Iridium-catalyzed asymmetric intramolecular allylic amidation: Enantioselective synthesis of chiral tetrahydroisoquinolines and saturated nitrogen heterocycles

Article abstract of DOI:10.1002/anie.201006039

For the first time iridium catalysis has been used for the synthesis of chiral tetrahydroisoquinolines with excellent yields and high enantioselectivities (see scheme; cod=1,5-cyclooctadiene, DBU=1,8- diazabicyclo[5.4.0]undec-7-ene). These products are important chiral building blocks for the synthesis of biologically active compounds, in particular alkaloids. Copyright

Full text of DOI:10.1002/anie.201006039

Communications
DOI: 10.1002/anie.201006039
Asymmetric Catalysis
Iridium-Catalyzed Asymmetric Intramolecular Allylic Amidation:
Enantioselective Synthesis of Chiral Tetrahydroisoquinolines and
Saturated Nitrogen Heterocycles**
Johannes F. Teichert, Martꢀn Faꢁanꢂs-Mastral, and Ben L. Feringa*
Tetrahydroisoquinolines represent a large class of natural
compounds with interesting and diverse biological proper-
ties.[1,2] Therefore, these heterocycles are important targets
for organic synthesis and much effort has been directed
towards the development of efficient enantioselective routes
to prepare chiral tetrahydroisoquinolines.[3–8] The methods
À
investigated thus far include palladium-catalyzed C H acti-
vation of arylethylamines,[9] transition-metal-catalyzed hydro-
Scheme 1. Retrosynthetic approach and twofold use of the trifluoroace-
tamide group.
genation of imines[10] or heteroaromatic compounds[7], as well
as Lewis acid promoted ionic cyclizations,[11] and organo-
catalytic Mannich reactions.[12] However, these procedures
are limited by the fact that either electron-rich phenylethyl-
amine derivatives are required, or only alkyl groups can be
introduced at the stereogenic center, or a number of steps are
required to reach the unprotected tetrahydroisoquinoline.
The iridium-catalyzed asymmetric allylic substitution[13–18]
with phosphoramidites[19–23] as chiral ligands represents a
powerful synthetic method, which has found application in a
wide variety of natural product syntheses.[13,19] One major
advantage of asymmetric iridium-catalyzed allylic substitu-
tion is its tolerance towards a large variety of nucleophiles,
including ammonia.[24–28] The use of amides as nucleophiles
for these transformations has only been reported for potas-
sium trifluoroacetamide as ammonia surrogate[24] or in allylic
amidation reactions through decarboxylative pathways.[29,30]
Furthermore, using amides as nucleophiles has the advantage
that alkylamine chains can be introduced at a stereogenic
center through iridium-catalyzed allylic amidation.
Herein, we report a new catalytic asymmetric approach
towards chiral tetrahydroisoquinolines and saturated chiral
nitrogen heterocycles. Specifically, a synthetic protocol for an
intramolecular iridium-catalyzed allylic amidation reaction to
construct chiral tetrahydroisoquinolines is presented
(Scheme 1). This transformation should serve as a reliable
method to access these valuable chiral building blocks for the
synthesis of natural products, as it furnishes a terminal olefin
as well as a secondary amine after removal of the trifluoro-
acetic acid group. Both of these functionalities serve as the
basis for further facile functionalization.
The usefulness of the trifluoroacetamide group in our
approach is twofold. First, it serves as a protecting group
during the synthesis of the allylic carbonates, whose key step
depends on a palladium-catalyzed cross-coupling reaction
that leads to the introduction of the allylic moiety
(Scheme 1).[31] Here, unprotected amines are generally not
accepted. Second, the amide serves as the actual nucleophile
of the iridium-catalyzed allylic substitution, which furnishes
the tetrahydroisoquinoline core. An important property of
the trifluoroacetamide group is that the secondary amine
moiety can easily be deprotected without jeopardizing the
adjacent sensitive allylic–benzylic stereocenter.[32]
We set off to investigate the influence of bases on the
catalytic system, which comprised of preformed iridacycle 3[15]
in THF at 508C. The choice of base was found to be highly
influential for the conversion and enantioselectivity of allylic
carbonate 1[33] into the corresponding protected chiral tetra-
hydroisoquinolines 2 (Table 1). The use of DBU resulted in
70% conversion and 81% ee (Table 1, entry 1),[34] while other
organic bases such as TBD and DABCO, which have been
used earlier in combination with Ir catalysts for allylic
substitutions,[35,36] led to significantly lower conversions and
enantioselectivities (Table 1, entries 2 and 5). Inorganic bases
such as K3PO4 and Cs2CO3 (Table 1, entries 3 and 4)
performed similarly, with disappointingly low conversions
and enantioselectivities.
We then went on to optimize the catalytic system. With
the preformed iridacycle 3, at elevated temperatures (508C)
the reaction did not reach full conversion overnight, and the
desired tetrahydroisoquinoline 2 was isolated in only 33%
yield (Table 2, entry 1). However, we were delighted to find
that the in situ formed iridacycle, which was prepared from
catalytic amounts of the phosphoramidite ligand L1 and
[{Ir(cod)Cl}2], showed a higher activity and led to full
conversion with similar enantioselectivities (83% ee;
[*] J. F. Teichert, Dr. M. Faꢀanꢁs-Mastral, Prof. Dr. B. L. Feringa
Stratingh Institute for Chemistry, University of Groningen
Nijenborgh 4, 9747 AG, Groningen (The Netherlands)
Fax: (+31)50-363-4296
E-mail: b.l.feringa@rug.nl
[**] M.F.-M. thanks the Spanish Ministry of Science and Innovation
(MICINN) for a postdoctoral fellowship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201006039.
688
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 688 –691

Table 1: Base screening.[a]
observed at room temperature (Table 2, entries 4 and 5).
Notably, when the corresponding acetamide was subjected to
the optimized reaction conditions, no allylic amidation
occurred even at elevated temperatures (Table 2, entry 6),
thus indicating that trifluoroacetamides possess ideal elec-
tronic and/or acidic requirements for the asymmetric trans-
formation envisaged.
To probe the substrate scope of the catalytic system, a
collection of chiral tetrahydroisoquinolines was synthesized,
which carried the most common substitution patterns seen in
natural products.[1,2] Furthermore, the method was extended
to a number of saturated nitrogen heterocycles of various ring
sizes (Table 3). Tetrahydroisoquinolines with donor substitu-
Entry
Base
Conversion [%][b]
ee [%][c]
1
2
3
4
5
DBU
TBD
K3PO4
Cs2CO3
DABCO
70
0
10
15
10
81
n.d.
19
22
31
Table 3: Product scope of intramolecular allylic amidation.[a]
Entry
Product
T [8C] Yield [%][b]
ee [%][c]
[a] Reaction conditions: 1 (1.0 equiv, 16.9 mg, 0.05 mmol), 3 (5.0 mol%,
3.45 mg, 0.0025 mmol), and base (1.0 equiv, 0.05 mmol) were dissolved
in THF (1 mL) and stirred under N2 at 508C for 20 h. [b] Determined by
1H NMR spectroscopy. [c] Determined by HPLC analysis using a chiral
stationary phase; see the Supporting Information. DABCO=1,4-
Diazabicyclo[2.2.2]octane, DBU=1,8-Diazabicyclo[5.4.0]undec-7-ene,
TBD=1,5,7-Triazabicyclo[4.4.0]dec-5-ene.
1
2
5
6
RT
RT
RT
97
89
92
78
95
2
3
94
91
94
Table 2: Catalyst optimization.[a]
4
5
6
7
8
9
RT
50
50
56 [d] (100% conv.)[e] 96
68 [d] (100% conv.)[e] 88
7[f]
10 50
25 [d] (100% conv.)[e] 92
[a] See the Experimental Section. [b] Yield of isolated product. [c] Deter-
mined by HPLC analysis using a chiral stationary phase; see the
Supporting Information. [d] Products are volatile. [e] Determined by
1H NMR spectroscopy. [f] A side reaction was observed, see Scheme 2.
Entry
R
Catalyst
T [8C]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
CF3 (1)
CF3 (1)
CF3 (1)
CF3 (1)
CF3 (1)
CH3 (4)
3
50
50
RT
50
RT
50
33 [e]
73
81
83
80
94
95
n.d.
Ir/L1
Ir/L1
Ir/L2
Ir/L2
Ir/L2
55 [d]
82
90
n.d.
ents, such as methoxy, dioxo, and methyl groups (2, 5, and 6;
Table 3, entries 1–3) were all synthesized in very good yields
and with excellent enantioselectivities ranging from 91–
95%.[38,39] Furthermore, the unsubstituted tetrahydroisoqui-
noline 7 could be isolated with similarly good results (78%,
94% ee; Table 3, entry 4).
We were interested in expanding the method of the
asymmetric intramolecular allylic amidation to the synthesis
of other chiral nitrogen-containing heterocycles. Along these
lines, five-, six-, and seven-membered chiral heterocycles 8–10
were synthesized (Table 3, entries 5–7). For these reactions to
proceed smoothly, elevated temperatures of 508C were
needed to ensure full conversion. In all cases, however, very
good to excellent enantioselectivities (up to 96% ee) were
found, thus demonstrating the versatility of our new catalytic
transformation.
[a] Reaction conditions: 1 or 4 (1.0 equiv, 0.05 mmol), Ir catalyst
(5.0 mol%, 0.0025 mmol), and DBU (1.0 equiv, 8 mL, 0.05 mmol) were
dissolved in THF (1 mL) and stirred under N2 at the indicated
temperature until full conversion (as evidenced by TLC). [b] Yield of
isolated product. [c] Determined by HPLC analysis using a chiral
stationary phase; see the Supporting Information. [d] Reaction did not
reach full conversion.
Table 2, entry 2). Lowering the temperature (RT) did not
affect the enantioselectivity but again resulted in incomplete
conversion (Table 2, entry 3). Turning to the related methoxy-
substituted phosphoramidite L2,[37] we found the product of
the intramolecular asymmetric allylic amidation in excellent
enantioselectivities (95% ee), with an even higher yield
Angew. Chem. Int. Ed. 2011, 50, 688 –691
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
689

Communications
Notably, when we investigated the synthesis of chiral
Experimental Section
azepane 10 (Table 3, entry 7), an unexpected side reaction
occurred. When allylic carbonate 11 was reacted under the
optimized reaction conditions (Scheme 2), chiral azepane 10
was found along with linear diene 12 as the major product.
The product distribution seemed to be independent of
temperature and amounts of DBU employed.[40] This side
reaction was only observed for 10 and not in the synthesis of
piperidine 9 or pyrrolidine 8, thus indicating that the
mechanism of the formation of the seven-membered ring
involves special spatial constraints.[41]
General procedure for the asymmetric allylic amidation reaction
(Table 3): [{Ir(cod)Cl}2] (2.5 mol%, 3.36 mg, 5.0 mmol) and L2
(5.0 mol%, 6.00 mg, 10.0 mmol) were dissolved in dry THF (1 mL)
under N2. Then, DBU (1.00 equiv, 0.03 mL, 0.20 mmol) was added
and the reaction mixture was heated at 508C for 30 min. Then, it was
brought to the appropriate temperature and the allylic carbonate
(1.0 equiv, 0.20 mmol) was added. The reaction mixture was stirred
until full conversion was achieved (as evident by TLC). All volatile
components were removed under reduced pressure to yield the crude
product as an orange oil. This was purified by column chromatog-
raphy on silica gel to yield the desired trifluoroacetamide.
Received: September 27, 2010
Keywords: allylic compounds · asymmetric catalysis · iridium ·
.
nitrogen heterocycles · phosphoramidites
[1] P. M. Dewick, Medicinal Natural Products—A Biosynthetic
Approach, 2nd ed., Wiley, Chichester, 2001.
[2] J. D. Scott, R. M. Williams, Chem. Rev. 2002, 102, 1669.
[3] M. Chrzanowska, M. D. Rozwadowska, Chem. Rev. 2004, 104,
3341.
Scheme 2. Observed side reaction. cod=1,5-cyclooctadiene.
[4] P. Siengalewicz, U. Rinner, J. Mulzer, Chem. Soc. Rev. 2008, 37,
2676.
[5] J. L. Vicario, D. Badꢀa, L. Carrillo, J. Etxebarria, Curr. Org.
Chem. 2003, 7, 1775.
The ease of protecting-group removal, that is, the chiral
secondary trifluoroacetamide, was demonstrated with the
conversion of 2 into 13 (Scheme 3). The product was simply
stirred at room temperature in the presence of an excess
[6] J. Royer, M. Bonin, L. Micouin, Chem. Rev. 2004, 104, 2311.
[7] Y. G. Zhou, Acc. Chem. Res. 2007, 40, 1357.
[8] For a single example of palladium-catalyzed allylic amidation
with 88% ee, see: K. Ito, S. Akashi, B. Saito, T. Katsuki, Synlett
2003, 1809.
[9] J. J. Li, T. S. Mei, J. Q. Yu, Angew. Chem. 2008, 120, 6552; Angew.
Chem. Int. Ed. 2008, 47, 6452.
[10] J. Szawkalo, S. J. Czarnocki, A. Zawadzka, K. Wojtasiewicz, A.
Leniewski, J. K. Maurin, Z. Czarnocki, J. Drabowicz, Tetrahe-
dron: Asymmetry 2007, 18, 406.
[11] S. S. Kinderman, M. M. T. Wekking, J. H. van Maarseveen, H. E.
Schoemaker, H. Hiemstra, F. Rutjes, J. Org. Chem. 2005, 70,
5519.
Scheme 3. Deprotection of trifluoroacetamide.
[12] M. S. Taylor, N. Tokunaga, E. N. Jacobsen, Angew. Chem. 2005,
117, 6858; Angew. Chem. Int. Ed. 2005, 44, 6700.
[13] G. Helmchen, A. Dahnz, P. Dꢁbon, M. Schelwies, R. Weihofen,
Chem. Commun. 2007, 675.
[14] B. Bartels, C. Garcꢀa-Yebra, G. Helmchen, Eur. J. Org. Chem.
2003, 1097.
amount of K2CO3 in MeOH/H2O (7:1) to give the corre-
sponding chiral tetrahydroisoquinoline 13 in excellent yield,
without loss of enantiomeric excess. Through the terminal
olefin and the secondary amine functional groups, 13 is a
highly versatile chiral building block for the synthesis of
tetrahydroisoquinoline-derived structures.
[15] C. A. Kiener, C. T. Shu, C. Incarvito, J. F. Hartwig, J. Am. Chem.
Soc. 2003, 125, 14272.
[16] S. Spiess, C. Welter, G. Franck, J. P. Taquet, G. Helmchen,
Angew. Chem. 2008, 120, 7764; Angew. Chem. Int. Ed. 2008, 47,
7652.
In conclusion, we have developed a new asymmetric
synthesis of chiral nitrogen-containing heterocycles, espe-
cially tetrahydroisoquinolines, which are important building
blocks for the synthesis of biologically active products. Our
approach is based on the first intramolecular asymmetric
iridium-catalyzed allylic amidation, and the desired products
are accessible in excellent yields and enantioselectivities. The
trifluoroacetamide group serves two purposes in this
approach; initially it is used as a protecting group during
the synthesis stage of the starting materials, but later on its
enhanced nucleophilicity is exploited for the key asymmetric
allylic amidation. We have also demonstrated that the
deprotection required to form the corresponding amine can
be readily executed.
´
[17] D. Markovic, J. F. Hartwig, J. Am. Chem. Soc. 2007, 129, 11680.
[18] T. Ohmura, J. F. Hartwig, J. Am. Chem. Soc. 2002, 124, 15164.
[19] J. F. Teichert, B. L. Feringa, Angew. Chem. 2010, 122, 2538;
Angew. Chem. Int. Ed. 2010, 49, 2486.
[20] C. Gnamm, G. Franck, N. Miller, T. Stork, K. Brꢂdner, G.
Helmchen, Synthesis 2008, 3331.
[21] P. Dꢁbon, M. Schelwies, G. Helmchen, Chem. Eur. J. 2008, 14,
6722.
[22] C. Welter, R. M. Moreno, S. Streiff, G. Helmchen, Org. Biomol.
Chem. 2005, 3, 3266.
[23] S. Fꢂrster, G. Helmchen, Synlett 2008, 831.
[24] M. J. Pouy, A. Leitner, D. J. Weix, S. Ueno, J. F. Hartwig, Org.
Lett. 2007, 9, 3949.
690
www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 688 –691

[25] M. J. Pouy, L. M. Stanley, J. F. Hartwig, J. Am. Chem. Soc. 2009,
131, 11312.
[35] G. Lipowsky, N. Miller, G. Helmchen, Angew. Chem. 2004, 116,
4695; Angew. Chem. Int. Ed. 2004, 43, 4595.
[26] J. F. Teichert, B. L. Feringa, Synthesis 2010, 1200.
[27] I. Lyothier, C. Defieber, E. M. Carreira, Angew. Chem. 2006,
118, 6350; Angew. Chem. Int. Ed. 2006, 45, 6204.
[28] C. Defieber, M. A. Ariger, P. Moriel, E. M. Carreira, Angew.
Chem. 2007, 119, 3200; Angew. Chem. Int. Ed. 2007, 46, 3139.
[29] O. V. Singh, H. Han, J. Am. Chem. Soc. 2007, 129, 774.
[30] O. V. Singh, H. S. Han, Org. Lett. 2007, 9, 4801.
[31] In our case this is a Stille reaction with tributylstannylpropenol,
see the Supporting Information for details.
[32] This is a major advantage over N-protecting groups, which have
to be removed by hydrogenolysis.
[33] The 6,7-dimethoxy substitution pattern is part of a large number
of tetrahydroisoquinoline compounds with biological activity.
Hence, we chose this substrate as the starting point of our
studies.
[36] C. T. Shu, A. Leitner, J. F. Hartwig, Angew. Chem. 2004, 116,
4901; Angew. Chem. Int. Ed. 2004, 43, 4797.
[37] K. Tissot-Croset, D. Polet, S. Gille, C. Hawner, A. Alexakis,
Synthesis 2004, 2586.
[38] The reactions were scaled up to 1.0 mmol and gave the same
results in terms of yields and enantioselectivities.
[39] The absolute configuration of the products was determined by
comparison of the optical rotation of compound 2 with the
literature value (Ref. [8]). The absolute configuration of the
other products was assigned by analogy.
[40] In the absence of the Ir catalyst, no reaction was observed, thus
indicating that the Ir catalyst is essential for this transformation
to take place.
[41] For a discussion of possible mechanisms for this transformation,
see the Supporting Information.
[34] With catalytic amounts of DBU the same enantioselectivity was
found, however the reactions did not result in full conversion.
The role of the base has to be elucidated further.
Angew. Chem. Int. Ed. 2011, 50, 688 –691
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
691