Enantioselective synthesis of β-arylbutenolides via palladium(0) catalysed asymmetric coupling cyclisation reaction of racemic allenic carboxylic acids with aryl iodides

Article abstract of DOI:10.1016/S0957-4166(01)00046-5

Bis(oxazoline) 2c was used as a chiral ligand in the Pd(0) catalysed enantioselective coupling cyclisation of 2,3-allenoic acids with aryl iodides affording butenolides in reasonable yields and with e.e.s of up to 53%.

Full text of DOI:10.1016/S0957-4166(01)00046-5

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 193–195
Enantioselective synthesis of b-arylbutenolides via palladium(0)
catalysed asymmetric coupling cyclisation reaction of racemic
allenic carboxylic acids with aryl iodides
Shengming Ma,* Zhangjie Shi and Shulin Wu
Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
354 Fenglin Lu, Shanghai 200032, PR China
Received 27 November 2000; accepted 22 January 2001
Abstract—Bis(oxazoline) 2c was used as a chiral ligand in the Pd(0) catalysed enantioselective coupling cyclisation of 2,3-allenoic
acids with aryl iodides affording butenolides in reasonable yields and with e.e.s of up to 53%. © 2001 Elsevier Science Ltd. All
rights reserved.
The synthesis of butenolides is of interest to synthetic
chemists because of their potential biological activities
as well as the wide ranging occurrence of butenolides in
natural products.1 Recently, we have developed some
efficient methodologies for the synthesis of butenolides
from 2,3-allenoic acids.2 In particular, polysubstituted
butenolides can be produced via the Pd(0)/Ag(I) co-
catalysed coupling cyclisation reaction of 2,3-allenoic
acids and organic halides (Eq. (1)).2a In this reaction,
the Pd atom at least partially takes part in the process
of the formation of the chiral centre in the butenolide,
which led us to investigate the catalytic enantioselective
synthesis of butenolides.3,4 Herein, we wish to report
our recent investigations into this asymmetric buteno-
lide synthesis.
reaction conditions were screened, but the results were
disappointing (Table 1, entries 2–8). Thus, minimising
the cycloisomerisation reaction was necessary to opti-
mise the reaction yield.
The original reaction was heterogeneous as a result of
the poor solubility of Ag2CO3 and K2CO3 in CH3CN, it
is also known that Ag(I) catalyses the cycloisomerisa-
tion of 2,3-allenoic acids under certain conditions,6
thus, a new reaction was developed free of K2CO3 and
Ag2CO3. After screening numerous conditions (Table 1,
entries 1–8), we observed that when (i-Pr)2NEt was
used as the base in place of K2CO3 and Ag2CO3, the
reaction afforded the expected product 3a in moderate
yield but with only a trace amount of 3a%. When the
temperature was lowered to −15°C, butenolide 3a was
obtained in 52% yield and 52% e.e. as the sole product
(Table 1, entry 9). At this point, 31% of racemic 1a was
recovered,7 indicating that the low yield was caused by
incomplete conversion and the reaction is not a kinetic
resolution process of the starting 2,3-allenoic acids. It is
interesting to observe that when other different bis(oxa-
zoline) ligands 2a–2b and 2d–2e5 (Fig. 1) were used
R'
cat. Pd(0) and Ag+
CH3CN
R
+
R'X
R
O
COOH
O
(1)
Under our original coupling cyclisation conditions,2a
when Pd2(dba)3·CHCl3 and ligand 2c5 (Fig. 1) were
used instead of Pd(PPh3)4 as the catalyst at a tempera-
ture of 85°C, the reaction between 1a and iodobenzene
afforded the coupling cyclisation product 3a in 56%
yield with zero e.e. (Table 1, entry 1). However, when
this reaction was carried out at room temperature, 3a
was isolated in 24% yield and 37% e.e. together with the
formation of cycloisomerisation product 3a% in 60%
yield with an e.e. of 7% (Table 1, entry 2). Several
O
N
O
O
N
O
N
N
R
R
R
R
2a R = Ph
2b R = i-Pr
2c R = Bn
2d R = Ph
2e R = Bn
* Corresponding author.
Figure 1.
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00046-5

194
S. Ma et al. / Tetrahedron: Asymmetry 12 (2001) 193–195
Table 1. The coupling cyclisation of 1a and iodobenzene under different conditionsa
H
Ph
.
H
n-C7H15
H
Pd2(dba)3 CHCl3, 2
base, additive
CH3CN
+
+
PhI
O
O
n-C7H15
n-C7H15
O
O
COOH
3a
3a'
1a
Entry
2
PhI (equiv.)
Base (equiv.)
Additive (mol%)
Temp. (°C)
Time (h)
3ac,d % (e.e. %)
3a%c,d % (e.e.)
1b
2
3
2c
2c
2c
2c
2c
2c
2c
2c
2c
2a
2b
2d
2e
4
4
2
2
4
4
4
2
6
3.5
6
6
3.5
K2CO3 (4)
K2CO3 (4)
K2CO3 (4)
K2CO3 (4)
KHCO3 (1)
Ag2CO3 (5)
Ag2CO3 (5)
–
TBAB (100)
Ag2CO3 (3.6)
Ag2CO3 (3.6)
Ag2CO3 (3.6)
–
–
–
–
–
–
85
rt
rt
rt
rt
rt
rt
rt
−15
−15
−15
−20
−15
2
11
11
12
10.5
10.5
16
10
48
56 (0)
24 (37)
20 (35)
47 (4)
0
0
60 (7)
19 (1.7)
0
0
0
0
18.1 (1)
0
0
14 (0)
0
0
4
5b
6b
7b
8
KNa(C4H4O6) (4)e
Cs2CO3 (4)
0
28 (5)
51 (12)
52 (52)
71 (0)
52 (13)
42 (0)
37.5 (2.6)
(i-Pr)2NEt (1.2)
(i-Pr)2NEt (1.2)
(i-Pr)2NEt (1.2)
(i-Pr)2NEt (1.2)
(i-Pr)2NEt (1.2)
(i-Pr)2NEt (1.2)
9
10
11
12
13
96
55.5
55
66
a 6 Equiv. of PhI, 5 mol% Pd2(dba)3·CHCl3, and 20 mol% of 2 were used unless otherwise stated.
b 10 mol% of 2c was used.
c Isolated yield.
d E.e.% values were determined by HPLC using an OD chiral column (eluent: n-hexane:2-propanol=9:1).
e Sodium potassium tartarate was used as the base.
Table 2. The results of Pd(0) catalysed coupling cyclisation of 1 and aryl iodides
Ar
R2
R1
H
R2
.
Pd2(dba)3 CHCl3 (2.5 mol%),
2c (20 mol%)
R1
+ ArI
O
-15 oC
(i-Pr)2NEt, CH3CN,
COOH
O
3
1
H
R2
1a R1 = n-C7H15, R2 = H
1b R1 = n-C4H9, R2 = H
1c R1 = c-C6H11, R2 = H
1d R1 = Ph, R2 = Me
R1
O
O
3'
Entry
1
ArI (equiv.)
Time (days)
3a (%)
E.e.b (%)
1
2
3
4
5
6
7
8
1a
1a
1a
1a
1b
1b
1b
1b
1c
1d
Phenyl (6)
2
2
2
2.5
2.5
2.5
2.5
2.5
2.5
2.5
3a (52)
3b (51)
3c (52)
3d (53)
3e (47)
3f (44)
3g (53)
3h (51)
3i (36)
3j% (100)
52
43
49
47
50
53
53
46
0
4-Methylphenyl (4)
4-Methoxycarbonylphenyl (2)
4-Methoxyphenyl (2)
Phenyl (6)
4-Methylphenyl (4)
4-Methoxycarbonylphenyl (2)
4-Methoxyphenyl (2)
Phenyl (6)
9
10
Phenyl (1.2)
–
a Isolated yield.
b E.e.s were determined by HPLC.
the coupling reaction went smoothly with aromatic
iodides bearing either electron withdrawing or donating
groups, and afforded the corresponding butenolides in
similar yields and e.e.s. The length of the carbon chain
in the 4-position of the 2,3-allenoic acid was found to
instead of 2c as the ligand in this reaction, the results
were poor (Table 1, entries 10–13).
Using these reaction conditions, we examined the
behaviour of other substrates (Table 2). We found that

S. Ma et al. / Tetrahedron: Asymmetry 12 (2001) 193–195
195
Ph
n-Hep
.
Pd2(dba)3 CHCl3 (2.5 mol%)
n-Hep
2c (20 mol%)
(i-Pr)2NEt, CH3CN, -15 oC
+
PhI
O
COOH
O
(+)-3a
time (h)
yield (%)
e.e. (%)
8
20
30
48
192
17
30
34
57
76
47
51
52
51
29
Scheme 1.
have little effect on the yield and e.e. and additionally,
the e.e. can be affected greatly by steric hindrance from
the substituent group, such as c-hex (Table 2, entry 9).
When 2-methyl-4-phenyl-2,3-butenoic acid was used
only the cycloisomerisation product 3j% was formed
(Table 2, entry 10).
94978; (c) Lee, G. C. M. Eur. Pat. EP. 372,940, 1990;
Chem. Abstr. 1990, 113, 191137j; (d) Ducharme, Y.; Gau-
thier, J. Y.; Prasit, P.; Leblanc, Y.; Wang, Z.; Leger, S;
Thrien, M. PCT Int. Appl. WO 95,00,501, 1995; Chem.
Abstr. 1996, 124, 55954y; (e) Lee Gary, C. M.; Gast, M. E.
PCT Int. Appl. WO. 91 16,055, 1991: Chem. Abstr. 1992,
116, 59197m.
The e.e. of the product 3a did not change with length-
ened reaction time of 48 hours. However, after 8 days,
the reaction afforded (+)-3a in a higher yield (76%) but
with lower enantioselectivity (29% e.e.) (Scheme 1).
2. (a) Ma, S.; Shi, Z. J. Org. Chem. 1998, 63, 6387; (b) Ma,
S.; Duan, D.; Shi, Z. Org. Lett. 2000, 2, 1419; (c) Ma, S.;
Wu, S. J. Org. Chem. 1999, 64, 9314; (d) Ma, S.; Shi, Z.;
Yu, Z. Tetrahedron Lett. 1999, 40, 2393; (e) Ma, S.; Shi,
Z.; Yu, Z. Tetrahedron 1999, 40, 12137.
In conclusion, we have developed a silver(I) free reac-
tion for the coupling cyclisation reaction of aromatic
iodides and 2,3-allenoic acids.8,9 Based on this, the first
example of the catalytic enantioselective synthesis of
b-aryl butenolides was observed. According to the
results presented in this paper, the reaction is believed
to occur via carbopalladation of the allene to form a
p-allyl-palladium intermediate, followed by an enan-
tioselective intramolecular allylic substitution.2a Further
studies in this area are now being carried out in our
laboratory.
3. (a) Gawronski, J. K.; van Oeveren, A.; van der Deen, H.;
Leung, C. W.; Feringa, B. L. J. Org. Chem. 1996, 61, 1513;
(b) Gawronski, J. K.; Chen, Q.; Geng, Z.; Huang, B.;
Martin, M. R.; Mateo, A. I.; Brzostowska, M.; Rych-
lewska, U.; Feringa, B. Chirality 1997, 9, 537.
4. For such a reaction with Ag+, the direct cyclisation
catalysed by Ag+ followed by transmetallation and reduc-
tive elimination would make any enantioselective synthesis
of butenolides from racemic 2,3-allenoic acids impossible.
See Ref. 2a.
5. The ligands 2a–2e were prepared according to the known
method: Matt, P. Von; Lloyd-Jones, G. C.; Minidis, A. B.
E.; Pfaltz, A.; Macko, L.; Neuburger, M.; Zehnder, M.;
Ruegger, H.; Pregosin, P. S. Helv. Chim. Acta 1995, 78,
265.
Acknowledgements
6. (a) Marshall, A.; Wolf, M. A.; Wallace, E. M. J. Org.
Chem. 1997, 62, 367; (b) Marshall, A.; Bartley, S.; Wal-
lace, M. J. Org. Chem. 1996, 61, 5729; (c) Marshall, A.;
Hinkle, W. J. Org. Chem. 1996, 61, 4247.
7. Both the yield and e.e. value of 1a were determined by its
conversion to the corresponding ethyl ester via evapora-
tion of the solvent followed by addition of DMF, (i-
Pr)2NEt and EtI.
We thank the National Natural Science Foundation of
China (Project No. 29932020) and the Major State
Basic Research Development Program (Grant No.
G2000077500) for financial support.
References
8. Venkruijsse, H. D.; Brandsma, L. Synthesis of Acetylenes,
Allenes and Cumulenes. A Laboratory Manual; Elsevier:
Amsterdam, The Netherlands, 1981; p. 33.
1. (a) Brima, T. S. US 4,968,817, 1990; Chem. Abstr. 1991,
114, 185246y; (b) Tanabe, A. Jpn. Kokai. Tokyo. Koho
JP. 63,211,276 [88,211,276], 1988; Chem. Abstr. 1989, 110,
9. Bestmann, H. J.; Hartung, H. Chem. Ber. 1966, 99, 1198.
.