Synthesis of β-substituted α-amino acids with use of iridium-catalyzed asymmetric allylic substitution

Article abstract of DOI:10.1021/jo034638f

The asymmetric synthesis of β-substituted α-amino acids with use of iridium-catalyzed allylic substitution was described. The Ir-catalyzed allylic substitution of diphenylimino glycinate with allylic phosphates proceeded smoothly even at 0 °C and gave branch products with high enantioselectivity (up to 97% ee), when chiral bidentate phosphite bearing the 2-ethylthioethyl group was employed. In addition, both diastereomers of the branch products were synthesized stereoselectively by simply switching the base employed. These methods were also applied to the asymmetric synthesis of quaternary α-amino acids.

Full text of DOI:10.1021/jo034638f

Syn th esis of â-Su bstitu ted r-Am in o Acid s w ith Use of
Ir id iu m -Ca ta lyzed Asym m etr ic Allylic Su bstitu tion
Takatoshi Kanayama, Kazumasa Yoshida, Hideto Miyabe, Tetsutaro Kimachi, and
Yoshiji Takemoto*
Graduate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku, Kyoto 606-8501, J apan, and
Faculty of Pharmaceutical Sciences, Mukogawa Women’s University, 11-68 Koshien Kyuban-cho,
Nishinomiya 663-8179, J apan
takemoto@pharm.kyoto-u.ac.jp
Received May 14, 2003
The asymmetric synthesis of â-substituted R-amino acids with use of iridium-catalyzed allylic
substitution was described. The Ir-catalyzed allylic substitution of diphenylimino glycinate with
allylic phosphates proceeded smoothly even at 0 °C and gave branch products with high
enantioselectivity (up to 97% ee), when chiral bidentate phosphite bearing the 2-ethylthioethyl
group was employed. In addition, both diastereomers of the branch products were synthesized
stereoselectively by simply switching the base employed. These methods were also applied to the
asymmetric synthesis of quaternary R-amino acids.
SCHEME 1. Tr a n sition Meta l-Med ia ted
Asym m etr ic Allylic Su bstitu tion
In tr od u ction
The transition metal-catalyzed asymmetric allylic sub-
stitution is a useful reaction in organic synthesis.1 Since
the allylic alkylation with dialkyl malonates has been
intensively studied, good yields and high enantioselec-
tivities can now be obtained with a proper combination
of a transition metal and a chiral ligand.2-5 In contrast
to the symmetric C-nucleophiles, the same reaction of
3-substituted allylic alcohol derivatives B with unsym-
metrical C-nucleophiles A is a tough and challenging
task, because both regio- and diastereoselectivities as
well as enantioselectivity should be controlled to give the
desired stereoisomers (Scheme 1). Over the past few
* Corresponding author. Fax: 81-075-753-4569.
(1) Recent reviews: (a) Trost, B. M. Chem. Pharm. Bull. 2002, 50,
1. (b) Trost, B. M.; Lee, C. B. In Catalytic Asymmetric Synthesis II;
Ojima, I., Ed.; Wiley-VCH: Weinheim, Germany, 2000; p 593. (c)
Hayashi, T. In Catalytic Asymmetric Synthesis; Ojima, I., Ed.; Wiley-
VCH: Weinheim, Germany, 2000; p 193. (d) Pfaltz, A.; Lautens, M.
Compr. Asymmetric Catal. I-III 1999, 2, 833-884.
(2) (a) Matsushima, Y.; Onitsuka, K.; Kondo, T.; Mitsudo, T.;
Takahashi, S. J . Am. Chem. Soc. 2001, 123, 10405. (b) You, S.-L.; Zhu,
X.-Z.; Luo, Y.-M.; Hou, X.-L.; Dai, L.-X. J . Am. Chem. Soc. 2001, 123,
7471. (c) Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336. (d)
Blacker, A. J .; Clarke, M. L.; Loft, M. S.; Mahon, M. F.; Humphries,
M. E.; Williams, J . M. J . Chem. Eur. J . 2000, 6, 353. (e) Hayashi, T. J .
Organomet. Chem. 1999, 576, 195. (f) Pre´toˆt, R.; Pfaltz, A. Angew.
Chem., Int. Ed. Engl. 1998, 37, 323. (g) Williams, J . M. J . Synlett 1996,
705.
(3) (a) Fuji, K.; Kinoshita, N.; Tanaka, K.; Kawabata, T. Chem.
Commun. 1999, 2289. (b) Bartels, B.; Helmchen, G. Chem. Commun.
1999, 741. (c) J anssen, J . P.; Helmchen, G. Tetrahedron Lett. 1997,
38, 8025.
(4) (a) Trost, B. M.; Dogra, K.; Hachiya, I.; Emura, T.; Hughes, D.
L.; Krska, S.; Reamer, R. A.; Palucki, M.; Yasuda, N.; Reider, P. J .
Angew. Chem., Int. Ed. 2002, 41, 1929. (b) Glorius, F.; Neuburger, M.;
Pfaltz, A. Helv. Chim. Acta 2001, 84, 3178. (c) Kaiser, N.-F.; Bremberg,
U.; Larhed, M.; Hallberg, A. Angew. Chem., Int. Ed. 2000, 39, 3596.
(d) Trost, B. M.; Hildbrand, S.; Dogra, K. J . Am. Chem. Soc. 1999,
121, 10416.
years, research has been focused on finding catalysts and
chiral ligands, which favor the formation of branched
chiral products D and E in the allylic substitution of
R-amino esters A with B.6,7 We have already reported
the Pd-mediated asymmetric allylic alkylation of diphe-
nylimino glycinate 1 with several allylic acetates in the
presence of the chiral phase transfer catalyst (PTC) 11
(6) (a) Nakoji, M.; Kanayama, T.; Okino, T.; Takemoto, Y. Org. Lett.
2001, 3, 3329. (b) Chen, G.; Deng, Y.; Gong, L.; Mi, A.; Cui, X.; J iang,
Y.; Choi, M. C. K.; Chan, A. S. C. Tetrahedron: Asymmetry 2001, 12,
1567. (c) You, S.-L.; Hou, X.-L.; Dai, L.-X.; Cao, B.-X.; Sun, J . Chem.
Commun. 2000, 1933. (d) Trost, B. M.; Ariza, X. J . Am. Chem. Soc.
1999, 121, 10727. (e) Kuwano, R.: Ito, Y J . Am. Chem. Soc. 1999, 121,
3236. (f) Hiroi, K.; Hidaka, A.; Sezaki, R.; Imamura, Y. Chem. Pharm.
Bull. 1997, 45, 769. (g) Genet, J .-P.; J uge, S.; Achi, S.; Mallart, S.;
Montes, J . R.; Levif, G. Tetrahedron 1988, 44, 5263.
(7) (a) Nakoji, M.; Kanayama, T.; Okino, T.; Takemoto, Y. J . Org.
Chem. 2002, 67, 7418. (b) Weiss, T. D.; Helmchen, G.; Kazmaier, U.
Chem. Commun. 2002, 1270. (c) Kazmaier, U.; Zumpe, F. L. Eur. J .
Org. Chem. 2001, 4067. (d) Kazmaier, U.; Zumpe, F. L. Angew. Chem.,
Int. Ed. 2000, 39, 802. (e) Kazmaier, U.; Zumpe, F. L. Angew. Chem.,
Int. Ed. 1999, 38, 1468. (f) Trost, B. M.; Ariza, X. Angew. Chem., Int.
Ed. Engl. 1997, 36, 2635. (g) Baldwin, I. C.; Williams, J . M. J .; Beckett,
R. P. Tetrahedron: Asymmetry 1995, 6, 1515.
(5) (a) Lloyd-J ones, G. C.; Pfaltz, A. Angew. Chem., Int. Ed. Engl.
1995, 34, 462. (b) Lehmann, J .; Lloyd-J ones, G. C. Tetrahedron 1995,
51, 8863.
10.1021/jo034638f CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/04/2003
J . Org. Chem. 2003, 68, 6197-6201
6197

Kanayama et al.
TABLE 1. Asym m etr ic Allylic Alk yla tion of 2-7 w ith 1
in th e P r esen ce of Ch ir a l P TC 11
TABLE 2. Ir -Ca ta lyzed Allylic Alk yla tion of 4 a n d 7
w ith 1 in th e P r esen ce of Ch ir a l liga n d s 13-18
ee, %
8a 9a
ligand,
yield, %
ratio
8a :9a
entry substrate conditions 8a + 9a
1
2
3
4
5
6
7
8
4
4
4
4
7
7
7
7
13, rt
14, rt
15, rt
16, rt
16, 0 °C
15, 0 °C
17, 0 °C
18, 0 °C
29
7
6
11
82
0
43
4
69:31
86:14
67:33
73:27
82:18
a
23a
b
46
25
66
68a
95
93
97
yield, %
ee, %
ratio
entry substrate conditions 8a + 9a ′ 10a 8a :9a ′ 8a 9a ′
1
2
3
4
5
6
2
3
4
5
6
7
Aa
Bb
Bb
Bb
Bb
Bb
13
31
40
15
23
47
71
0
0
0
0
5:95
c
85
c
77:23 49
75:25 46
80:20 29
83:17 57
83:17 26
65:35
75:25
29
41
39
53
c
c
c
c
a
The enantiomers of 8a and 9a were obatined as a major
b
product. Not determined.
0
a
Pd2(dba)3‚CHCl3 (3 mol %), chiral ligand 12 (9 mol %), 0 °C.
[Ir(cod)Cl]2, and (PhO)3P) (entries 2-6). In all cases with
the iridium catalyst, the regioselectivity was almost
perfect and 10a could not be observed. Furthermore, the
other diastereomer 8a among the branch products was
obtained as a major product with moderate enantiose-
lectivity. The absolute configuration of the C-2 stereo-
genic center of the products 8a and 9a ′ was assumed to
be S, because the alkylation10 and Pd-mediated allylic
alkylation7a of 1 with the chiral PTC 11 generally produce
(2S)-amino acid derivatives. Unfortunately, the chemical
yield and ee of 8a could not be improved by simply
changing the leaving groups of the substrates. To over-
come this situation, we next investigated the effect of
chiral Ir-ligands 13-1811 in place of the chiral PTC 11
on the enantioselectivity (Table 2). The reaction of 1 with
benzoate 4 was carried out in the presence of 50% KOH
(3 equiv), [Ir(cod)Cl]2 (10mol %), and chiral phosphite
(20mol %). Indeed, it was revealed that the enantiose-
lectivity, but not the diastereoselectivity, was dramat-
ically affected by the substituent (R) of the ligands 13-
18.
b
[Ir(cod)Cl]2 (10 mol %), (PhO)3P (40 mol %), room temperature.
c Not determined.
(Table 1), giving the chiral products C with high enan-
tioselectivity (up to 96% ee).6a Unlike the palladium
catalyst, some transition metals such as Ir,3 Mo,4 and W5
promote the allylic alkylation at the more substituted
terminus of the allylic substrate. Trost et al. recently
reported that the Mo-catalyzed asymmetric allylic alky-
lation with azlactones occurs at the more substituted
terminus with high regio-, diastereo-, and enantioselec-
tivity.8 However, there are no reports concerning the
asymmetric synthesis of both diastereomers D and E as
a major product with use of the same starting materials
and the same chiral ligand. In this article, we detail the
enantioselective Ir-catalyzed allylic substitutions of 1 in
the presence of various chiral phosphites, and the sig-
nificant effect of bases on diastereoselectivity of the
obtained branch products 8 and 9 as well as application
of the reaction to asymmetric synthesis of quaternary
amino acids.9
Whereas addition of the known chiral phosphites 133a
and 1411f in place of (PhO)3P gave the same product 8a
as a major product with moderate enantioselectivity, that
of new ligands 15 and 16 afforded 8a in 95% and 93% ee
yield at the expense of the chemical yield, respectively
(entries 1-4). These results suggest that phosphite
ligands comprised of a BINOL and a primary alcohol are
more effective to enhance the enantioselectivity of the
product. As hydrolysis of 4 to cinnamyl alcohol occurred
Resu lts a n d Discu ssion
Asym m etr ic Allylic Alk yla tion of Allylic Alcoh ol
Der iva tives 2-7 w ith 1 in th e P r esen ce of Ch ir a l
P TC 11 a n d Ch ir a l P h osp h ites 13-24. Our previous
work7a prompted us to examine PTC 11 as a chiral
catalyst in transition metal-catalyzed allylic substitutions
of diphenylimino glycinate 1 (Table 1). We first carried
out the Pd-catalyzed reaction of 1 and acetate 2 in the
presence of 11 and chiral ligand 12 (entry 1). Although
the branch product 9a ′ was obtained with good diastereo-
and enantioselectivity, the regioselectivity of 9a ′ to 10a
was low. Then we examined the Ir-catalyzed allylic
substitutions of 1 and several allylic alcohol derivatives
3-7 under the phase-transfer conditions (50% KOH,
(10) (a) O’Donnell, M. J . Aldrichim. Acta 2001, 34, 3. (b) Corey, E.
J .; Xu, E.; Noe, M. C. J . Am. Chem. Soc. 1997, 119, 12414. (c) Lygo,
B.; Crosby, J .; Peterson, J . A. Tetrahedron Lett. 1999, 40, 8671-8674.
(d) Okino, T.; Takemoto, Y. Org. Lett. 2001, 3, 1515.
(11) (a) Ansell, J .; Wills, M. Chem. Soc. Rev. 2002, 31, 259. (b) Reetz,
M. T.; Mehler, G. Angew. Chem., Int. Ed. 2000, 39, 3889. (c) Deeren-
berg, S.; Schrekker, H. S.; van Strijdonck, G. P. F.; Kamer, P. C. J .;
van Leeuwen, P. W. N. M.; Fraanje, J .; Goubitz, K. J . Org. Chem. 2000,
65, 4810. (d) Arena, C. G.; Drommi, D.; Faraone, F. Tetrahedron:
Asymmetry 2000, 11, 2765. (e) Selvakumar, K.; Valentini, M.; Pregosin,
P. S.; Albinati, A. Organometallics 1999, 18, 4591. (f) Ovchinnikov, V.
V.; Cherkasova, O. A.; Verizhnikov, L. V. Zh. Obshch. Khim. 1982,
52, 707.
(8) Trost, B. M.; Dogra, K. J . Am. Chem. Soc. 2002, 124, 7256.
(9) Kanayama, T.; Yoshida, K.; Miyabe, H.; Takemoto, Y. Angew.
Chem., Int. Ed. 2003, 42, 2054-2056.
6198 J . Org. Chem., Vol. 68, No. 16, 2003

Iridium-Catalyzed Asymmetric Allylic Substitution
TABLE 3. Ir -Ca ta lyzed Allylic Alk yla tion of 7 w ith 1 in
th e P r esen ce of Ch ir a l Liga n d s 19-24
TABLE 4. Ir -Ca ta lyzed Allylic Su bstitu tion of 1 a n d 7
w ith Va r iou s Rea ction Con d ition s
yield, %
ee, %
ratio
entry
reaction conditions
8a :9a 10a 8a :9a 8a 9a
1
2
3
4
5
6
7
8
9
11, 50%KOH, Tol.
PTC,a 50%KOH, Tol.
solid KOH, Tol.
KN(SiMe3)2, THF
CsOH‚H2O, Tol.
NaH, THF
LiBr, DBU, THF
LDA, THF
LiN(SiMe3)2, THF
30
41
71
28
43
29
20
56
82
0
0
0
0
0
0
23
3
83:17 95 87
51:49 95 60
70:30 97
b
79:21 48 72
70:30 95 59
62:38 91 73
30:70 44 63
11:89
b
96
<1
12:88 56 92
a
b
PTC: [CH3(CH2)15N(CH3)3]Br. Not determined.
ee, %
yield, %
8a + 9a
ratio
8a :9a
cations had a more significant influence on the diaste-
reoselectivity (8a /9a ) than on the enantioselectivity of 8a .
At first, we expected that the diastereoselectivity should
be improved by the dual asymmetric induction, if chiral
PTC 11 can form a chiral complex with the resulting
enolate. Then, we carried out the substitution of 1 and 7
in the presence of either chiral PTC 11 or achiral PTC
[CH3(CH2)15NMe3]Br using 50% KOH as a base (entries
1 and 2). However, the diastereoselectivity of 8a was
almost the same or decreased compared with the result
without PTC’s, while high ee values were still main-
tained. Other bases such as solid KOH, KN(SiMe3)2,
CsOH‚H2O, and NaH had a marginal effect on the
stereoselectivity except entry 4, producing the same
isomer 8a as a major product (entries 3-6). On the other
hand, use of bases such as LiBr/DBU, LDA, and LHMDS,
which would generate the lithium enolate, affected the
diastereoselectivity, producing the other isomer 9a as a
major product (entries 7-9). In terms of chemical yield
and stereoselectivity, 50% KOH in toluene [method A]
and LHMDS in THF [method B] are the best conditions
for diastereoselectively preparing 8a and 9a , respectively.
In this manner, these two methods allow us to synthesize
both diastereomers 8a and 9a with high enantioselec-
tivity.
Adopting the two protocols (methods A and B) as our
standard, various allylic substrates 25-29 were exam-
ined (Table 5). Although the requisite phosphates 25-
29 were synthesized from the corresponding allylic
alcohols, phosphate 27 was used without purification for
the allylic substitution due to its lability to column
chromatography. In general, the Ir-catalyzed allylic
substitution was not affected by the para and meta
substituent of the aromatic ring of 25-28. Furthermore,
phosphate 29, bearing a naphthyl group, could be em-
ployed as an allylic substrate, giving the desired products
8f and 9f in comparable yields. Thus, with method A,
the corresponding branch products 8b-f were obtained
diastereoselectively (8/9 ) 71/29 to 79/21) with excellent
enantioselectivity (>90% ee). Similarly, with method B,
other branch products 9b-f could be synthesized stereo-
selectively (8/9 ) 13/87 to 10/90, >88% ee). As described
above, these two protocols were applicable to several
allylic substrates and were demonstrated to be a versatile
tool for asymmetric synthesis of both diastereomers 8a -f
and 9a -f.
entry
ligand
8a
9a
1
2
3
4
5
(R)-19
(R)-20
(R)-21
(R)-22
(S)-22
17
62
67
37
52
71:29
71:29
75:25
81:19
79:21
82
97
96
95
68b
52
42
a
a
25b
Not determined. b The enantiomers of 8a and 9a were obatined
a
as a major product.
predominantly under these conditions, phosphate 7 was
employed as an allylic substrate, which should be resis-
tant to the hydrolysis. After several experiments, it was
revealed that the best result (82% yield, 8a :9a ) 82:18,
97% ee) was obtained when the reaction was performed
at 0 °C with 7 in the presence of chiral ligand 16 (entry
5). On the other hand, carbon-analogue 15 did not
promote the reaction efficiently (entry 6). In addition, the
choice of the chiral phosphite bearing the sulfur atom in
the proper position on the alkyl group was essential to
improve both the chemical yield and stereoselectivity of
8a (entries 7 and 8). The acceleration of the reaction rate
with 16 is attributed to the formation of six-membered
chelation with 16, which enhances nucleophilicity of the
Ir catalyst. To clarify the reason for the excellent
performance of phosphite 16, we synthesized various
phosphites and reexamined their potential (Table 3).
Although replacement of the EtS group in 16 by a PhS
group retarded the reaction, that by other alkyl groups
such as BnS and i-BuS gave similar results in terms of
stereoselectivity (entries 1-3). From the results of the
allylic substitution with (R)-22 and (S)-22 as a chiral
ligand, the chirality of BINOL affected the enantioselec-
tivity more strongly than that of sulfide (entries 4 and
5). Unfortunately, phosphites 23-24, introducing the
substituents (Ph, Me) on the 3- and 3′-position of BINOL,
did not promote the reaction and no desired products
were afforded due to the severe steric hindrance.
Dia ster eoselective Syn th esis of Both Ster eoiso-
m er s 8 a n d 9 by Sw itch in g a Ba se. Having established
higher levels of enantioselectivity of 8a , our attention was
focused on the diastereoselectivity (8a /9a ). Since the
diastereoselectivity of 8a /9a was marginally affected by
chiral ligands such as PTC and phosphites, we explored
the effect of the countercations of the resulting enolate
by the reaction of 1 with various bases, and the results
are shown in Table 4. Noteworthy is that the counter-
J . Org. Chem, Vol. 68, No. 16, 2003 6199

Kanayama et al.
TABLE 5. Ir -Ca ta lyzed Allylic Su bstitu tion of 1 w ith
Va r iou s Su bstr a tes 25-29
ee, %
F IGURE 1. The plausible allyl-Ir(III) complex F and F′.
yield, %
ratio
entry
substrate
methoda
8 + 9
8:9
8
9
SCHEME 3. Deter m in a tion of Con figu r a tion of 8a
a n d 9a a n d Th eir Syn th etic Ap p lica tion to Ch ir a l
Heter ocyclic Com p ou n d s
1
2
3
4
5
6
7
8
9
25
26
A
A
A
A
A
B
B
B
B
B
77
78
86
79
97
81
78
70
81
84
78:22
71:29
79:21
76:24
77:23
10:90
10:90
13:87
10:90
11:89
97
98
90
97
94
62
93
46
73
67
63
61
70
74
69
88
94
90
94
96
27b
28
29
25
26
27b
28
10
29
a
Method A: In toluene at 0 °C. The ratio of 1:25-29:50% KOH:
[Ir(cod)Cl]2:(R)-16 was 100:100:300:10:20. Method B: In THF at
0 °C. The ratio of 1:25-29:LiN(SiMe3)2:[Ir(cod)Cl]2:(R)-16 was 150:
b
100:150:10:20. The crude phosphate 27 was used without puri-
fication.
SCHEME 2. Ir -Ca ta lyzed Allylic Alk yla tion of
Meth yl Ester 30 w ith 7
and absolute configurations of products 8a and 9a were
revealed to be correct as depicted in Scheme 3.7a The
configurations of 8b-f and 9b-f were then assumed by
analogy. From the results described above, the stereo-
chemical course of the reaction can be explained as
follows (Figure 1). Initially, two π-allyl complexes F and
F ′ (σ-allyl complexes could not be denied) would be
formed predominantly by attack of the iridium(I)-ligand
16 complex on the allylic substrate 7.13 However, the
complex F ′ should be disfavored due to the steric interac-
tion between the ethylthio group of the ligand and the
phenyl group of 7. Therefore, the nucleophilic attack of
the enolate of 1 at the allylic carbon trans to the
phosphorus atom would give the chiral products 8 and 9
with high enantioselectivity. Although we cannot explain
the different behavior of the bases at this stage, it might
be attributed to the geometry of the enolate of 1. Namely,
it was assumed that the use of KOH as a base would give
the E-enolate G predominantly, but, in contrast, the
Z-enolate H would be formed with use of LiN(SiMe3)2 as
a base.14
We next applied these methods to the methyl ester 30
to identify the importance of the tert-butyl group of 1 for
stereoselectivity (Scheme 2). Treatment of 30 by method
A gave the same diastereomer 31 as a major product in
good enantioselectivity (90% ee), but the diastereoselec-
tivity decreased significantly (73% combined yield, 31:
32 ) 60:40). Similarly, subjection of 30 to method B gave
the other diastereomer 32 with good enantioselectivity
(89% ee) but with moderate diastereoselectivity (87%
combined yield, 31:32 ) 21:79). These results suggest
that (1) the tert-butyl group of 1 would be essential for
good diastereoselectivity but not for enantioselectivity
and (2) stereodifferentiation of the enantiotopic faces of
the allyl-metal complex coordinated by chiral ligand 16
is highly independent of the nucleophiles employed.
Ap p lica tion of 8 a n d 9 to Asym m etr ic Syn th esis
of th e Con for m a tion a lly Con str a in ed Am in o Acid s
a n d Deter m in a tion of th eir Absolu te a n d Rela tive
Con figu r a tion s. As described in Scheme 3, two obtained
diastereomers 8a and 9a were transformed into pipecolic
acid derivatives 34 and 36 in 4 and 6 steps, respectively,
by using the intramolecular olefin metathesis reaction
as a key step. Furthermore, the hydrogenation of 36 gave
rise to the product 37,12a which is known as a useful
conformationally constrained amino acid.12 By compari-
son of their spectra (37 and antipode of 37), the relative
Asym m etr ic Allylic Alk yla tion of Allylic P h os-
p h a tes 7 a n d 29 w ith Va r iou s Im in o Ala n in a tes 38-
40 in th e P r esen ce of Ch ir a l P h osp h ite 16. Since we
succeeded in the enantioselective synthesis of the branch
products 8a -f and 9a -f, we finally investigated the
asymmetric synthesis of quaternary amino acids using
(12) (a) Liu, D.-G.; Gao, Y.; Wang, X.; Kelley, J . A.; Burke, T. R., J r.
J . Org. Chem. 2002, 67, 1448. (b) Liu, D.-G.; Wang, X.-Z.; Gao, Y.; Li,
B.; Yang, D.; Burke, T. R., J r. Tetrahedron 2002, 58, 10423.
(13) O’Donnell, M. J .; Wu, S. Tetrahedron: Asymmetry 1992, 3, 591.
(14) Lipkowitz, K. B.; Cavanaugh, M. W.; Baker, B.; O’Donnell, M.
J . J . Org. Chem. 1991, 56, 5181.
6200 J . Org. Chem., Vol. 68, No. 16, 2003

Iridium-Catalyzed Asymmetric Allylic Substitution
TABLE 6. Ir -Ca ta lyzed Allylic Su bstitu tion of Va r iu ou s
Im in o Ala n in a te 38-40 w ith 7 a n d 29
SCHEME 4. Deter m in a tion of Con figu r a tion of 42a
similar reaction of 29 with other alaninates 39 and 40
gave no efficient enhancement in terms of stereoselec-
tivity (entries 6-8). In contrast to method A (entries 1
and 2), it should be noted that the enantioselectivities of
both products 41b and 42b were good to high (75-87%
ee). The configuration of 42a was determined by chemical
transformation of 42a into the known compound 438 as
described in Scheme 4. Furthermore, the configuration
of 41a ,b and 42b was assumed by analogy to the allylic
alkylation of 1 and 7, giving 8a and 9a . The above results
suggest that chiral quaternary amino acids could be
synthesized by using the Ir-catalyzed allylic alkylation
of alaninate 38 and allylic phosphate with acceptable
yield and stereoselectivity.
ee, %
yield, %
41 + 42
ratio
41:42
entry
substrate
methoda
41
42
1
2
3
4
5
6
7
38 + 7
A
A
18
22
17
49
59
47
64
50:50
50:50
65:35
55:45
17:83
19:81
25:75
92
91
91
93
84
82
87
30
35
46
66
82
75
83
38 + 29
40 + 29
40 + 29
38 + 29
39 + 29
40 + 29
A + 11b
A + PTCc
B
B
B
Con clu sion
a
Method A: In toluene at 0 °C. The ratio of 38-40:7-29:50%
KOH:[Ir(cod)Cl]2:(R)-16 was 100:100:300:10:20, Method B: In THF
at 0 °C. The ratio of 38-40:29:LiN(SiMe3)2:[Ir(cod)Cl]2:(R)-16 was
150:100:150:10:20. b 11 (10 mol %). c PTC: [CH3(CH2)15N(CH3)3]Br
(10 mol %).
We have developed the first enantioselective
Ir-catalyzed allylic substitutions of diphenylimino glyci-
nate 1 by using chiral bidentate ligand 16 (up to 98%
ee), and also succeeded in the diastereoselective asym-
metric synthesis of both diastereomers 8 and 9 by simply
switching the base employed. In addition, these methods
could be applied to the asymmetric synthesis of chiral
quaternary amino acids starting from imino alaninate
38 and phosphates 7 and 29.
the Ir-catalyzed allylic alkylation of various alaninates
38-4015 (Table 6). The adoption of method A to phos-
phate 7 and alaninate 38 gave 41a and 42a as a 1/1
mixture in low yield (entry 1). The reaction of 29 and 38
also proceeded slowly and gave a similar result (entry
2).
It is noteworthy that the enantioselectivity of the
products 41a ,b was high, but that of 42a ,b was low.
Although chiral and achiral PTC’s were added to the
reaction mixture to improve the chemical yield (entries
3 and 4), the chemical yield and diastereoselectivity of
41b are only somewhat increased. In contrast to method
A, the adoption of method B to phosphate 29 and
alaninate 38 afforded 42b as a major product (41b:42b
) 17:83) in 59% yield (entry 5). In this case, both
diastereo- and enantioselectivity were acceptable. A
Ack n ow led gm en t. This research was supported by
grants from the J apan Health Sciences Foundation and
Grant-in-Aid for Scientific Research (C) from the Min-
istry of Education, Culture, Sports, Science and Tech-
nology, J apan.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails and characterization of 8a -f, 9a -f, 16, 37, 41a ,b, and
42a ,b. This material is available free of charge via the Internet
at http://pubs.acs.org.
J O034638F
J . Org. Chem, Vol. 68, No. 16, 2003 6201