Complete retention of Z geometry in allylic substitution catalyzed by an iridium complex

Article abstract of DOI:10.1021/ol990033n

(equation presented) The Z geometry of methyl (Z)-3-monosubstituted-2-alkenyl carbonate was completely retained in iridium complex-catalyzed allylic amination. The reaction of methyl (Z)-2-nonenyl carbonate with piperidine in the presence of a catalytic amount of [Ir(COD)Cl]₂ and P(OPh)₃ at 50°C for 2 h gave a 98:2 mixture of (Z)-1-(2-nonenyl)piperidine and 1-(1-n-hexyl-2-propenyl)piperidine in 86percent yield. No E isomer was obtained. Various (Z)-allylic amines were obtained in 91-100percent selectivity by allylic amination of methyl (Z)-3-monosubstituted-2-alkenyl carbonate.

Full text of DOI:10.1021/ol990033n

ORGANIC
LETTERS
1999
Vol. 1, No. 2
265-267
Complete Retention of Z Geometry in
Allylic Substitution Catalyzed by an
Iridium Complex
Ryo Takeuchi* and Norihito Shiga
Department of EnVironmental Science, Faculty of Science and Graduate School of
Integrated Science, Yokohama City UniVersity, 22-2 Seto, Kanazawa-ku,
Yokohama 236-0027, Japan
rtakeuch@yokohama-cu.ac.jp
Received April 13, 1999
ABSTRACT
The Z geometry of methyl (Z)-3-monosubstituted-2-alkenyl carbonate was completely retained in iridium complex-catalyzed allylic amination.
The reaction of methyl (Z)-2-nonenyl carbonate with piperidine in the presence of a catalytic amount of [Ir(COD)Cl]2 and P(OPh)3 at 50 °C for
2 h gave a 98:2 mixture of (Z)-1-(2-nonenyl)piperidine and 1-(1-n-hexyl-2-propenyl)piperidine in 86% yield. No E isomer was obtained. Various
(Z)-allylic amines were obtained in 91−100% selectivity by allylic amination of methyl (Z)-3-monosubstituted-2-alkenyl carbonate.
The chemistry of (π-allyl)metal complexes is central in
organic synthesis.1 The reaction of a (π-allyl)metal complex
with a nucleophile provides various synthetically useful
processes. One of the important aspects of the chemistry of
(π-allyl)metal complexes is syn-anti isomerization2 via a
π-σ-π process which greatly affects the regio- and ste-
reoselectivity of allylic substitution. Nucleophilic attack at
the unsubstituted allylic terminus of a terminally monosub-
stituted syn π-allyl complex gives an (E)-alkene, while that
of an anti π-allyl complex gives a (Z)-alkene.
A wide variety of transition-metal complexes are reported
to be a catalyst for allylic substitution. Of the metal
complexes studied, palladium is the most general and
versatile catalyst.3 It is well-known that a syn (π-allyl)-
palladium complex is thermodynamically more stable than
an anti complex.4 (π-Allyl)palladium complexes undergo
a rapid syn-anti isomerization prior to the nucleophilic
attack. Thus, palladium complex catalyzed allylic substitution
of (Z)-3-monosubstituted-2-alkenyl esters resulted in the loss
of Z geometry.5 A mixture of (E)- and (Z)-alkenes is
obtained. A limited example of the retention of Z geometry
has been reported.6 Åkermark found that 2,9-disubstituted-
1,10-phenanthroline could induce a preference for the anti
configuration and reported Z-selective allylic alkylation of
(Z)-2-hexenyl acetate.7 (Z)-Alkene was obtained in moderate
(4) (a) Faller, J. W.; Mattina, M. J. Inorg. Chem. 1972, 11, 1296. (b)
Faller, J. W.; Thomsen, M. E.; Mattina, M. J. J. Am. Chem. Soc. 1971, 93,
2642.
(5) (a) Hayashi, T.; Kishi, K.; Yamamoto, A.; Ito, Y. Tetrahedron Lett.
1990, 31, 1743. (b) Trost, B. M.; Verhoeven, T. R. J. Am. Chem. Soc. 1980,
102, 4730. (c) Trost, B. M.; Keinan, E. J. Org. Chem. 1979, 44, 3451.
(6) (a) Hutzinger, M. W.; Oehlschlager, A. C. J. Org. Chem. 1991, 56,
2918. (b) Luo, R.-T.; Negishi, E. J. Org. Chem. 1985, 50, 4762. (c) Luo,
R.-T.; Negishi, E. Tetrahedron Lett. 1985, 26, 2177.
(7) (a) Sjo¨gren, M. P. T.; Hansson, S.; Åkermark, B.; Vitagliano, A.
Organometallics 1994, 13, 1963. (b) Sjogren, M.; Hansson, S.; Norrby, P.;
Åkermark, B.; Cucciolito, M. E.; Vitagliano, A. Organometallics 1992, 11,
3954. (c) Åkermark, B.; Hansson, S.; Vitagliano, A. J. Am. Chem. Soc.
1990, 112, 4587.
(1) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. Principles
and Applications of Organotransition Metal Chemistry; University Science
Books: Mill Valley, CA, 1987; p 881.
(2) (a) Crabtree, R. H. The Organometallic Chemistry of the Transition
Metals; Wiley: New York, 1994; p 112. (b) Reference 1, p 175. (c) Ward,
Y. D.; Villanueva, L. A.; Allred, G. D.; Payne, S. C.; Semones, M. A.;
Liebeskind, L. S. Organometallics 1995, 14, 4132.
(3) Tsuji, J. Palladium Reagents and Catalysts; Wiley: New York, 1995;
p 290.
10.1021/ol990033n CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/29/1999

to good selectivity, but the formation of (E)-alkene could
not be suppressed completely.
Table 1. Reaction of 1a with Piperidinea
We reported that an iridium complex is a new and efficient
catalyst for allylic alkylation.8 The regioselectivity of iridium
complex catalyzed allylic alkylation depends on the geometry
of the allylic system, i.e., the regioselectivity of alkylation
of a syn (π-allyl)iridium intermediate is different from that
of an anti (π-allyl)iridium intermediate. The alkylation
occurred prior to syn-anti isomerization. This result prompted
us to investigate Z-selective allylic substitution. We first
succeeded with the complete retention of Z geometry in the
allylic amination5a,c,9 of (Z)-3-monosubstituted-2-alkenyl
carbonate (Scheme 1).
entry
(Z)-3a /
no. substrate catalystb P/Ir time/h yield/%c (E)-3a /4a d
1
2e
3
4
5
6
7
8
9f
10g
1a 1
1a 1
1a 1
1a 1
1a 1
1a 1
1a 1
1a 2
1a 1
1a 1
A/P(OPh)3
A/P(OPh)3
A/P(OEt)3
A/PPh3
2
2
2
2
0
1
3
2
2
2
2
24
24
24
23
2
4
24
2
86
14
7
31
15
88
69
43
90
83
98/0/2
92/0/8
96/0/4
8/78/14
100/0/0
98/0/2
97/0/3
97/0/3
97/0/3
97/0/3
A
A/P(OPh)3
A/P(OPh)3
A/P(OPh)3
B/P(OPh)3
C/P(OPh)3
3
a A mixture of 1a (2 mmol), [Ir(COD)Cl]2 (0.04 mmol), ligand, and
piperidine (5 mL) was stirred under argon at 50 °C. b Catalyst legend: A,
[Ir(COD)Cl]2: B, [Ir(COD)2]BF4; C, [Ir(COD)OMe]2. c Isolated yield.
d Determined by GLC. e Piperidine (10 mmol) in refluxing THF (5 mL).
f Catalyst 0.04 mmol. g Catalyst 0.04 mmol.
Scheme 1
was necessary for high yield. The reaction of 5 equiv of
piperidine with 1a1 at reflux in THF for 24 h gave products
in 14% yield (entry 2). The starting material was recovered
in 80% yield. P(OPh)3 was the most efficient ligand. The
use of P(OEt)3 and PPh3 decreased the yield of products
(entries 3 and 4). The ratio of P to Ir was also important.
The reaction of 1 or 2 equiv of P(OPh)3 with Ir gave products
in good yield (entries 1 and 6), while that of 3 equiv of
P(OPh)3 with Ir resulted in a slight decrease of the yield
(entry 7). [Ir(COD)2]BF4/P(OPh)3 and [Ir(COD)OMe]2/
P(OPh)3 showed comparable catalytic activities and selectivi-
ties (entries 9 and 10). Carbonate was more reactive than
acetate. The reaction of (Z)-2-nonenyl acetate (1a2) with
piperidine gave a decrease in the yield of products at
prolonged reaction time (entry 8).
The reactions of 1a1 with various amines were examined
(Table 2). Amines were used as a solvent. Allylic amines
Table 2. Reaction of 1a1 with Aminesa
entry no.
amine
product
yield/%b (Z)-3/(E)-3/4c
1d
2e
3f
4g
pyrrolidine
3b, 4b
3c, 4c
3d , 4d
3e, 4e
71
83
62
23
94/0/6
95/0/5
98/0/2
96/0/4
morpholine
diethylamine
n-butylamine
a A mixture of 1a1 (2 mmol), [Ir(COD)Cl]2 (0.04 mmol), P(OPh)3 (0.16
mmol), and amine (5 mL) was stirred under argon at 50 °C. b Isolated yield.
c Determined by GLC. d Reacted for 3 h. e Reacted for 24 h. f [Ir(COD)Cl]2
(0.08 mmol) and P(OPh)3 (0.32 mmol). Reacted for 4 h. g Reacted for 3 h.
The reaction of methyl (Z)-2-nonenyl carbonate (1a1) with
piperidine in the presence of a catalytic amount of [Ir(COD)-
Cl]2 and P(OPh)3 gave a mixture of (Z)-3a and 4a. The
reaction using piperidine as a solvent at 50 °C for 2 h gave
a 98:2 mixture of (Z)-3a and 4a in 86% yield (Table 1; entry
1).10 No E isomer was obtained. A large excess of piperidine
(Z)-3b-e were obtained in 94-98% selectivity. No E isomer
was obtained. The structure of the amine had a substantial
(8) (a) Takeuchi, R.; Kashio, M. J. Am. Chem. Soc. 1998, 120, 8647.
(b) Takeuchi, R.; Kashio, M. Angew. Chem., Int. Ed. Engl. 1997, 36, 263.
(9) For Pd complex catalyzed allylic amination, see: (a) Åkermark, B.;
Åkermark, G.; Hegedus, L. S.; Zetterberg, K. J. Am. Chem. Soc. 1981,
103, 3037. (b) Stakem, F. G.; Heck, R. F. J. Org. Chem. 1980, 45, 3584.
(10) General Procedure for Allylic Amination of Allylic Esters. A
typical procedure is described for the reaction of 1a1 with piperidine. A
mixture of methyl (Z)-2-nonenyl carbonate (1a1; 400 mg, 2.0 mmol),
triphenyl phosphite (49.6 mg, 0.16 mmol), [Ir(COD)Cl]2 (26.9 mg, 0.04
266
Org. Lett., Vol. 1, No. 2, 1999

effect on the yield of products. The reactions of 1a1 with
secondary amines gave the corresponding products in good
yield (entries 1-3). In contrast to results obtained in the
reactions with secondary amines, the reaction of 1a1 with
n-butylamine decreased the yield considerably (entry 4).
Bulky amines are therefore necessary for high yields of (Z)-
3.11
As shown in Table 3, (Z)-1b-f were successfully reacted
with piperidine to give (Z)-3f-j with high selectivities. No
E isomer was obtained. Terminal alkene was tolerated in
of 1e with piperidine for 22 h gave (Z)-3i exclusively in 33%
yield. Increasing the amount of the catalyst to 8 mol %
accelerated the reaction to give (Z)-3i in 70% yield (entry
4). The alkoxy functionality was also tolerated and did not
alter the regio- and stereoselectivity of the allylic amination;
allylic amine (Z)-3j was obtained with 98% selectivity (entry
5).
As seen in allylic alkylations,8a the selectivity of the allylic
amination depends on the geometry of the allyl system. The
reaction of (E)-1a1 with piperidine at 50 °C for 2 h gave a
28:1:71 mixture of (E)-3a, (Z)-3a, and 4a in 93% yield. The
syn-(π-allyl)- and anti-(π-allyl)iridium intermediates gave
different results. This difference in regioselectivity is reason-
ably explained as follows. When the amine approaches the
substituted allylic terminus of the anti-(π-allyl)iridium
intermediate, the substituent and iridium moiety are close
together and thereby increase steric repulsion (Scheme 2).12
Table 3. Reaction of 1 with Piperidinea
entry no. substrate product time/h yield/%b (Z)-3/(E)-3/4c
1
2
3
4d
1b
1c
1d
1e
1f
3f, 4f
3g, 4g
3h , 4h
3i, 4i
3j, 4j
8
15
3
5
4
83
81
89
70
78
94/0/6
91/0/9
100/0/0
100/0/0
98/0/2
5
Scheme 2
a A mixture of 1 (2 mmol), [Ir(COD)Cl]2 (0.04 mmol), P(OPh)3 (0.16
mmol), and piperidine (5 mL) was stirred under argon at 50 °C. b Isolated
yield. c Determined by GLC. d [Ir(COD)Cl]2 (0.08 mmol) and P(OPh)3 (0.32
mmol).
the allylic amination. The reaction of 1d with piperidine was
regiospecific to give (Z)-3h in 89% yield (entry 3). The steric
congestion around the carbon-carbon double bond affected
the reaction time and the yield of the product. The reaction
The transition state of the amination at the substituted allylic
terminus of the anti-(π-allyl)iridium intermediate is therefore
less stable than that of the syn-(π-allyl)iridium intermediate.
Thus, amination of the anti-(π-allyl)iridium intermediate
preferentially occurs at the unsubstitued allylic terminus.
In summary, we have succeeded with the complete
retention of Z geometry in the allylic amination. This unique
feature of iridium catalysis will have wide synthetic applica-
tion.
mmol), and piperidine (5.0 mL) was stirred at 50 °C for 2 h under an Ar
atmosphere. The progress of the reaction was monitored by GLC. After
1a1 was consumed, the reaction mixture was diluted with ether. The ethereal
solution was extracted with 6 M HCl. The combined acidic layers were
neutralized with NaOH and extracted with ether. The organic layer was
dried with MgSO4 and filtered. After evaporation of the solvent, the residue
was purified by column chromatography (n-hexane/ethyl acetate (70/30))
to give (Z)-3a and 4a (360 mg; yield 86%). (Z)-1-(2-Nonenyl)piperidine
((Z)-3a): 1H NMR (400 MHz, C6D6) δ 1.00 (t, J ) 7.0 Hz, 3H), 1.30-
1.51 (m, 10H), 1.66 (quintet, J ) 5.1 Hz, 4H), 2.19 (q, J ) 7.1 Hz, 2H),
2.48 (t, J ) 5.1 Hz, 4H), 3.09 (d, J ) 6.7 Hz, 2H), 5.64 (dtt, J ) 11.0, 7.3,
1.6 Hz, 1H), 5.77 (dtt, J ) 11.0, 6.7, 1.5 Hz, 1H); 13C NMR (67.8 MHz,
CDCl3) δ 14.0, 22.6, 24.4, 26.0 (2C), 27.4, 28.9, 29.5, 31.7, 54.5 (2C),
55.9, 126.4, 132.8. Anal. Calcd for C14H27N: C, 80.31; H, 13.00; N, 6.69.
Found: C, 80.07; H, 13.01; N, 6.65. 1-(1-n-Hexyl-2-propenyl)piperidine
(4a): Compound 4a could not be isolated in pure form. A partial 1H NMR
spectrum was obtained from the mixture of (Z)-3a. 1H NMR (400 MHz,
C6D6): δ 5.12 (dd, J ) 17.2, 2.2 Hz, 1H), 5.21 (dd, J ) 10.3, 2.2 Hz, 1H),
5.83 (ddd, J ) 17.2, 10.3, 6.9 Hz, 1H).
Acknowledgment. This work was supported by a Grant-
in-Aid for Scientific Research on Priority Areas (No. 283,
“Innovative Synthetic Reactions”) from the Ministry of
Education, Science, Sports and Culture, Government of
Japan.
Supporting Information Available: Text giving experi-
mental procedures and compound characterization data. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(11) Cone angles of secondary amines are larger than those of primary
amines; see: Seligson, A. L.; Trogler, W. C. J. Am. Chem. Soc. 1991, 113,
2520.
(12) See reference 8a and references therein.
OL990033N
Org. Lett., Vol. 1, No. 2, 1999
267