Allylic amination via decarboxylative C-N bond formation

Article abstract of DOI:10.1055/s-2005-918949

This manuscript details the development of a palladium-catalyzed allylic amination that proceeds via decarboxylation of allylic carbamates. Both saturated and aromatic heterocycles undergo decarboxylative rearrangement in good yields. The mechanism of allylation of heteroaromatic amines involves the formation of π-allyl palladium complexes followed by decarboxylation of the carbamate. Finally, the heteroaromatic anion equivalent is allylated to provide allylic amines. Georg Thieme Verlag Stuttgart.

Full text of DOI:10.1055/s-2005-918949

LETTER
2759
Allylic Amination via Decarboxylative C–N Bond Formation
D
ecarboxylati
h
A
m
e
ination lli R. Mellegaard-Waetzig, Dinesh Kumar Rayabarapu, Jon A. Tunge*
Department of Chemistry, University of Kansas, 1251 Wescoe Hall Drive, Lawrence, KS 66045-7582, USA
Fax +1(785)8645396; E-mail: tunge@ku.edu
Received 17 August 2005
ylation of allylic carbamates does occur under milder
conditions (25–70 °C) in the presence of palladium; how-
ever, reactions of the resulting amide nucleophiles are
presently limited to protonation9 and carbonylation.10 In
Abstract: This manuscript details the development of a palladium-
catalyzed allylic amination that proceeds via decarboxylation of
allylic carbamates. Both saturated and aromatic heterocycles under-
go decarboxylative rearrangement in good yields. The mechanism
of allylation of heteroaromatic amines involves the formation of p- addition, Tsuji has reported a single example of nickel-
allyl palladium complexes followed by decarboxylation of the
carbamate. Finally, the heteroaromatic anion equivalent is allylated
to provide allylic amines.
catalyzed decarboxylative allylic amination using
Ni[P(OEt)3]4 that likely proceeds through a p-allyl nickel
complex.11,12 Herein we report that Pd(0) and Ru(II) cata-
Key words: palladium-catalyzed, decarboxylation, allylic amina-
tion
lyze the decarboxylative formation of allylic amines from
allylic carbamates.
O
Ph
Ph
The formation of reactive intermediates through decar-
boxylation of ester derivatives is a potentially powerful
strategy for the formation of carbon–carbon as well as car-
bon–heteroatom bonds. Previously we and others have
shown that, in the presence of ruthenium(II) or palladi-
um(0) catalysts, CO2 is extruded from allyl b-keto esters
at room temperature, and the resulting enolates and allyl
fragments are coupled to form g,d-unsaturated ketones.1,2
Given the activity of these catalysts toward the decarbox-
ylative formation of enolates,3,4 we became curious
whether they could also effect the generation of nucleo-
philes other than enolates via decarboxylation. It is well-
known that allyl ethers can be generated by loss of CO2
from allyl carbonates,5 however, the analogous amination
reactions are rare. Thus, the development of catalytic de-
carboxylative allylic amination reactions is the focus of
this report (Scheme 1).6,7
NaH
N
O
NH
H
160–180 °C
H3C
H3C
Scheme 2
Initially, potential catalysts were screened for their ability
to mediate the conversion of allylic carbamate 1a to
allylic amines 2a and 3a (Scheme 3). Based on literature
precedent, it was anticipated that ruthenium(II) catalysts
would provide the branched product selectively while
palladium(0) would preferentially form the linear
product.13,14 For example, the Pd-catalyzed allylation of
morpholine with cinnamyl acetate provides the linear
allylic amine with moderate regioselectivity (84:16
2a:3a),14 while the ruthenium-catalyzed amination of
cinnamyl carbonate with piperidine favors formation of
the branched product (16:84).13
O
R2
Cp*Ru(bpy)Cl
Ph
N
R3
O
O
O
2a
O
R1
O
R3
+
Conditions
Ph
O
N
R2
O
O
R1
O
R2
R3
Pd(PPh3)4
1a
N
R1
Ph
3a
Scheme 1
Scheme 3
At the outset, it was known that allylic carbamates under-
go regio- and stereospecific thermal decarboxylative rear-
rangement under rather harsh conditions (Scheme 2).8
Our previous work suggests that such a process might be
catalyzed by palladium(0) or ruthenium(II) through for-
mation of p-allyl metal intermediates. Indeed, decarbox-
Initial catalyst screening shows that (pentamethylcyclo-
pentadienyl)ruthenium chloride tetramer [Cp*RuCl]4 is
an effective catalyst for the conversion of 1a to allylic
amines (Table 1), however, the regioselectivity is low.
Furthermore, in contrast to our previous observations, the
addition of 2,2¢-bipyridine (bpy) to [Cp*RuCl]4 produced
a catalyst that was less active than [Cp*RuCl]4 itself.2a
The best conditions for the formation of branched product
3a involve treatment of the allyl carbamate with
SYNLETT 2005, No. 18, pp 2759–2762
0
3
.1
1
.2
0
0
5
Advanced online publication: 12.10.2005
DOI: 10.1055/s-2005-918949; Art ID: S07805ST
© Georg Thieme Verlag Stuttgart · New York

2760
S. R. Mellegaard-Waetzig et al.
LETTER
Table 2 Palladium-Catalyzed Decarboxylative Allylic Aminationa
[Cp*RuCl]4 and tetramethylethylenediamine (TMEDA)
at 40 °C in CH2Cl2. Under these conditions, products 2a
and 3a were formed in a 1:3 ratio at 86% conversion.
Interestingly, prolonged standing in the presence of the
ruthenium–TMEDA catalyst resulted in equilibration of
the branched product 3a to the thermodynamically more
stable product 2a. This equilibration between 2a and 3a
was not observed with the analogous [Cp*RuCl]4/bpy
mixture. Finally, attention was turned to the use of a
palladium(0) catalyst. As expected, catalysis by Pd(PPh3)4
favors the formation of linear product 2a, with the highest
activity and regioselectivity (32:1) being realized in
CH2Cl2 at 40 °C.
Entry Reactant
Product
Yield
(%)b
O
1
91
Ph
N
Ph
O
N
O
2a
O
1a
2
68
87
O
O
Ph
Ph
N
Ph
Ph
2b
Ph
O
1b
N
Ph
Ph
3
N
2c
2d
Ph
O
N
1c
Table 1 Screening of Catalyst for Decarboxylative Amination
4
56c
84
O
O
Ph
Ph
NH
Entry
Conditionsa,b
Yield of 2a Yield of 3a
Ph
Ph
O
1d
N
H
(%)
(%)
5
Ph
Ph
1
[Cp*RuCl]4
40
60
toluene, 85 °C, 12 h
Ph
Ph
N
N
Ph
O
N
N
2e
O
O
1e
2
[Cp*RuCl]4
CH2Cl2, 40 °C, 30 min
38
36
34
25
95
92
97
62
64
66
75
5
6
95
79
Ph
O
Ph
Ph
O
1f
3
[Cp*RuCl]4
CH2Cl2, 40 °C, 24 h
2f
7
O
4c
5d
6
[Cp*RuCl]4, bpy
CH2Cl2, 40 °C, 4 d
N
O
O
N
2g
1g
O
[Cp*RuCl]4, TMEDA
CH2Cl2, 40 °C, 4 h
8
9
76
92
CH3
O
CH3
H3C
Ph
O
N
CH3
Ph
N
[Cp*RuCl]4, TMEDA
CH2Cl2, 40 °C, 3 d
O
O
O
1h
2h
CH3
O
CH3
7
Pd(PPh3)4
toluene, 85 °C, 12 h
8
O
N
N
1i
O
2i
CH3
8
Pd(PPh3)4
CH2Cl2, 40 °C, 1 h
3
10
66
CH3
O
O
1:1.3
N
N
a Reactions utilized 2.5 mol% [Cp*RuCl]4 and 10 mol% ligand
(where applicable) or 10 mol% Pd(PPh3)4.
b Ratios determined by 1H NMR spectroscopy.
c Conversion 88%.
1j
O
O
O
3j
2j
CH3
N
d Conversion 86%.
11
83
O
O
1k
N
N
Given the higher regioselectivity of the palladium-cata-
lyzed decarboxylative amination, this transformation was
chosen for further investigation of substrate scope. A
variety of tertiary carbamates including those derived
from anilines, cyclic amines, and acyclic amines undergo
the decarboxylative coupling in 1–4 hours at 40 °C in
CH2Cl2 upon treatment with 10 mol% Pd(PPh3)4
(Table 2). A secondary allylic amine could also be formed
under these conditions (Table 2, entry 4, 1d), however,
the yield was lower due to formation of a substantial
quantity of undesired diallylation product.
2k
O
O
12
13
78
–
O
O
N
N
O
O
1l
2l
Ph
O
N
CH3
H
O
N
CH3
Ph
1m
a All reactions were carried out under N2 with Pd(PPh3)4 (0.05 mmol)
and carbamate (0.500 mmol) in CH2Cl2 (2.0 mL) at 40 °C for 1–4 h.
b Isolated in >19:1 regioisomeric purity except where indicated.
c Diallylated product (20% yield) was also isolated.
A variety of allylic proelectrophiles are also compatible
with this amination chemistry. Cyclic, acyclic, aryl-sub-
stituted, and alkyl-substituted allylic carbamates all pro-
vide good yields of allylic amination products when
coupled with morpholine. However, attempts to couple a
cyclic p-allyl electrophile with a secondary aniline nu-
cleophile result in quantitative elimination to form cyclo-
hexadiene and N-ethyl aniline (Table 2, entry 13, 1m).
Synlett 2005, No. 18, 2759–2762 © Thieme Stuttgart · New York

LETTER
Decarboxylative Allylic Amination
2761
Table 3 Palladium-Catalyzed Decarboxylative Allylation of
After demonstrating the feasibility of decarboxylative
amination, the more difficult task of coupling hetero-
aromatic amines with allyl electrophiles was addressed.
Heteroaromatic amines are prevalent in pharmaceuticals,
so convenient methods for their allylation are desirable.15
Despite its potential utility, allylic amination using
heteroaromatic amines has not received as much attention
as the analogous allylation of alkyl amines.16,17
Heteroaromatic Aminesa
Entry Reactant
Product
Yield
(%)b
1
O
77
N
N
N
N
O
2n
1n
To begin, the decarboxylative coupling of cyclohexenyl
pyrazoyl carboxylate (1n) was investigated (Scheme 4).
2
3
4
90
88
O
N
N
N
N
N
N
N
O
1o
2o
2p
O
O
N
O
N
N
N
5 mol% Pd(PPh3)4
N
N
N
O
25 °C, CH2Cl2
4 h, 77%
1p
81
1n
2n
O
O
1:1.6
Scheme 4
N
N
N
R
N
R
N
N
N
N
N
2q'
1q
2q
Treatment of 1n with Pd(PPh3)4 in CD2Cl2 at room tem-
perature led to complete conversion to the allylic pyrazole
in 4 hours. A control reaction run in the absence of
palladium showed no reaction after 3 days under the same
conditions. Cyclohexenyl imidazole carboxylate and
cyclohexenyl 1,2,4-triazole carboxylate react similarly to
provide high yields of the allylic heterocyclic amines
within 1 hour at ambient temperature (Table 3). It is note-
worthy that the analogous Pd(PPh3)4-catalyzed reaction
between cyclohexenyl acetate and 1H-1,2,4-triazole gen-
erates a lower (78%) yield of allylic amination product
and requires 15 hours in refluxing THF for reaction com-
pletion.16b Using our methodology, benzotriazole could
also be coupled with allyl electrophiles to provide synthet-
ically useful allyl benzotriazoles,18 however, the regio-
selectivity is poor (entries 4, 8). The low regioselectivity
observed is consistent with other reported alkylations of
benzotriazole.8b,16c
5
6
7
8
–
O
HN
N
O
1r
71
65
O
N
N
N
N
N
O
2s
1s
O
N
N
O
N
2t
1t
79
1.1:1
O
O
N
N
R
N
N
R 2u'
N
N
N
N
N
1u
2u
a All reactions were carried out under N2 with Pd(PPh3)4 (0.039
mmol) and carbamate (0.78 mmol) in CH2Cl2 (5.0 mL) at 25 °C for
1–4 h.
In contrast to the other heteroaromatics, pyrrole was not a
suitable reaction partner for decarboxylative amination.
Treatment of allylic pyrrole carbamate (1r) with
Pd(PPh3)4 under our standard conditions resulted exclu-
sively in elimination to form cyclohexadiene and pyrrole
(Scheme 5). There are two reasonable explanations for
formation of cyclohexadiene rather than allylic amine
from 1r. First, if allylation preceeds decarboxylation, then
successful substrates require an additional nucleophilic
nitrogen (i.e. imidazole, Scheme 5). Alternatively the pre-
dominance of elimination may simply be a result of the
higher basicity of pyrrolide (pKa = 23)19 as compared to
the anions of imidazole (pKa = 18.6), pyrazole
(pKa = 19.8), and benzotriazole (pKa = ca. 12).
b Isolated yields.
To probe the mechanism of allylation of heteroaromatic
amines, the rearrangement of an unsymmetrical heteroar-
omatic allylic carbamate (1v) was performed (Scheme 6).
In this case, if nucleophilic attack occurs prior to decar-
boxylation, then the carbamate will function as a protect-
ing group and decarboxylative amination will produce
only the product of alkylation at the more hindered nitro-
gen (2v). Alternatively, if decarboxylation occurs prior to
alkylation, then a mixture of 2v and 2w can form. Decar-
boxylative amination of 1v under our standard conditions
produces a 1.1:1 mixture of 2v and 2w. The observation
of substantial quantities of 2w indicates that decarboxyla-
tion is occurring prior to allylation. Furthermore, it is
interesting to note that the major product is the result of
alkylation at the more hindered nitrogen. The analogous
reaction between cyclohexenyl acetate and 4-methyl
imidazole produces 2v and 2w as a 1:2 mixture favoring
alkylation at the less hindered nitrogen. Thus, decarbox-
N
O
O
N
N
O
N
PdLn
Pd0
N
O
N
+
– CO2
Scheme 5
Synlett 2005, No. 18, 2759–2762 © Thieme Stuttgart · New York

2762
S. R. Mellegaard-Waetzig et al.
LETTER
(3) (a) Lou, S.; Westbrook, J. A.; Schaus, S. E. J. Am. Chem.
Soc. 2004, 126, 11440. (b) Shim, J.-G.; Nakamura, H.;
Yamamoto, Y. J. Org. Chem. 1998, 63, 8470. (c) Nokami,
J.; Mandai, T.; Watanabe, H.; Ohyama, H.; Tsuji, J. J. Am.
Chem. Soc. 1989, 111, 4126.
ylative allylic amination provides different selectivities
than standard allylic amination, which may prove to be
beneficial for the synthesis of hindered heteroaromatic
amines.
(4) (a) Behenna, D. C.; Stoltz, B. M. J. Am. Chem. Soc. 2004,
126, 15044. (b) Trost, B. M.; Xu, J. J. Am. Chem. Soc. 2005,
127, 2846.
(5) (a) Guibe, F.; M’Leux, Y. S. Tetrahedron Lett. 1981, 22,
3591. (b) Bäckvall, J. E.; Nordberg, R. E.; Vaberg, J.
Tetrahedron Lett. 1983, 24, 411. (c) Larock, R. C.; Lee, N.
H. Tetrahedron Lett. 1991, 32, 6315. (d) Shim, J.-G.;
Yamamoto, Y. J. Org. Chem. 1998, 63, 3067.
CH3
O
N
N
5 mol%
Pd(PPh3)4
CH3
O
N
N
N
N
+
CD2Cl2
45 min
57%
CH3
1v
2v
2w
(6) Decarboxylative allylic amination starting with allylic
carbamates has been reported using stoichiometric
BF3·OEt2: Wang, C.-L. J.; Calabrese, J. C. J. Org. Chem.
1991, 56, 4341.
(7) The reverse reaction involving the coupling of amines, CO2,
and allyl halides to form allylic carbamates has been
reported: McGhee, W. D.; Riley, D. P.; Christ, M. E.; Christ,
K. M. Organometallics 1993, 12, 1429.
(8) (a) Synerholm, M. E.; Gilman, N. W.; Morgan, J. W.; Hill,
R. K. J. Org. Chem. 1968, 33, 1111. (b) Katritzky, A. R.;
Zhang, G.-F.; Fan, W.-Q.; Wu, J.; Pernak, J. J. Phys. Org.
Chem. 1993, 6, 567.
(9) (a) Cook, G. R.; Yu, H.; Sankaranarayanan, S.; Shanker, P.
S. J. Am. Chem. Soc. 2003, 125, 5115. (b) Cook, G. R.;
Shanker, P. S.; Pararajasingham, K. Angew. Chem. Int. Ed.
1999, 38, 110. (c) Jeffrey, P. D.; McCombie, S. W. J. Org.
Chem. 1982, 47, 587.
(10) Knight, J. G.; Ainge, S. W.; Harm, A. M.; Harwood, S. J.;
Maughan, H. I.; Armour, D. R.; Hollinshead, D. M.; Jaxa-
Chamiec, A. A. J. Am. Chem. Soc. 2000, 122, 2944.
(11) Tsuji, J.; Minami, I.; Shimizu, I. Chem. Lett. 1984, 1721.
(12) Our attempts to utilize Ni(0) catalysts for decarboxylative
allylation yielded mainly 1,5-dienes via allyl
homodimerization.
(13) Kondo, T.; Ono, H.; Satake, N.; Mitsudo, T.-a.; Watanabe,
Y. Organometallics 1995, 14, 1945.
Scheme 6
In conclusion, we have developed a catalytic synthesis of
allylic amines via decarboxylation of allylic carbamates.
This method is particularly well-suited for the decarbox-
ylative coupling of heteroaromatic amines with allyl elec-
trophiles. The allylation of heteroaromatic amines follows
a mechanism where decarboxylation preceeds alkylation.
This unique reaction feature may be responsible for pro-
ducing selectivities that are different from those observed
for amination of allylic acetate derivatives.
General Procedure for Decarboxylative Allylation of Saturated
Amines
To a dry, air-free 25-mL Schlenk flask was added Pd(PPh3)4 (0.05
mmol). The flask was then charged with substrate 1 (0.50 mmol)
and dry CH2Cl2 (2.0 mL). This solution was heated to 40 °C and al-
lowed to stir for 1–4 h. After cooling the solution to r.t., the mixture
was filtered through a short Celite and silica gel pad and washed
with CH2Cl2 (3 × 5 mL). The filtrate was concentrated and the resi-
due was purified on a silica gel column using hexane–EtOAc
(15:85) as eluent.
(14) Bergbreiter, D. E.; Weatherford, D. A. J. Org. Chem. 1989,
54, 2726.
General Procedure for Decarboxylative Allylation of Hetero-
aromatic Amines
(15) Katritzky, A. R.; Pozharskii, A. F. Handbook of
Heterocyclic Chemistry, 2nd ed.; Pergamon: Oxford, 2002.
(16) For leading examples of the allylation of heteroaromatic
amines see: (a) Trost, B. M.; Kuo, G.-H.; Benneche, T. J.
Am. Chem. Soc. 1988, 110, 621. (b) Arredondo, Y.;
Moreno-Mañas, M.; Pleixats, R.; Villarroya, M.
Tetrahedron 1993, 49, 1465. (c) Arnau, N.; Arredondo, Y.;
Moreno-Mañas, M.; Pleixats, R.; Villarroya, M. J.
Heterocycl. Chem. 1995, 32, 1325. (d) Trost, B. M.;
Krische, M. J.; Berl, V.; Grenzer, E. M. Org. Lett. 2002, 4,
2005. (e) Konkel, M. J.; Vince, R. J. Org. Chem. 1996, 61,
6199.
To a dry, air-free 25-mL Schlenk flask was added substrate (0.780
mmol), Pd(PPh3)4 (0.039 mmol, 5 mol%), and dry CH2Cl2 (5 mL).
The resulting colorless solution was allowed to stir at r.t. for 1–12
h. Reaction completion was generally indicated by a change in color
to bright yellow. The solution was then concentrated and directly
purified by flash column chromatography using hexane–EtOAc
(90:10) as the eluent. The imidazole-containing products were
triturated with hexane and filtered prior to chromatography.
Acknowledgment
(17) (a) Johannsen, M.; Jorgensen, K. A. Chem. Rev. 1998, 98,
1689. (b) Watson, I. D. G.; Styler, S. A.; Yudin, A. K. J. Am.
Chem. Soc. 2004, 126, 5086. (c) Ohmura, T.; Hartwig, J. F.
J. Am. Chem. Soc. 2002, 124, 15164. (d) Feuerstein, M.;
Laurenti, D.; Doucet, H.; Santelli, M. Chem. Commun. 2001,
43.
(18) (a) Katritzky, A. R.; Wang, X.; Denisenko, A. J. Org. Chem.
2001, 66, 2850. (b) Katritzky, A. R.; Qi, M. J. Org. Chem.
1997, 62, 4116. (c) Le, Z.-G.; Chen, Z.-C.; Hu, Y.; Zheng,
Q.-G. Heterocycles 2004, 63, 1077. (d) Mashraqui, S. H.;
Kumar, S. M.; Chandrasekhar, D. Bull Chem. Soc. Jpn.
2001, 74, 2133.
We thank the Petroleum Research Fund (39562-G1) for partial
support of this work.
References
(1) (a) Tsuda, T.; Chujo, Y.; Nishi, S.-I.; Tawara, K.; Saegusa,
T. J. Am. Chem. Soc. 1980, 102, 6381. (b) Shimizu, I.;
Yamada, T.; Tsuji, J. Tetrahedron Lett. 1980, 21, 3199.
(2) (a) Burger, E. C.; Tunge, J. A. Org. Lett. 2004, 6, 2603.
(b) Burger, E. C.; Tunge, J. A. Org. Lett. 2004, 6, 4113.
(c) Burger, E. C.; Tunge, J. A. Chem. Commun. 2005, 2835.
(d) Burger, E. C.; Tunge, J. A. Eur. J. Org. Chem. 2005,
1715.
(19) For the pKa in DMSO, see: Bordwell, F. G.; Drucker, G. E.;
Fried, H. E. J. Org. Chem. 1981, 46, 632.
Synlett 2005, No. 18, 2759–2762 © Thieme Stuttgart · New York