The development of a Selectfluor-mediated oxidative Mannich reaction

Article abstract of DOI:10.1016/j.tetlet.2011.05.012

Selectfluor is an easy to handle, air-stable and powerful electrophilic fluorination agent that can be used in a novel oxidative Mannich-type carbon-carbon bond forming process to allylate electron deficient tertiary amines in moderate to good yields. This process is rapid, operationally simple, insensitive to atmospheric conditions and of potentially broad use, especially in the construction of homoallylamines.

Full text of DOI:10.1016/j.tetlet.2011.05.012

Tetrahedron Letters 52 (2011) 3543–3546
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
The development of a Selectfluor®-mediated oxidative Mannich reaction
Matthew H. Daniels a, , Jed Hubbs a,
⇑
a Department of Medicinal Chemistry, Merck Research Laboratories, 33 Avenue Louis Pasteur, Boston, MA 02115, United States
a r t i c l e i n f o
a b s t r a c t
SelectfluorÒ is an easy to handle, air-stable and powerful electrophilic fluorination agent that can be used
in a novel oxidative Mannich-type carbon–carbon bond forming process to allylate electron deficient ter-
tiary amines in moderate to good yields. This process is rapid, operationally simple, insensitive to atmo-
spheric conditions and of potentially broad use, especially in the construction of homoallylamines.
Ó 2011 Elsevier Ltd. All rights reserved.
Article history:
Received 7 April 2011
Revised 2 May 2011
Accepted 3 May 2011
Available online 11 May 2011
Keywords:
SelectfluorÒ
Mannich reaction
Allylation
Oxidation
Nucleophilic addition
1. Introduction
Cl
2 BF4
F
N+
SelectfluorÒ (1) is a commercially available electrophilic fluori-
nating reagent that has seen broad application in recent years.1 It is
stable and easy to handle yet is an extremely powerful donor of an
electron deficient fluorine atom to a wide number of nucleophiles.
Studies have shown that 1 is the strongest of several commercially
available, commonly used fluorinating reagents, including N-fluo-
ropyridinium salts and N-fluorosulfonimides.2 Reagent 1 has been
-
F
N+
1
OH
Ph
F
N
Bn
Ph
N
Bn
N
H
Ph
Selectfluor®
1
2
3
4
Scheme 1. Reagents and conditions: 4 equiv 1, 0.2 M in MeCN, rt, 2.5 h.
utilized in the synthesis of
a
-fluorothioethers,3 2-deoxy-2-fluoro
sugars4 and enediones,5 as well as for bromination of olefins6
and oxidation of benzylic alcohols and aldehydes.7 Compound 1
also has utility in the deprotection of p-methoxybenzylidine pro-
tected acetals, 1,3-dithiane protected ketones, and 2-terahydopyr-
anyl ethers.8 Additionally, recent research has shown 1 to be useful
in gold-catalyzed oxidative functionalization processes.9
1
-HF
F
N
Bn
Ph
N+ Ph
Bn
N+ Ph
Bn
N+ Ph
Ph
2
A
B
C
During the course of a medicinal chemistry program, it was dis-
covered that compounds such as 2 could be converted to dif-
F
2 equiv 1, -2 H+
then H2O, -H+
H2O
F
luorohydroxypiperidines such as
3
upon treatment with
1
B
C
OH
Ph
(Scheme 1).10 The mechanism of this transformation (Scheme 2)
was postulated to proceed via the oxidation of 2 by 1 to a cyclic
iminium ion B via intermediate A. Fluorination of B with excess
equivalents of 1 followed by hydrolysis provided 3. Preliminary
investigations of the scope of the difluorohydroxylation reaction,
however, provided mixtures of product 3 and dealkylated amine
4. It was suspected that the comparable C–H acidities of the two
N
Bn
- PhCHO
N
H
Ph
3
4
Scheme 2. Proposed mechanism of formation of 3 and 4.
benzylic methines led to poor selectivity for the formation of B
as opposed to C.
We believed that this reaction could be exploited to allow
nucleophiles other than water to react with the in situ generated
iminium ion, enabling tertiary amines to act as electrophiles in
⇑
Corresponding author.
E-mail address: matthew_daniels2@merck.com (M.H. Daniels).
Present address: Satori Pharmaceuticals, 75 Rogers St., Cambridge, MA 02142,
United States.
0040-4039/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2011.05.012

3544
M. H. Daniels, J. Hubbs / Tetrahedron Letters 52 (2011) 3543–3546
Table 1
Reactions of nucleophiles with 5 mediated by 1a
F
1
-HF
Ph
N
Ph
N+ Ph
EtO
MeO
TMS
H
H
1, MeCN
O
O
N
Ph
N
Ph
Nu, RT
5
D
EtO2C
EtO2C
R
5
6-11
N+ Ph
N
Entry
Nu
R
Product
Yieldb (%)
EtO
EtO
TMS
BF3K
CH2
CH2
O
O
1
2
6
6
41
58
H2C
E
6
H2C
O
Scheme 3. Proposed mechanism for conversion of 5 to 6.
O
O
O
3
4
7
8
18
10
Mannich-type reactions. Such oxidative Mannich-type reactions
have been previously reported. For example, DDQ has been used
to oxidize 2-amino silyl ketenes in situ, which can then be treated
with organometallic nucleophiles to give Mannich-type prod-
ucts.11 Other conditions developed include ruthenium12 or rho-
dium13 based catalysts to oxidize unactivated alkyl amines, as
well as the use of N-tert-butylbenzenesulfinimidoyl chloride as
an organic oxidant.14 All of these techniques, while useful, suffer
from drawbacks such as low temperatures, long reaction times,
complicated procedures or expensive transition metal catalysts.
In contrast, 1 is relatively inexpensive, operationally simple to
use and a competent oxidant at room temperature.1
O
OTMS
O
OTMS
5
9
6
OTMS
O
6
7
10
11
16
45
Me
N
Me
N
To explore the possibility of a Mannich-type reaction using 1,
initial nucleophile screens were conducted using the piperidinyl
acetate 5. This substrate was chosen due to the enhanced C–H acid-
ity of the exocyclic methylene carbon, reducing formation of
hydroxydifluoropiperidines. Scheme 3 shows the proposed mecha-
nism for transformation of 5 into compound 6. Treatment of 5 with
1 likely provides fluorinated intermediate D. Elimination of HF
then delivers iminium ion E, which can then be quenched nucleo-
philes such as allyltrimethylsilane to give compounds such as 6.
a
Compound 5 (1 mmol), 1 (1.5 mmol), and Nu (2 mmol) were stirred in 5 ml
MeCN for 90 min at room temperature.
b
Isolated yield.
Table 2
Reactions of allyl nucleophiles with 8 in the presence of 1 and basea
Ph
N
Ph
Table
1 shows the results of reacting various carbon
nucleophiles with ester 5 to provide compounds 6–11. Reactions
generally proceeded rapidly at room temperature, providing poor
to moderate yields of the desired products. Diverse carbon
nucleophiles including allylmetal compounds (entries 1 and 2),
aromatic compounds (entries 3 and 7) and silyl enol ethers (entries
4–6) all provided some yield of the desired products. Potassium
allyltrifluoroborate gave the highest yields under these conditions.
For compounds 8, 9 and 10, relative stereochemistry and diastere-
oselectivity were not determined.
In considering parameters for optimization of an oxidative
Mannich reaction mediated by 1, we reasoned that addition of a
base would promote the initial oxidation step. Unfortunately,
treating compound 12 with allyltrimethylsilane in the presence
of 1 and 1,1-diazabicyclo[2.2.2]octane (DABCO), a strong amine
base closely related to SelectfluorÒ, gave no isolable product 13
(Table 2, entry 1). Although neither diisopropylethylamine nor
potassium carbonate provided 13 (entries 2 and 3), switching the
nucleophile to potassium allyltrifluoroborate and the base to either
Proton Sponge™ (PS) or 4-dimethylaminepyridine (DMAP) did in
fact cleanly deliver homoallylamine 13 in up to 72% isolated yield.
Notably, complete conversion was observed in only 10 min at
room temperature. Further optimization (not shown) suggested
that 1.5 equiv of 1 and 1.05 equiv of PS were ideal.
1, Nu
N
MeCN, Base, RT
EtO2C
EtO2C
12
13
Entry
Nu
Base
Equiv base
Yieldb (%)
TMS
TMS
TMS
1c
2
DABCO
DIPEA
K2CO3
1.1
2
0
0
0
H2C
H2C
H2C
H2C
3
2
BF3K
NMe2NMe2
PS
4
5
1.1
1.1
72
52
BF3K
DMAP
H2C
a
Compound 7 (1 mmol), 1 (1.5 mmol), nucleophile (2 mmol) and base were
stirred in 5 ml MeCN at rt for 10–20 min.
b
Isolated yield.
Reaction run for 180 min.
c
This system still proved to be sensitive to the identity of the
nucleophile, as shown in Table 3. While N-methylpyrrole and silyl
enol ethers give moderate yields of product (entries 4–7), most
other nucleophiles, including allyltrimethylsilane, gave rather poor
yields of the desired products.
Revisiting the original substrate 5 also provided drastically
diminished results, furnishing compound 6 in only 10% isolated
yield. Protected glycine derivative 21, however, was treated with
1 and PS to give 22 in 52% isolated yield (Scheme 4). Given the po-
tential interest in assembling unnatural amino acids from a readily

M. H. Daniels, J. Hubbs / Tetrahedron Letters 52 (2011) 3543–3546
3545
Table 3
Table 4
Reactions of nucleophiles with 12 mediated by 1a
Scope of substrate in 1-mediated allylationa
Ph
Ph
N
EWG
EWG
1, Base
R
R
MeCN, RT
N
R'
N
R'
CH2
1, PS, Nu
BF3K
N
H2C
23-27
28-32
MeCN, RT
EtO2C
EtO2C
13-20
R
Entry
1
SM
Base
P
Yieldb (%)
12
Bn2N
F3C
CN
23
24
25
26
TMSNHAc
None
28
29
30
31
58
58
56
41
Entry
1
Nu
R
Product
Yieldb (%)
13
2
3
4
N
Ph
TMS
CH2
13
H2C
Bn2N
COMe
N
DMAP
OTMS
O
O
MeOC
2
3
14
15
25
28
DMAP
Ph
OTBDMS
5
N
27
None
32
41
Ph
O
O
F3C
a
Substrate (1 mmol), 1 (1.5 mmol), base (if used, 1.1–1.5 mmol) and potassium
Me
N
Me
N
allyltrifluoroborate (2 mmol) were stirred in 5 mL MeCN for 10–20 min at room
temperature.
4
5
16
17
40
54
b
Isolated yield.
OMe
OMe
O
OTBDMS
With this knowledge in hand, we began to investigate alterna-
tive substrates, varying both the participating and non-participat-
ing substituents on the nitrogen ring (Table 4). Ultimately, we
found that while a variety of electron deficient tertiary alkyl
amines could react with nucleophiles in the presence of 1, optimi-
zation was still needed for a given substrate. For example, it was
necessary to use N-trimethylsilylacetamide to obtain an acceptable
yield (58%) of 28 from 23 (entry 1). The use of DMAP in this trans-
formation provided 28 in only 18% yield. Some substrates gave
highest yields in the absence of base (entries 2 and 5). While the
substrate scope was generally good, one notable exception was
the failure of any N-aryl amine to provide any product other than
fluorination of the phenyl ring.
OMe
Me
Me
OMe
OMe
6
18
55
OTMS
O
Me
Me Me
OMe
7
8
19
20
46
22
Me
O
S
OTMS
S
TMS
N
N
a
12 (1 mmol), 1 (1.5 mmol), PS (1.05 mmol) and Nu (2 mmol) were stirred in
5 mL MeCN for 10 min at room temperature.
b
Isolated yield.
In conclusion, we have developed a novel method for an oxida-
tive Mannich-type reaction mediated by SelectfluorÒ (1). This reac-
tion works on a variety of electron-poor tertiary alkyl amine
substrates with a diverse array of carbon nucleophiles in poor to
good yields. While substrate-specific optimization is required, this
protocol does offer a useful route to complex tertiary amines.
1, Base
CH2
Bn2N
Bn2N
MeCN, RT
CO2Me
21
CO2Me
22
BF3K
H2C
Acknowledgments
Base: PS, 52% yield
DMAP, 80% yield
Portions of this work were conducted as part of the Cooperative
Education Program with Northeastern University, Boston, MA
(M.H.D.). We thank Merck colleague Bruce Adams for assistance
with NMR analyses.
Scheme 4. Reagents and conditions: 1.5 equiv 1, 1.1 equiv PS or DMAP, 2.0 equiv
allyl-BF3K, 0.2 M in MeCN, 20 min, rt.
available glycine derivative, another round of optimization was ini-
tiated using ester 21 as the substrate. It was discovered that the
use of 1 equiv of DMAP instead of PS could provide compound
22 in 80% yield.15 As a side investigation, this substrate was used
to test the sensitivity of the reaction conditions. While the use of
a 4:1 mixture of acetonitrile and water did lead to dealkylation
of 21 (cf. Scheme 2), the 1-mediated conversion of 21 to 22 in
the presence of DMAP proceeded smoothly under air, nitrogen
and argon (80%, 77% and 84% yield, respectively). The use of an ar-
gon atmosphere with 4 Å molecular sieves to absorb ambient
water also gave a yield 80% yield. These results suggest that the
use of 1 as an oxidant in Mannich-type reactions is extremely ro-
bust, insensitive to ambient conditions and operationally simple.
References and notes
1. Nyffler, P. T.; Durón, S. G.; Burkart, M. D.; Vincent, S. P.; Wong, C.-H. Angew.
Chem., Int. Ed. 2005, 44, 192–212.
2. Toullec, P. Y.; Devillers, I.; Frantz, R.; Togni, A. Helv. Chim. Acta 2004, 87, 2706–
2711.
3. Lal, G. S. J. Org. Chem. 1993, 58, 2791.
4. Vincent, S. P.; Burkart, M. D.; Tsai, C.-Y.; Zhang, Z.; Wong, C.-H. J. Org. Chem.
1999, 64, 5264–5279.
5. Blank, S. J.; Stephens, C. E. Tetrahedron Lett. 2006, 47, 6849–6850.
6. Ye, C.; Shreeve, J. M. J. Org. Chem. 2004, 69, 8561–8563.
7. Banks, R. E.; Lawrence, N. J.; Popplewell, A. L. Synlett 1994, 831.
8. Liu, J.; Wong, C.-H. Tetrahedron Lett. 2002, 43, 4037–4039.
9. (a) Melhado, A. D.; Brenzovich, W. E., Jr.; Lackner, A. D.; Toste, F. D. J. Am. Chem.
Soc. 2010, 132, 8885; (b) Zhang, G.; Cui, L.; Wang, Y.; Zhang, L. J. Am. Chem. Soc.
2010, 132, 1474; (c) Hopkinson, M. N.; Ross, J. E.; Giuffredi, G. T.; Gee, A. D.;

3546
M. H. Daniels, J. Hubbs / Tetrahedron Letters 52 (2011) 3543–3546
Gouverneur, V. Org. Lett. 2010, 12, 4904–4907; (d) Zhang, G.; Peng, Y.; Cui, L.;
Zhang, L. Angew. Chem., Int. Ed. 2009, 48, 3112–3115.
dimethylaminopyridine (134 mg, 1.1 mmol), potassium allyltrifluoroborate
(296 mg, 2.0 mmol) and (531 mg, 2.0 mmol). slight exotherm was
1
A
10. Stanton, M. G.; Hubbs, J.; Sloman, D.; Hamblett, C.; Andrade, P.; Angagaw, M.;
Bi, G.; Black, R. M.; Crispino, J.; Cruz, J. C.; Fan, E.; Farris, G.; Hughes, B. L.;
Kenific, C. M.; Middleton, R. E.; Nikov, G.; Sajonz, P.; Shah, S.; Shomer, N.;
Szewczak, A. A.; Tanga, F.; Tudge, M. T.; Shearman, M.; Munoz, B. Bioorg. Med.
Chem. Lett. 2010, 20, 755–758.
11. Shimizu, M.; Itou, H.; Miura, M. J. Am. Chem. Soc. 2005, 127, 3296–3297.
12. Murahashi, S.-I.; Komiya, N.; Terai, H.; Nakae, T. J. Am. Chem. Soc. 2003, 125,
15312–15313.
13. Catino, A. J.; Nichols, J. M.; Nettles, B. J.; Doyle, M. P. J. Am. Chem. Soc. 2006, 128,
5648–5649.
14. Matsu, J.; Tanaki, Y.; Ishibashi, H. Org. Lett. 2006, 8, 4371–4374.
15. The following procedure is representative. To a stirring solution of 21 (270 mg,
1.0 mmol) in acetonitrile (5.0 mL) was added successively 4-N,N-
observed. The reaction mixture was allowed to stir for 20 min at room
temperature, at which point the solvent was removed in vacuo. The resulting
residue was taken up in a minimal amount of DCM, poured into hexanes and
filtered over silica gel, eluting with Et2O. The filtrate was concentrated under
vacuum and purified by flash chromatography (10% EtOAc/hexanes) to obtain
22 (247 mg, 80%) as a white solid. 1H NMR (600 MHz, CDCl3): d 7.37 (dt,
J = 16.1, 7.6 Hz, 4H), 7.30 (t, J = 7.5 Hz, 4H), 7.23 (dd, J = 14.2, 7.0 Hz, 2H), 5.72
(m, 1H), 5.04 (m, 2H), 3.92 (d, J = 13.9 Hz, 2H), 3.75 (s, 3H), 3.53 (d, J = 14.0 Hz,
2H), 3.44 (m, 1H), 2.52 (m, 2H). 13C NMR (150 MHz, CDCl3): d 173.1, 139.8,
135.3, 129.2, 128.5, 127.3, 117.2, 61.0, 54.8, 51.3, 34.4. HPLC–MS (APCI):
[MH]+ = 310.1, expected 310.2.