Sequential catalytic isomerization and allylic substitution. Conversion of racemic branched allylic carbonates to enantioenriched allylic substitution products

Article abstract of DOI:10.1021/ja0644273

A catalytic protocol for the conversion of readily accessible racemic, branched aromatic allylic esters to branched allylic amines, ethers, and alkyls has been developed. Palladium-catalyzed isomerization of branched allylic esters to terminal allylic esters, followed by sequential iridium-catalyzed allylic substitution, gave the branched allylic products in good yield with high regioisomeric and enantiomeric selectivity. Both electron-rich and electron-poor branched allylic esters gave products in >90% ee. High enantiomeric excesses were also observed for the products from the reactions of 2-thienyl acetates and dienyl carbonates. Copyright

Full text of DOI:10.1021/ja0644273

Published on Web 08/19/2006
Sequential Catalytic Isomerization and Allylic Substitution. Conversion of
Racemic Branched Allylic Carbonates to Enantioenriched Allylic Substitution
Products
Shashank Shekhar, Brian Trantow, Andreas Leitner, and John F. Hartwig*
Department of Chemistry, Yale UniVersity, P.O. Box 208107, New HaVen, Connecticut 06520-8107
Received June 22, 2006; E-mail: john.hartwig@yale.edu
Table 1. Optimization of Reaction Conditions for Conversion of
We report a sequence of two catalytic processes occurring
through metal-allyl intermediates that leads to enantioselective
substitution of branched allylic esters with nitrogen, oxygen, and
carbon nucleophiles. This process lies at the intersection of develop-
ments in sequential catalytic processes1 and the challenge of control-
ling both regioselectivity and enantioselectivity in allylic substitu-
tion. Enantioselective allylic substitution of branched allylic electro-
philes is desirable because these reagents are easily prepared from
a vinyl Grignard reagent and the wide array of available aldehydes.
Iridium-catalyzed enantioselective allylic substitution of linear
allylic electrophiles has become a powerful tool to prepare optically
active allylic amines and ethers,2-9 but iridium-catalyzed reactions
of branched allylic electrophiles occur with low enantioselectivi-
ties.7,9 Reactions of racemic, branched isomers of monosubstituted
allylic esters catalyzed by Mo10 and W11 complexes form enan-
tioenriched branched substitution products, but no examples of
Mo- or W-catalyzed enantioselective allylic substitutions with
heteroatom nucleophiles have been reported. A few examples of
Pd- and Rh-catalyzed enantioselective allylic substitutions of
monosubstituted allylic esters have been reported to form products
in high enantiomeric excess, but these methods are either restricted
in the scope of substrates that react with high regio- and enanti-
oselectivity12 or use ligands that are cumbersome to synthesize.13
The alternative approach reported here involves a sequence of
Pd-catalyzed isomerization and Ir-catalyzed enantioselective sub-
stitution of the resulting linear isomer. This sequence forms allylic
substitution products from cinnamyl and dienyl carbonates and
acetates in high yields, with high regioselectivities, and with high
enantiomeric excess.
Our studies on the Pd-catalyzed isomerization are summarized
in Table 1. Oxidative addition of allylic carbonates and acetates to
Pd(0) is reversible and can cause isomerization of the allylic esters.14
Pd(II) complexes also catalyze rearrangements of allylic esters and
imidates.15 Thus, we tested both Pd(0) and Pd(II) complexes for a
productive process to generate linear allylic esters for use in crude
form in the Ir-catalyzed allylation of nitrogen, oxygen, and carbon
nucleophiles. Because the Ir-catalyzed reactions occur in high
enantiomeric excess in THF solvent, it was necessary to develop a
Pd-catalyzed process in this medium. Entries 1-7 show the effect
of the identity of the palladium catalyst on the rate of isomerization.
The reactions catalyzed by Pd(PPh3)4 (entry 2) were much faster
than those catalyzed by (CH3CN)2PdCl2 (entry 1), but a palladium
complex lacking excess PPh3 would be more desirable because PPh3
could complex to the [(COD)IrCl]2 7 catalyst precursor or the active
cyclometalated Ir(I) species4,6 and interfere with the subsequent Ir-
catalyzed allylic substitution. Thus, we tested combinations of PPh3
and Pd(dba)2 as catalyst for the isomerization, and the Pd complex
generated from a mixture of 0.2 mol % of Pd(dba)2 and 0.4 mol %
of PPh3 led to complete isomerization of 1a to 2a over a convenient
time period (entry 4). Lower ratios of PPh3 to Pd (entry 5) and
Branched Carbonate to Linear Carbonatesa
yieldb (%)
of 2
entry
[Pd] (%)
L (%)
time (h)
1
2
3
4
5
6
7
8
(CH3CN)2PdCl2 (1)
Pd(PPh3)4 (1)
Pd(dba)2 (0.5)
Pd(dba)2 (0.2)
Pd(dba)2 (0.2)
Pd(dba)2 (0.1)
Pd/C (1)
24
0.6
0.5
3.0
12
24
12
12
95
100
100
100
60
PPh3 (1.0)
PPh3 (0.4)
PPh3 (0.3)
PPh3 (0.2)
70
10
20
Pd(PPh3)4-poly (0.5)
a All reactions were conducted with 1a (1 M) in THF at room
temperature. b GC yield.
Scheme 1
lower loadings of Pd and PPh3 led to incomplete isomerization
(entry 6). Isomerizations catalyzed by Pd on carbon in the absence
of any ligand or by Pd(PPh3)4-polystyrene were slow (entries 7
and 8).
Experiments to combine the isomerization step with allylic substi-
tution were conducted with the combination of Pd(dba)2 (0.2 mol
%) and PPh3 (0.4 mol %) as catalyst. Addition of the iridium catalyst
and benzylamine to the crude reaction solution formed only 10%
of the allylic amine. However, simple filtration of the crude isomer-
ization product through silica to remove the bulk of the palladium
catalyst, followed by addition of the iridium catalyst and benzyl-
amine, led to formation of the substitution products in high yield
and enantioselectivity (Table 2, entry 1). The scope of the sequential
isomerization and enantioselective allylic substitution of branched,
racemic allylic esters is summarized in Table 2. The Ir-catalyzed
process was conducted with three different phosphoramidites: L1,
which first gave high enantiomeric excesses for allylic amination,
the naphthyl analogue L2 that improved regio- and enantioselec-
tivity in some cases,5 and L3, the ligand with edited stereochem-
istry.6,8
In many cases, high yield and selectivities were observed for
the amination reactions when using L3, although certain reactions
required L2 to occur with high selectivities. For example, reactions
of electron-poor cinnamyl carbonates (entries 5-7), dienyl carbon-
ates (entries 10 and 11), and a cinnamyl acetate (entry 8) occurred
in higher yield in the presence of the catalyst generated from L2
than from the catalyst generated from L3. The branched carbonate
1a isomerized to the linear isomer in 3 h at room temperature, and
the crude allylic carbonate reacted over 6 h with the amines after
addition of the iridium catalyst.
Electron-poor allylic carbonates 1b-e also underwent the
sequential catalytic process to give the optically active allylic amines
in good yields with high regio- and enantioselectivities when L2
9
11770
J. AM. CHEM. SOC. 2006, 128, 11770-11771
10.1021/ja0644273 CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Enantioselective Allylic Substitution of Branched
Electrophiles
The isomerization and substitution sequence was also suitable
for the synthesis of allyl aryl ethers and allylic malonates. These
etherification and alkylation reactions catalyzed by Ir complexes
of L1 formed higher ratios of 4 to 5 than reactions catalyzed by Ir
complexes of L3. Carbonate 1a underwent the combination of
isomerization and substitution with lithium phenoxide (entry 12)
to form the branched allyl aryl ether in 76% yield, 9:1 regioselec-
tivity, and 94% ee with the catalyst containing L1. Good enantio-
selectivity was also observed for reactions conducted with L3. The
isomerization of 1a, followed by reaction with sodium dimethyl-
malonate (entry 13) in the presence of the iridium catalyst containing
ligand L1, also formed the corresponding branched allylic malonate
with high regioselectivity and enantioselectivity.
In conclusion, we have developed a catalytic protocol for the
conversion of readily accessible branched aromatic allylic esters
to branched allylic products in good yield with high regio- and
enantioselectivity with a broad range of nucleophiles. This proce-
dure is based on a complementary, sequential use of allylpalladium
and allyliridium chemistry, the palladium-catalyzed reaction for
isomerization, and the iridium-catalyzed reaction for enantioselec-
tive C-N, C-O, and C-C bond formation.
yieldb
eed
(%)
entry
R, X (1)
Nu (3)
L
(%)
4/5c
1
2
3
4
C6H5, OCO2Me (1a) PhCH2NH2 (3a)
C6H5, OCO2Me (1a) morpholine (3b)
C6H5, OCO2Me (1a) PhNH2 (3c)
L3 76 98/2 93
L3 89 98/2 94
L3 83 98/2 94
L2 81 99/1 82
L3 63 99/1 82
L2 82 99/1 95.5
2-FC6H4
morpholine (3b)
morpholine (3b)
morpholine (3b)
morpholine (3b)
PhCH2NH2 (3a)
OCO2Me (1b)
3-OMeC6H4,
OCO2Me (1c)
4-CF3C6H4,
OCO2Me (1d)
4-ClC6H4,
5
6
7
L2 72 99/1 94
L2 83 99/1 96
OCO2Me (1e)
8e-g 4-OMeC6H4
L1 75 99/1 87
L2 83 99/1 94
L3 77 99/1 86
L3 79 98/2 93
L2 57 98/2 88
OAc (1f)
9e-g 2-thienyl, OAc (1g) PhCH2NH2 (3a)
10g,h CH3CHdCH,
OCO2Me (1h)
PhCH2NH2 (3a)
11g,i (CH3)2CHCHdCH, PhCH2NH2 (3a)
OCO2Me (1i)
L2 70 96/4 88
12j,k C6H5
LiOPh (3d)
L1 76 90/10 94
L3 62 85/15 92
OCO2Me (1a)
C6H5
OCO2Me (1a)
13j
NaCH(CO2Me)2 (3f) L1 77 97/3 94
L3 72 80/20 91
Acknowledgment. We thank the NSF for support of this work.
A.L. is a recipient of the FWF Austria Schro¨dinger Fellowship.
We also thank Boehringer Ingelheim for an unrestricted gift, and
Johnson-Matthey for iridium chloride.
Supporting Information Available: Full experimental section. This
material is available free of charge via the Internet at http://pubs.acs.org.
a All reactions were carried out with 1 mmol of 1 and 1.2 mmol of 3 in
THF (1.0 mL) at room temperature in the presence of 0.002 mmol Pd(dba)2,
0.004 mmol PPh3, 0.015 mmol [(COD)IrCl]2, and 0.030 mmol of L1-L3
unless otherwise noted. b Average of isolated yields from two runs.
c Determined by 1H NMR spectra of crude reaction mixtures. d Determined
by chiral HPLC. e Isomerization was performed at 60 °C. f EtOH (0.4 mL)
was used as cosolvent. g [(COD)IrCl]2 (0.020 mmol) and L1-L3 (0.040
mmol) were used. h Pd(dba)2 (0.4%) and PPh3 (0.8%) were used for
isomerization. i Pd(dba)2 (0.3%) and PPh3 (0.6%) were used for isomer-
ization. j THF (2.0 mL) was used. k 3d (2 mmol) was used.
References
(1) (a) Wasilke, J.-C.; Obrey, S.; Baker, R. T.; Bazan, G. C. Chem. ReV. 2005,
105, 1001. (b) Fogg, D. E.; dos Santos, E. N. Coord. Chem. ReV. 2004,
248, 2365.
(2) For seminal work on iridium-catalyzed allylic substitutions with achiral
catalysts, see refs 3a,b.
Scheme 2
(3) (a) Takeuchi, R.; Kashio, M. Angew. Chem., Int. Ed. Engl. 1997, 36, 263.
(b) Takeuchi, R.; Ue, N.; Tanabe, K.; Yamashita, K.; Shiga, N. J. Am.
Chem. Soc. 2001, 123, 9525. (c) Ohmura, T.; Hartwig, J. F. J. Am. Chem.
Soc. 2002, 124, 15164. (d) Lopez, F.; Ohmura, T.; Hartwig, J. F. J. Am.
Chem. Soc. 2003, 125, 3426. (e) Leitner, A.; Shu, C.; Hartwig, J. F. Proc.
Natl. Acad. Sci. U.S.A. 2004, 101, 5830. (f) Shu, C.; Hartwig, J. F. Angew.
Chem., Int. Ed. 2004, 43, 4794. (g) Shu, C.; Leitner, A.; Hartwig, J. F.
Angew. Chem., Int. Ed. 2004, 43, 4797. (h) Bartels, B.; Helmchen, G.
Chem. Commun. 1999, 741. (i) Lipowsky, G.; Miller, N.; Helmchen, G.
Angew. Chem., Int. Ed. 2004, 43, 4595. (j) Weihofen, R.; Dahnz, A.;
Tverskoy, O.; Helmchen, G. Chem. Commun. 2005, 3541.
(4) Kiener, C. A.; Shu, C.; Incarvito, C.; Hartwig, J. F. J. Am. Chem. Soc.
2003, 125, 14272.
(5) Leitner, A.; Shu, C.; Hartwig, J. F. Org. Lett. 2005, 7, 1093.
(6) Leitner, A.; Shekhar, S.; Pouy, J. M. J. Am. Chem. Soc. 2005, 127, 15506.
(7) Bartels, B.; Garcia-Yebra, C.; Rominger, F.; Helmchen, G. Eur. J. Inorg.
Chem. 2002, 2569.
(8) Welter, C.; Dahnz, A.; Brunner, B.; Streiff, S.; Dubon, P.; Helmchen, G.
Org. Lett. 2005, 7, 1239.
(9) Polet, D.; Alexakis, A.; Tissot-Croset, K.; Corminboeuf, C.; Ditrich, K.
Chem. Eur. J. 2006, 12, 3596.
(10) (a) Belda, O.; Moberg, C. Acc. Chem. Res. 2004, 37, 159. (b) Trost, B.
M.; Hachiya, I. J. Am. Chem. Soc. 1998, 120, 1104. (c) Malkov, A. V.;
Spoor, P.; Viander, V.; Kocovsky, P. Tetrahedron Lett. 2001, 42, 509.
(d) Glorius, F.; Neuburger, M.; Pfaltz, A. HelV. Chim. Acta 2001, 84,
3178.
(11) (a) Trost, B. M.; Hung, M.-H. J. Am. Chem. Soc. 1983, 105, 7757. (b)
Lloyd-Jones, G. C.; Pfaltz, A. Angew. Chem., Int. Ed. Engl. 1995, 34,
462. (c) Lehmann, J.; Lloyd-Jones, G. C. Tetrahedron 1995, 51, 8863.
(12) (a) Hayashi, T.; Kishi, K.; Yamamoto, A.; Ito, Y. Tetrahedron Lett. 1990,
31, 1743. (b) Pretot, R.; Pfaltz, A. Angew. Chem., Int. Ed. 1998, 37, 323.
(c) Pamies, O.; Dieguez, M.; Claver, C. J. Am. Chem. Soc. 2005, 127,
3646. (d) Dieguez, M.; Pamies, O.; Claver, C. J. Org. Chem. 2005, 70,
3363. (e) Hayashi, T.; Okada, A.; Suzuka, T.; Kawatsura, M. Org. Lett.
2003, 5, 1713.
(13) You, S.-L.; Zhu, X.-Z.; Luo, Y.-M.; Hou, X.-L.; Dai, L.-X. J. Am. Chem.
Soc. 2001, 123, 7471.
(14) Amatore, C.; Jutand, A.; Mensah, L.; Meyer, G.; Fiaud, J.-C.; Legros,
J.-T. Eur. J. Org. Chem. 2006, 1185.
was used as ligand (entries 4-7). The Pd-catalyzed isomerization
of electron-poor, branched allylic carbonates (1b-e) to the corre-
sponding linear carbonates was slightly slower than the isomer-
ization of the neutral carbonate 1a. However, the allylic amination
product was obtained in 6 h from the isomerized linear carbonates.
Reactions of electron-rich cinnamyl esters were initiated with
acetate, rather than carbonate, leaving groups because the electron-
rich carbonates were unstable toward silica gel purification.
Although the Pd-catalyzed isomerization of electron-rich branched
allylic acetates (1f,g) was slower (6-12 h at 60 °C) than the
isomerization of the allylic carbonates (1a-e), the subsequent allylic
amine products were obtained in good yields with high regio- and
enantioselectivities. For example, the product from reaction of
p-methoxy-substituted 1f and benzylamine was obtained in 94.4%
ee when L2 was used as ligand (entry 8). The product from reaction
of thienyl derivative 1g with benzylamine in the presence of the
catalyst containing L3 formed the allylic amine product in 93% ee
(entry 9). Dienyl carbonates 1h,i underwent the catalytic isomer-
ization and allylic substitution to give optically active amines in
moderate yield and good enantioselectivities (entries 10 and 11).
(15) (a) Overman, L. E.; Knoll, F. M. Tetrahedron Lett. 1979, 20, 321. (b)
Overman, L. E.; Ziller, J.; Zipp, G. G. J. Org. Chem. 1997, 62, 1449. (c)
Overman, L. E.; Owen, C. E.; Pavan, M. M.; Richards, C. J. Org. Lett.
2003, 5, 1809.
JA0644273
9
J. AM. CHEM. SOC. VOL. 128, NO. 36, 2006 11771