Synthesis of α-CN and α-CF3 N-heterocycles through tandem nucleophilic additions

Article abstract of DOI:10.1021/ol2011902

Using a readily available secondary aminoalkyne as starting material, a powerful strategy was discovered to prepare precursors of biologically important unnatural cyclic aminoacids and fluorinated N-heterocycles with important ring sizes (e.g., 5-7) in a

Full text of DOI:10.1021/ol2011902

ORGANIC
LETTERS
2011
Vol. 13, No. 13
3450–3453
Synthesis of r-CN and r-CF3 N-Heterocycles
through Tandem Nucleophilic Additions
Junbin Han, Bo Xu,* and Gerald B. Hammond*
Department of Chemistry, University of Louisville, Louisville, Kentucky 40292, United States
Bo.xu@louisville.edu; gb.hammond@louisville.edu
Received May 4, 2011
ABSTRACT
Using a readily available secondary aminoalkyne as starting material, a powerful strategy was discovered to prepare precursors of biologically
important unnatural cyclic aminoacids and fluorinated N-heterocycles with important ring sizes (e.g., 5À7) in a one-pot reaction using two
nucleophilic additions in a tandem fashion.
N-Heterocycles of different ring sizes and substitution
patterns are among the most important structural classes in
medicinal chemistry. Among them, 5-, 6-, and 7-membered
rings are the most common.1 Recently, we reported the
synthesis of functionalized 5-, 6-, and 7-membered N-hetero-
cycles, via a Cu(I)-catalyzed, one-pot, tandem hydroamina-
tion/alkynylation, a process we called cyclization-triggered
addition.2 We wanted to investigate if this protocol could be
extended to the tandem addition of two different nucleo-
philes to an alkyne, beginning with an intramolecular hydro-
amination of aminoalkyne 1 and ending with the addition
of a second nucleophile to the in situ generated enamine
or imine. We are now pleased to report that our one-pot
cyclization-triggered addition furnishes R-CN and R-CF3
substituted N-heterocycles in very good to excellent yields.
Literature reports on tandem additions of two different
nucleophiles to an alkyne in one pot are indeed rare.3 For
example, Li and co-workers reported the tandem addition
of an amine and alkyne to R,β-unsaturated esters through a
proposed iminium intermediate.4 Che and co-workers have
reported a gold(I) catalyzed tandem synthesis of pyrrolo-
[1,2-a]quinolones.5 A gold catalyzed one-pot synthesis of 1,
2-dihydroquinoline derivativesfrom amines, internal alkynes,
and terminal alkynes has also been reported by Bertrand and
co-workers.6 Finally, Schafer and co-workers have reported a
titanium catalyzed one-pot synthesis of R-aminoacids from
terminal alkynes.7
We used the reaction of aminoalkyne 1a and TMS-CN
(2a) as our initial model because the reported synthesis of
R-cyano-N-heterocycles;oxidative cyanation of a cyclic
tertiary amine8;often suffer from limited scope and lack of
regioselectivity. Using 5% CuBr as catalyst, our tandem
reaction gives 30% of the desired product 3a with excellent
regioselectivity (Table 1, entry 1). When the reaction is carried
out at higher temperature under microwave condition
the yield of 3a is excellent (Table 1, entry 2). Gold also works
well in this reaction (Table 1, entry 3). Under microwave
(1) (a) Zhang, C.; De, C. K.; Mal, R.; Seidel, D. J. Am. Chem. Soc.
2008, 130, 416. (b) Trost, B. M.; Maulide, N.; Livingston, R. C. J. Am.
Chem. Soc. 2008, 16502. (c) Fustero, S.; Moscardo, J.; Jimenez, D.;
Perez-Carrion, M. D.; Sanchez-Rosello, M.; Del Pozo, C. Chem.;Eur.
J. 2008, 14, 9868. (d) Vicario, J. L.; Badia, D.; Carrillo, L. New methods
for the asymmetric synthesis of nitrogen heterocycles 2005; Research
Signpost: Kerala, India, 2005. (e) El Ashry, E. S. H.; El-Nemr, A. Synthesis
of naturally occurring nitrogen heterocycles from carbohydrates; Black-
well Pub.: Oxford, U.K., 2005.
(4) Zhou, L.; Jiang, H.-f.; Li, C.-J. Adv. Synth. Catal. 2008, 350, 2226.
(5) (a) Liu, X.-Y.; Che, C.-M. Angew. Chem., Int. Ed. 2008, 47, 3805.
(b) Liu, X.-Y.; Che, C.-M. Angew. Chem., Int. Ed. 2009, 48, 2367.
(6) Zeng, X.; Frey, G. D.; Kinjo, R.; Donnadieu, B.; Bertrand, G. J.
Am. Chem. Soc. 2009, 131, 8690.
(2) Han, J.; Xu, B.; Hammond, G. B. J. Am. Chem. Soc. 2010, 132,
916.
(7) Lee, A. V.; Schafer, L. L. Synlett 2006, 2973.
(3) (a) Doye, S. Synlett 2004, 1653. (b) Garcia Castro, I.; Tillack, A.;
Hartung, C. G.; Beller, M. Tetrahedron Lett. 2003, 44, 3217. (c) Haak,
E.; Bytschkov, I.; Doye, S. Eur. J. Org. Chem. 2002, 457.
(8) (a) Murahashi, S.-I.; Komiya, N.; Terai, H.; Nakae, T. J. Am.
Chem. Soc. 2003, 125, 15312. (b) Murahashi, S.-I.; Komiya, N.; Terai, H.
Angew. Chem., Int. Ed. 2005, 44, 6931.
r
10.1021/ol2011902
Published on Web 06/02/2011
2011 American Chemical Society

condition, the Cu(I)-catalyzed reaction is very fast (30 min),
but conventional heating also works very well if longer
reaction times are employed (Table 1, entry 4). Increasing
the scale of this reaction did not reduce the yield of product.
5-, and 6-membered rings. Complete regioselectivity was
observed in all cases.
Table 3. Scope of the Tandem Amination/Cyanation Process
Table 1. Screening for the Best Conditions for a Tandem
Amination/Cyanation
entry
catalyst
solvent
temp, time
yield %
1
CuBr
CuBr
dioxane rt, 12 h
30
95
90
92
90
82
2
dioxane mwa, 100 °C, 0.5 h
3
IPrAuCl/AgOTf dioxane mwa, 100 °C, 0.5 h
4
CuBr
CuBr
CuBr
dioxane 100 °C, 12 h
5
6b
toluene mwa, 100 °C, 0.5 h
dioxane mwa, 100 °C, 0.5 h
a mw = microwave. b Using 10 equiv of water; IPrAuCl = Chloro-
[1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene]gold(I).
With optimized condition in hand, we turned our atten-
tion to less basic amines (Table 2).
Table 2. Screening for the Best Conditions for a Tandem
Amination/Cyanation Using Less Basic Aminoalkynes
The regioselectivity obeyed Baldwin’s rules,9 that is,
cyclization of 4-yn-amine 1a and 3-yn-amine 1b gave
five-membered ring products through 5-endodig and
5-exodig processes whereas the reaction of 5-yn-amine
(e.g., 1g) produced a six-membered ring through a 6-exodig
process, and the reaction of 6-yn-amine (e.g., 1j) furnished
a seven-membered ring through a 7-exodig process. When
a chiral aminoalkyne was used (1h), a moderate chiral
induction was observed (dr = 1:6, Table 3, entry 8).
Encouraged by the success of the tandem amination/
cyanation sequence, we turned our attention to the synthesis
The aromatic amine tethered alkyne 1b gave mostly the
hydration product (Table 2, entry 1). We then decided
to use a gold catalyst, IPrAuCl/AgOTf), capable of
strong alkyne activation and with good thermal stabi-
lity; this catalyst gave the desired product in excellent
yield (Table 2, entry 2). We obtained similar results
using the amide tethered alkyne 1b (Table 2, entries 3
and 4).
The scope of the tandem amination/cyanation sequence
is outlined in Table 3. This reaction worked extremely
well in all cases giving near-quantitative chemical yields of
(9) Baldwin, J. E. J. Chem. Soc., Chem. Commun. 1976, 734.
Org. Lett., Vol. 13, No. 13, 2011
3451

of trifluoromethylated N-heterocycles, a medicinally impor-
tant class of compounds10 whose synthesis usually requires
multisteps.10,11 Simply by reacting our in situ generated
enamine intermediate with a nucleophilic CF3 source, our
protocol would lead to a one-pot synthesis of a CF3-contain-
ing N-heterocycle. We used the reaction of aminoalkyne 1f
and TMS-CF3 to optimize reaction conditions. Using 5%
CuBr as catalyst, we observed none of the desired product 4a
(Table 4, entry 1); this result is reasonable if one realizes that
TMS-CF3 is a rather inert reagent unless a suitable activator
is present (i.e., a fluoride source to cleave the TMS group).
Although CsF was not successful (Table 4, entry 2), either
copper or gold catalysts, in combination with AgF, pro-
duced very good yields of 4a (Table 4, entries 3
and 4). CuF2 also furnished the desired product, albeit in
low yield (Table 4, entry 5). Because both gold and copper
worked equally well, we suspected that the real catalyst was
AgF. Indeed, we found this to be the case (Table 4, entry 8).
The optimized loading of AgF was found to be 1.5 equiv
(Table 4, entry 9). Higher temperatures have a deleterious
effect on the yield of 4 (Table 4, entry 11), and the reaction
proceeded well even at room temperature, although longer
times were needed (Table 4, entry 12).
extremely well in all cases giving near-quantitative
chemical yields of five-, and six-membered rings. Com-
plete regioselectivity was observed and its regioselec-
tivity also obeyed Baldwin’s rules. When a chiral
aminoalkyne was used (1k), a moderate chiral induc-
tion was observed (dr = 77:23, Table 5, entry 7). It
should be noted that AgF worked equally well with
basic aliphatic aminoalkynes and less basic aromatic
aminoalkynes.
Table 5. Tandem Amination/Trifluoromethylation Process
Table 4. Screening for the Best Condition for Tandem Amina-
tion/Trifluoromethylation Sequencea
entry
catalyst
temp °C
time
yield %
1
2
3
4
5
6
7
8
CuBr (5%)
100
100
100
100
100
100
100
100
100
100
120
rt
40 min
40 min
40 min
40 min
40 min
40 min
40 min
40 min
40 min
40 min
40 min
6 h
0
0
CuBr (5%)/CsF (1.0 equiv)
IPrAuCl (5%), AgF (1.0 equiv)
CuBr(5%)/AgF(2.0 equiv)
CuF2 (2.0 equiv)
AgF (2.0 equiv)
AgF (0.2 equiv)
AgF (1.0 equiv)
AgF (1.5 equiv)
AgF (4.0 equiv)
AgF (1.5 equiv)
80
82
15
90
40
70
91
82
82
88
9
10
11
12
AgF (1.5 equiv)
To determine the nucleophilic intermediate in the tandem
amination/trifluoromethylation process, we monitored
a mw = microwave; IPrAuCl = Chloro[1,3-bis(2,6-diisopropyl-
phenyl) imidazol-2-ylidene]gold(I).
1
this reaction by 19F NMR and H NMR under similar
conditions to those used in Table 2 (but using CD3CN as
solvent because of its better dissolving properties toward
thesilversalt). Fiveminutes intothereactionweobserveda
The scope of our tandem amination/trifluoromethylation
reaction is outlined in Table 5. The reaction worked
peak ascribed to CF3Ag in 19F NMR (δ À25.5, d, J2
=
FÀAg
94 Hz),12 together with a signal corresponding to CF3H
in 19F NMR (δ À79.9, d, J2FÀH = 79 Hz) and 1H NMR
(δ 6.70, d, J2FÀH = 79 Hz). The formation of CF3H can be
explained by protodemetalation of the CF3Ag intermedi-
ate. Thus, it is highly likely that CF3Ag is the real nucleo-
philic species in the reaction.12
(10) (a) Shevchenko, N. E.; Balenkova, E. S.; Roschenthaler, G.-V.;
Nenajdenko, V. G. Synthesis 2010, 120. (b) Bariau, A.; Jatoi, W. B.;
Calinaud, P.; Troin, Y.; Canet, J.-L. Eur. J. Org. Chem. 2006, 3421. (c)
Chaume, G.; Van Severen, M.-C.; Marinkovic, S.; Brigaud, T. Org. Lett.
2006, 8, 6123.
(11) (a) Nie, J.; Guo, H.-C.; Cahard, D.; Ma, J.-A. Chem. Rev. 2010,
111, 455. (b) Chu, L.; Qing, F.-L. Chem. Commun. 2010, 46, 6285. (c)
ꢀ
Okano, T.; Sakaida, T.; Eguchi, S. J. Org. Chem. 1996,61, 8826. (d) Caupene,
C.; Chaume, G. g.; Ricard, L.; Brigaud, T. Org. Lett. 2008, 11, 209.. (e) Gille,
S.; Ferry, A.; Billard, T.; Langlois, B. R. J. Org. Chem. 2003, 68, 8932.
(12) Tyrra, W. E. J. Fluorine Chem. 2001, 112, 149.
Org. Lett., Vol. 13, No. 13, 2011
3452

We conducted deuterium experiments to explore further
the mechanism of this tandem process (Scheme 1). For
both, tandem amination/cyanation and tandem amina-
tion/trifluoromethylation reactions, deuterium was de-
tected on the ring carbon as well as on the exocyclic
carbon. These results indicate that the proposed enamine
intermediate probably undergoes additional transforma-
tions like retrohydroamination or tautomerization before
the second nucleophilic attack takes place.
Scheme 2. Reaction of 1i in the Absence of a Second Attacking
Nucleophile
Scheme 1. Deuterium Experiments on Amination/Cyanation
and Amination/Trifluoromethylation Reactions
gel flash chromatography of this reaction mixture gave
products 11 and 12.
Because 11 and 12 are not detected in the reaction
mixture prior to silica gel chromatography, their forma-
tion can only be explained by invoking hydration
of the corresponding enamines during chromatography.
This experiment not only confirms the role of enamine as
intermediate, but also underscores the importance of hav-
ing a second attacking nucleophile (i.e., the tandem sequence)
for a clean reaction to take place. Furthermore, the isolation
of 11 implies that the hydroamination is indeed reversible.
Although enamine is generally considered a nucleophile, in
the presence of a suitable catalyst like copper, it can
function as an electrophile.13
Acknowledgment. We are grateful to the National
Science Foundation for financial support (CHE-0809683).
We acknowledge the support provided by the CREAM
Mass Spectrometry Facility (University of Louisville)
funded by NSF/EPSCoR grant # EPS-0447479.
To verify the above enamine formation mechanism, we
investigated the reaction of 1i in the absence of a second
nucleophile (Scheme 2). Without a second nucleophile, the
reaction of 1i produces a relatively complex mixture; its
NMR spectrum signaled the presence of enamine inter-
mediates and other byproduct. We tentatively assigned the
structure of the enamine intermediates as 9 and 10. Silica
Supporting Information Available. Experimental de-
tails and spectra. This material is available free of charge
via the Internet at http://pubs.acs.org.
(13) Koradin, C.; Gommermann, N.; Polborn, K.; Knochel, P.
Chem.;Eur. J. 2003, 9, 2797.
Org. Lett., Vol. 13, No. 13, 2011
3453