of stereoselectivity was anticipated to be low. Indeed,
enantiomeric catalysts C3 and ent-C3 effected formation of
either of the diastereomers 5a and 5b, respectively, with the
same degree of selectivity, 95:5, as found for the side chain
precursor 2g.
In summary, we have developed the first direct enantioselec-
tive allylic hydroxylation to give synthetically valuable branched
allylic alcohols directly. Essential for success was the use of a new,
air, and moisture stable single component Ir-catalyst and carbon-
ate anion as nucleophile. The system tolerates a wide variety of
solvents and reaction conditions as well as air. Excellent regio-
selectivities and enantioselectivities have been achieved with a
representative set of substrates.
1
CH2OCPh3 (1a) C2 (1)
3.5
7.5
19.5
0.7
6.5
19
92
90
86
72
82
90
81
84
74
77
77
95
96:4
99:1
98:2
98:2
97:3
97:3
99:1
96:4
98:2
99:1
99:1
99:1
95
88
95
96
87
93
89
85
89
86
95
90
2e Ph (1b)
Ph (1b)
4f Ph (1b)
C2 (1)
C3 (1)
C3 (4)
3
5e CH2CH2Ph (1c) C2 (1)
CH2CH2Ph (1c) C3 (1)
7f CH2CH2Ph (1c) C3 (4)
6
0.5
5.5
18.5
1.5
3.5
16
8
C7H15 (1d)
C2 (1)
C3 (1)
C3 (4)
C3 (1)
9
C7H15 (1d)
10 f C7H15 (1d)
11 c-C6H11 (1f)
12 C11H13N4S (1g) C3 (4)
a Solvent: DMF/H2O 10:1, rt. b Isolated yield of branched product.
c Determined by 1H NMR or GC. d Determined by HPLC or GC on a
chiral column. The absolute configurations were determined by com-
parison of the optical rotations with those of known compounds. e The
f
reaction was carried out at 50 °C. The reaction was carried out in
CH2Cl2/H2O 10:1 with additional 18-crown-6 at 40 °C.
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental procedures, results
b
of catalyst screening experiments, characterization of compounds,
determination of regio- and enantioselectivities. This material is
available free of charge via the Internet at http://pubs.acs.org.
Scheme 4. Application of the Iridium-Catalyzed Allylic
Hydroxylation in Natural Products Synthesis (E = CO2CH3)
’ AUTHOR INFORMATION
Corresponding Author
g.helmchen@oci.uni-heidelberg.de
’ ACKNOWLEDGMENT
This work was supported by the Deutsche Forschungsgemeinschaft
(SFB 623) and the Fonds der Chemischen Industrie. We thank
Mascha J€akel for samples of the catalysts and Prof. K. Ditrich (BASF
SE) for enantiomerically pure 1-arylethylamines.
’ REFERENCES
(1) Reviews covering the whole field: (a) Lu, Z.; Ma, S. Angew.
Chem., Int. Ed. 2008, 47, 258–297. (b) Helmchen, G.; Kazmaier, U.;
F€orster, S. In Catalytic Asymmetric Synthesis, 3rd ed.; Ojima, I., Ed.;
Wiley-VCH: New York, 2010; pp 497-461.
(2) (a) Helmchen, G.; Dahnz, A.; D€ubon, P.; Schelwies, M.;
Weihofen, R. Chem. Commun. 2007, 675–691. (b) Helmchen, G. In
Iridium Complexes in Organic Synthesis; Oro, L. A., Claver, C., Eds.;
Wiley-VCH: Weinheim, Germany, 2009; pp 211-250. (c) Hartwig,
J. F.; Stanley, L. M. Acc. Chem. Res. 2010, 43, 1461–1475.
(3) (a) Lopez, F.; Ohmura, T.; Hartwig, J. F. J. Am. Chem. Soc. 2003,
125, 3426–3427. (b) Stanley, L. M.; Bai, C.; Ueda, M.; Hartwig, J. F.
J. Am. Chem. Soc. 2010, 132, 8918–8920.
(4) (a) Cannon, J. S.; Kirsch, S. F.; Overman, L. E. J. Am. Chem. Soc.
2010, 132, 15185–15191. (b) Faller, J. W.; Wilt, J. C. Tetrahedron Lett.
2004, 45, 7613–7616. (c) Covell, D. J.; White, M. C. Angew. Chem., Int.
Ed. 2008, 47, 6448–6451. (d) Trost, B. M.; Organ, M. G. J. Am. Chem.
Soc. 1994, 116, 10320–10321.
The scope with respect to allylic substrates was explored using
catalysts C2 and C3, usually at a load of 1 mol %, which was found
to suffice in the case of substrate 1a (Scheme 3, Table 2, entry 1).
For the carbonate 1b (entries 2-4) superior results were
obtained with C3. In this case the enantioselectivity was even
better in CH2Cl2/H2O than in DMF as solvent (entry 4).
This once more demonstrates the remarkable robustness of the
catalyst. Generally, C3 was the catalyst of choice with respect to
enantioselectivity for substrates with sp3-bound substituents
(entries 5-12). Absolute configurations of the products were
as anticipated according to a general rule, which is valid for all
Ir-catalyzed allylic substitutions probed so far.2a
As an application in natural products synthesis, we have used
the new procedure for the preparation of an analog of the
antibiotic brefeldin A, 15-desmethyl-15-vinylbrefeldin A, which
appears of interest in conjunction with biological effects on the
(5) Evans, P. A.; Leahy, D. K.; Slieker, L. M. Tetrahedron: Asymmetry
2003, 14, 3613–3618.
(6) (a) Mbaye, M. D.; Renaud, J. L.; Demersman, B.; Bruneau, C.
Chem. Commun. 2004, 1870–1871. (b) Austeri, M.; Lindner, D.; Lacour, J.
2074
dx.doi.org/10.1021/ja109953v |J. Am. Chem. Soc. 2011, 133, 2072–2075

Journal of the American Chemical Society
COMMUNICATION
Chem.—Eur. J. 2008, 14, 5737–5741. (c) Kanbayashi, N.; Onitsuka, K.
J. Am. Chem. Soc. 2010, 132, 1206–1207.
(7) (a) Hodgson, D. M.; Humphreys, P. G. In Science of Synthesis;
Clayden, J. P., Ed.; Thieme: Stuttgart, Germany, 2007; Vol. 36, pp 583-
665. (b) Guzman-Martinez, A.; Hoveyda, A. H. J. Am. Chem. Soc. 2010, 132,
10634–10637.
(8) Lyothier, I.; Defieber, C.; Carreira, E. M. Angew. Chem., Int. Ed.
2006, 45, 6204–6207.
(9) (a) L€ussem, B. J.; Gais, H.-J. J. Am. Chem. Soc. 2003, 125, 6066–
6067. (b) Gais, H.-J.; Bondarev, O.; Hetzer, R. Tetrahedron Lett. 2005,
46, 6279–6283. (c) Takahata, H.; Suto, Y.; Kato, E.; Yoshimura, Y.;
Ouchi, H. Adv. Synth. Cat. 2007, 349, 685–693.
(10) Gnamm, C.; Franck, G.; Miller, N.; Stork, T.; Br€odner, K.;
Helmchen, G. Synthesis 2008, 3331–3350.
(11) Spiess, S.; Welter, C.; Franck, G.; Taquet, J.-P.; Helmchen, G.
Angew. Chem., Int. Ed. 2008, 47, 7652–7655.
(12) (a) Madrahimov, S. T.; Markovic, D.; Hartwig, J. F. J. Am. Chem.
Soc. 2009, 131, 7228–7229. (b) Pouy, A. J.; Stanley, L. M.; Hartwig, J. F.
J. Am. Chem. Soc. 2009, 131, 11312–11313.
(13) (a) Spiess, S.; Raskatov, J. A.; Gnamm, C.; Br€odner, K.;
Helmchen, G. Chem.—Eur. J. 2009, 15, 11087–11090. (b) Raskatov,
J. A.; Spiess, S.; Gnamm, C.; Br€odner, K.; Rominger, F.; Helmchen, G.
Chem.—Eur. J. 2010, 16, 6601–6615.
(14) Anton, D. R.; Crabtree, R. H. Organometallics 1983, 2, 621–627.
(15) Compounds of this class were first prepared and characterized
(spectral data and X-ray crystal structures) by Dr. J. A. Raskatov in our
laboratory; Raskatov, J. A.; J€akel, M.; Helmchen, G., unpublished.
(16) F€orster, S.; Persch, E.; Tverskoy, O.; Rominger, F.; Helmchen,
G.; Klein, C.; G€onen, B.; Br€ugger, B. Eur. J. Org. Chem. 2011, early view,
DOI: 0.1002/ejoc.201001297.
2075
dx.doi.org/10.1021/ja109953v |J. Am. Chem. Soc. 2011, 133, 2072–2075