Palladium-catalyzed allylic C-H fluorination

Article abstract of DOI:10.1021/ja407223g

The first catalytic allylic C-H fluorination reaction using a nucleophilic fluoride source is reported. Under the influence of a Pd/Cr cocatalyst system, simple olefin substrates undergo fluorination with Et₃N·3HF in good yields with high branched:linear regioselectivity. The mild conditions and broad scope make this reaction a powerful alternative to established methods for the preparation of allylic fluorides from prefunctionalized substrates.

Full text of DOI:10.1021/ja407223g

Communication
pubs.acs.org/JACS
Palladium-Catalyzed Allylic C−H Fluorination
Marie-Gabrielle Braun and Abigail G. Doyle*
Department of Chemistry, Princeton University, Princeton, New Jersey 08544, United States
S
* Supporting Information
ABSTRACT: The first catalytic allylic C−H fluorination
reaction using a nucleophilic fluoride source is reported.
Under the influence of a Pd/Cr cocatalyst system, simple
olefin substrates undergo fluorination with Et3N·3HF in
good yields with high branched:linear regioselectivity. The
mild conditions and broad scope make this reaction a
powerful alternative to established methods for the
Figure 1. Catalytic synthesis of allylic fluorides by functional group
exchange or C−H activation approaches.
preparation of allylic fluorides from prefunctionalized
substrates.
In an effort to identify a synthesis of allylic fluorides that
would obviate the need for substrate prefunctionalization, we
considered whether allylic fluorination could proceed instead
by C−H activation of simple olefin substrates. We recognized
that the electrophilic Pd(II)−sulfoxide catalyst system
developed by White and co-workers for allylic C−H
functionalization might provide such a platform14 while
retaining the high functional group tolerance and selectivity
of the programmed methods. This system has proven
remarkably general for allylic C−H functionalization, enabling
allylic esterification, amination, alkylation, Heck addition, and
dehydrogenation of terminal olefins via a π-allylpalladium
intermediate.15 However, it was apparent at the outset of our
studies that the adaptation to allylic fluorination would present
several challenges. First, although phosphine ligands were
necessary to impart reactivity and selectivity in the conversion
of prefunctionalized substrates to allylic fluorides in the
previously described methodologies using Pd catalysts, these
and other Lewis basic ligands are typically not compatible with
Pd(II)-mediated electrophilic C−H cleavage.16 Furthermore,
many fluoride sources show strong Lewis basicity that might
interfere with C−H activation or lead to elimination rather than
nucleophilic substitution at the π-allylpalladium intermediate. It
has also been shown that under certain conditions, the targeted
allylic fluoride products can themselves undergo reaction with
palladium(0) to form π-allyl complexes, thereby leading to
product decomposition or isomerization.17
ubstituting a C−F bond for a C−H bond in a target
S
molecule has emerged as a powerful strategy for the
optimization of pharmaceuticals, agrochemicals, and perform-
ance materials.1 Accordingly, methods that enable access to
these valuable fluorinated structures by the direct trans-
formation of a C−H bond to a C−F bond are of high
synthetic value. The last 10 years have witnessed notable
advances in the development of such catalytic methods for
2
2
Csp −H fluorination, but examples of catalytic fluorination of
aliphatic C−H bonds have only recently emerged. In 2012,
Sanford and co-workers reported a strategy relying on the use
of a directing group for Pd-catalyzed nucleophilic fluorination
of benzylic C−H bonds.2a,3 Concurrent with this work, both
Groves and Lectka described versatile radical C−H abstraction/
fluorination methods for aliphatic and benzylic fluorination
using Mn,4,5 Cu,6 and Fe7 catalysts. Despite these important
advances, numerous challenges remain, including the identi-
fication of methods that demonstrate unique selectivity,
improved operational convenience, and expanded substrate
scope. Herein we report a strategically distinct approach to
aliphatic C−H fluorination: a Pd(II) catalyst is shown to enable
the branched-selective synthesis of allylic fluorides from simple
olefin substrates using an inexpensive nucleophilic fluoride
source.
The allylic fluoride motif is featured in a range of medicinal
and imaging agents and also serves as a versatile building block
for the construction of numerous aliphatic fluorine-containing
structures.8 Synthetic methods for the preparation of allylic
fluorides currently require substrate prefunctionalization
(Figure 1). For example, our group,9 Gouverneur and
Brown,10 Nguyen,11 Wu,12 and recently Liu13 have demon-
strated that allylic fluorides may be accessed through transition-
metal-catalyzed fluorination of allylic halides, p-nitrobenzoates,
trichloroacetimidates, and phosphorothioates. Although these
methods exhibit high levels of reactivity, functional group
compatibility, and regioselectivity, they suffer from limitations
due to poor atom economy and multistep preparation of
substrates.
With these considerations in mind, our investigation began
with an examination of the allylic C−H fluorination of 1-decene
(1) using the White catalyst, benzoquinone (BQ) as an oxidant,
and a series of fluoride sources (Table 1). In our previously
described Pd-catalyzed enantio- and regioselective allylic
fluorination of allylic halides, AgF was uniquely effective as a
fluoride source;9 however, its use for allylic C−H fluorination
of 1 resulted in no detectable product formation (entry 1).
Reactions with alkali-metal fluorides such as potassium fluoride
Received: July 15, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja407223g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
Table 1. Fluoride Source and Cocatalyst Optimization
Table 2. Catalyst and Ligand Optimization
a
b
c
entry
catalyst
ligand
L1
conv (%)
yield (%)
b:l
1
2
3
4
5
6
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(TFA)2
Pd(TFA)2
Pd(TFA)2
−
82
88
23
38
96
94
33
7
51
65
9
7.0:1
6.0:1
6.6:1
L2
L3
a
b
c
“F−” source
M
conv (%)
yield (%)
b:1
d
entry
1
d
e
d
DMSO
L1
1
nd
AgF
−
−
−
−
−
100
12
40
81
38
82
37
35
33
47
0
0
nd
e
d
58
70
0
7.4:1
2
KF
nd
d
L2
7.3:1
e
3
KHF2
0
nd
f
d
7
L2
nd
4
Pyr·9HF
Et3N·3HF
Et3N·3HF
Et3N·3HF
Et3N·3HF
Et3N·3HF
Et3N·3HF
0
nd
e
8
L2
0
nd
5
33
51
6
6.6:1
7.0:1
7.9:1
7.2:1
8.2:1
5.3:1
a
6
7
CrCl
CoCl
MnCl
AlCl
CrF
Determined by GC using dodecane as an internal standard for
reactions carried out on a 0.2 mmol scale in a polypropylene vial.
Combined GC yields of 2b and 2l. Based on GC analysis of the
b
c
8
14
8
unpurified reaction mixture and not corrected for small response
9
d
e
f
variations. 1.4 equiv. Not determined. Reaction carried out without
BQ.
10
28
a
Determined by GC using dodecane as an internal standard for
reactions carried out on a 0.2 mmol scale in a polypropylene vial.
b
c
Combined GC yields of 2b and 2l. Based on GC analysis of the
unpurified reaction mixture and not corrected for small response
induced allylic C−H fluorination of 1 in improved yield (entry
2). Additionally, a brief survey of Pd(II) catalysts revealed that
Pd(TFA)2 was superior to Pd(OAc)2, affording 2 in 70% yield
with 7.3:1 b:1 selectivity (entry 6). On the other hand,
changing the solvent and reaction temperature had no
beneficial impact on the reaction yield. Notably, control
experiments determined that allylic fluoride 2 is not produced
in the absence of the Pd(II) catalyst or BQ (entries 7 and 8).22
Having established conditions suitable for the allylic
fluorination of 1-decene, we proceeded to examine the
substrate scope of the methodology (Table 3). Olefins bearing
a broad range of oxygen- and nitrogen-containing functional
groups were found to be competent substrates, delivering
products with benzyl ethers (3), esters (4), amides (5), and
phthalimides (6) (entries 2−5). Tolerance of the method to N-
containing heterocycles such as pyridines (7) and acidic
heteroatom−H bonds (8 and 9) is particularly significant
given the importance of these motifs in medicinal chemistry
(entries 6−8). An alkyl bromide-containing olefin was also well-
tolerated, underscoring the mildness of the nucleophilic
fluorination conditions (entry 9). In addition, fluorination of
arene- and alkyne-containing olefins afforded allylic fluorides 11
and 12 as the exclusive fluorine-containing products in 64 and
47% yield, respectively (entries 10 and 11); these findings are
noteworthy because chemoselectivity for allylic over benzylic or
propargylic fluorination is not observed in the other reported
methods for aliphatic C−H fluorination. An additional feature
that distinguishes this C−H fluorination is that the reactions
can be conducted under air using wet solvents.
d
e
variations. Not determined. Reaction carried out at 1 M.
(KF) were similarly unsuccessful, instead affording byproducts
from elimination or acetoxylation of the allylpalladium
intermediate (entry 2). Since nucleophiles bearing acidic X−
H bonds have been most successful for the Pd(II)−sulfoxide
catalyst system, we next turned our attention to the evaluation
of HF-containing fluoride reagents. While KHF2 and Olah’s
reagent18 were ineffective (entries 3 and 4), Et3N·3HF
provided allylic fluoride 2 in 33% yield with high branched:-
linear (b:l) selectivity (6.6:1) and minimal formation of these
byproducts (entry 5). This fluoride source has the benefit of
being inexpensive and commercially available, but to the best of
our knowledge, it has never been employed for Pd-catalyzed
fluorinations.
In an effort to improve the yield of this promising result, we
turned our attention to the inclusion of Lewis acid cocatalysts, a
strategy that was previously delineated by White and co-
workers for the purpose of activating the putative π-
allylpalladium−BQ intermediate toward nucleophilic function-
alization.19 A variety of metal−salen complexes were evaluated
(Table 1, entries 6−9). Most notably, addition of catalytic
amounts of (salen)CrCl (10 mol %) gave the desired
fluorinated product 2 in an enhanced 51% yield while
maintaining high levels of regioselectivity (entry 6). Our
observation that (salen)CrF was less reactive than (salen)CrCl
(entry 10) suggests that this cocatalyst does not play the role of
a fluoride delivery agent.20
Overall, the scope and accompanying regioselectivity of the
method compare favorably to those described for alternative
syntheses of allylic fluorides. The yields are modest but
synthetically useful, especially given the accessibility of the
starting materials. For example, whereas the terminal olefin
precursor to 10 in Table 3 is commercially available,
Further optimization studies evaluated the influence of the
sulfoxide ligand and the identity of the Pd precatalyst on the
reaction outcome using cocatalytic (salen)CrCl (Table 2).
Whereas monosulfoxide ligands provided poorer conversions
(entries 3 and 4), the bis(benzyl sulfoxide) ligand L214,21
B
dx.doi.org/10.1021/ja407223g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
preparation of the allylic chloride substrate previously necessary
to access this allylic fluoride requires seven steps.9 Moreover,
the C−H fluorinations proceed to complete conversion with
minimal diene generation, enabling straightforward product
isolation.23 Nevertheless, certain limitations were identified:
cyclic alkenes and olefins with substitution at the homoallylic
position, such as allylcyclohexane, were found to undergo
fluorination with significantly depressed yields (33% and 16%;
entries 12 and 13). Furthermore, allylbenzene proved to be a
poor substrate, providing exclusively the linear product 15 in
18% yield (entry 14);24 the reported instability of 15 at high
concentrations may explain this result.25
Table 3. Scope of the Allylic C−H Fluorination
C−H activation approaches can be particularly useful when
applied to the direct functionalization of bioactive natural
products or drug candidates. In these targets, allylic C−H
bonds can be metabolic hotspots;26 since fluorination is a
common strategy used in medicinal and agrochemistry to block
such sites, a direct allylic fluorination may prove quite useful.
To demonstrate this potential, we subjected the complex
steroid scaffold 16 to the Pd-catalyzed allylic C−H fluorination
conditions (Figure 2). In the event, allylic fluoride 17 was
isolated from this reaction in 59% yield with good
regioselectivity (b:l = 8.0:1).27
Figure 2. Late-stage functionalization of a natural product derivative
via direct allylic C−H fluorination. Conditions: (a) Pd(TFA)2 (15 mol
%), L2 (15 mol %), [(R,R)-salen]CrCl (10 mol %), Et3N·3HF (6.0
equiv), BQ (2.0 equiv), DCE (2 M), 23 °C, 72 h.
In conclusion, we have demonstrated the first example of a
direct allylic C−H fluorination reaction using a simple
nucleophilic fluoride reagent. The methodology furnishes a
diverse collection of synthetically valuable fluorinated products
under mild and operationally simple conditions. Furthermore,
in comparison with reported methods for C−H fluorination,
this approach exhibits chemoselective allylic functionalization.
Our future investigations will focus on elucidating the
mechanism of this reaction and rendering it enantioselective.28
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures, additional reaction optimization, and
spectroscopic data for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
agdoyle@princeton.edu
■
Notes
a
Combined isolated yields of branched and linear isomers for reactions
The authors declare no competing financial interest.
carried out on a 0.4 mmol scale in a polypropylene vial (averages of
b
two runs). Based on 19F NMR analyses of the purified products
ACKNOWLEDGMENTS
We thank Elias Wang for help with substrate synthesis and
Matthew K. Nielsen for experimental assistance. Financial
c
d
■
(averages of two runs). Reaction conducted in dioxane. Determined
by 19F NMR analysis using fluorobenzene as a quantitative internal
standard.
C
dx.doi.org/10.1021/ja407223g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(18) (a) Olah, G. A.; Nojima, M.; Kerekes, I. Synthesis 1973, 779−
780. (b) Olah, G. A.; Nojima, M.; Kerekes, I. Synthesis 1973, 780−783.
(c) Olah, G. A.; Welsh, J. T.; Vankar, Y. D.; Nojima, M.; Kerekes, I.;
Olah, J. A. J. Org. Chem. 1979, 44, 3872−3881.
(19) (a) Covell, D. J.; White, M. C. Angew. Chem., Int. Ed. 2008, 47,
6448−6451. (b) Reed, S. A.; White, M. C. J. Am. Chem. Soc. 2008, 130,
3316−3318. (c) Gormisky, P. E.; White, M. C. J. Am. Chem. Soc. 2011,
133, 12584−12589. (d) Qi, X.; Rice, G. T.; Lall, M. S.; Plummer, M.
S.; White, M. C. Tetrahedron 2010, 66, 4816−4826.
(20) (salen)M catalysts promote fluoride ring opening of epoxides,
and in one mechanistic study, a (salen)Co fluoride was shown to be
the active nucleophile. See: (a) Bruns, S.; Haufe, G. J. Fluorine Chem.
2000, 104, 247−254. (b) Haufe, G.; Bruns, S.; Runge, M. J. Fluorine
Chem. 2001, 112, 55−61. (c) Haufe, G.; Bruns, S. Adv. Synth. Catal.
2002, 344, 165−171. (d) Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc.
2010, 132, 3268−3269. (e) Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc.
2011, 133, 16001−16012.
support provided by the NSF (CAREER 1148750), Princeton
University, Eli Lilly, and Amgen and a fellowship to M.-G.B.
from the Ecole Polytechnique are gratefully acknowledged.
A.G.D. is an Alfred P. Sloan Foundation Fellow and a Camille
Dreyfus Teacher-Scholar.
REFERENCES
■
(1) (a) Jeschke, P. ChemBioChem 2004, 5, 570−589. (b) Muller, K.;
̈
Faeh, C.; Diederich, F. Science 2007, 317, 1881−1886. (c) Purser, S.;
Moore, P. R.; Swallow, S.; Gouverneur, V. Chem. Soc. Rev. 2008, 37,
320−330. (d) Cai, L.; Lu, S.; Pike, V. W. Eur. J. Org. Chem. 2008,
2853−2873.
(2) (a) Hull, K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc.
2006, 128, 7134−7135. (b) Wang, X.; Mei, T.-S.; Yu, J.-Q. J. Am.
Chem. Soc. 2009, 131, 7520−7521. (c) Chan, K. S. L; Wasa, M.; Wang,
X.; Yu, J.-Q. Angew. Chem., Int. Ed. 2011, 50, 9081−9084. (d) Lou, S.-
J.; Xu, D.-Q.; Xia, A.-B.; Wang, Y.-F.; Liu, Y.-K.; Du, X.-H.; Xu, Z.-Y.
Chem. Commun. 2013, 49, 6218−6220. (e) Truong, T.; Klimovica, K.;
Daugulis, O. J. Am. Chem. Soc. 2013, 135, 9342−9345.
(3) McMurtrey, K. B.; Racowski, J. M.; Sanford, M. S. Org. Lett. 2012,
14, 4094−4097.
(21) (a) Young, A. J.; White, M. C. Angew. Chem., Int. Ed. 2011, 50,
6824−6827. (b) Stang, E. M.; White, M. C. J. Am. Chem. Soc. 2011,
133, 14892−14895.
(22) Efforts to reduce the catalyst and cocatalyst loadings led to
significant reductions in reaction efficiency. See the Supporting
Information for additional optimization studies.
(4) Liu, W.; Huang, X.; Cheng, M.-J.; Nielsen, R. J.; Goddard, W. A.,
III; Groves, J. T. Science 2012, 337, 1322−1325.
(23) We have not yet been able to identify the mass balance of the
reactions, but it may be polymeric material.
(5) Liu, W.; Groves, J. T. Angew. Chem., Int. Ed. 2013, 52, 6024−
(24) Allylic fluorination of 1,1- and 1,2-disubstituted olefins also
resulted in low conversion and yield. See the Supporting Information
for more information.
(25) Lee, E.; Yandulov, D. V. J. Fluorine Chem. 2009, 130, 474−483.
(26) Ando, T.; Koseki, N.; Yasuhara, I. K.; Kasuga, N. M.; Ishiwatari,
T. Pestic. Sci. 1994, 40, 307−312.
(27) (a) Unfortunately, the diastereomeric ratio of 17b could not be
determined by standard spectroscopic techniques. (b) It is also
noteworthy that allylic fluorination of the free diol of 16 delivered a
yield and regioisomeric ratio (43% yield, b:l = 6.8:1) similar to those
with diacetate 16.
6027.
(6) Bloom, S.; Pitts, C. R.; Miller, D. C.; Haselton, N.; Holl, M. G.;
Urheim, E.; Lectka, T. Angew. Chem., Int. Ed. 2012, 51, 10580−10583.
Two examples of allylic C−H fluorination of α-methylstyrenes are
included in this report; in both cases, a single linear allylic fluoride is
the only possible product.
(7) Bloom, S.; Pitts, C. R.; Woltornist, R.; Griswold, A.; Holl, M. G.;
Lectka, T. Org. Lett. 2013, 15, 1722−1724.
(8) For reviews of the synthesis of allylic fluorides, see: (a) Pacheco,
M. C.; Purser, S.; Gouverneur, V. Chem. Rev. 2008, 108, 1943−1981.
(b) Hollingworth, C.; Gouverneur, V. Chem. Commun. 2012, 48,
2929−2942.
(9) (a) Katcher, M. H.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132,
17402−17404. (b) Katcher, M. H.; Sha, A.; Doyle, A. G. J. Am. Chem.
Soc. 2011, 133, 15902−15905.
(28) Allylic fluoride 3 was obtained as a racemic mixture under the
optimized conditions.
(10) (a) Hollingworth, C.; Hazari, A.; Hopkinson, M. N.; Tredwell,
M.; Benedetto, E.; Huiban, M.; Gee, A. D.; Brown, J. M.; Gouverneur,
V. Angew. Chem., Int. Ed. 2011, 50, 2613−2617. (b) Benedetto, E.;
Tredwell, M.; Hollingworth, C.; Khotavivattana, T.; Brown, J. M.;
Gouverneur, V. Chem. Sci. 2013, 4, 89−96.
(11) Topczewski, J. J.; Tewson, T. J.; Nguyen, H. M. J. Am. Chem.
Soc. 2011, 133, 19318−19321.
(12) Lauer, A. M.; Wu, J. Org. Lett. 2012, 14, 5138−5141.
(13) Zhang, Z.; Wang, F.; Mu, X.; Chen, P.; Liu, G. Angew. Chem.,
Int. Ed. 2013, 52, 7549−7553.
(14) Chen, M. S.; White, M. C. J. Am. Chem. Soc. 2004, 126, 1346−
1347.
(15) (a) White, M. C. Synlett 2012, 23, 2746−2748 and references
therein. Also see: (b) Bigi, M. A.; White, M. C. J. Am. Chem. Soc.
2013, 135, 7831−7834. (c) Delcamp, J. H.; Gormisky, P. E.; White, M.
C. J. Am. Chem. Soc. 2013, 135, 8460−8463. (d) Liu, G.; Yin, G.; Wu,
L. Angew. Chem., Int. Ed. 2008, 47, 4733−4736. (e) Nahra, F.; Liron,
F.; Prestat, G.; Mealli, C.; Messaoudi, A.; Poli, G. Chem.Eur. J. 2009,
15, 11078−11082. (f) Lin, S.; Song, C.-X.; Cai, G.-X.; Wang, W.-H.;
Shi, Z.-J. J. Am. Chem. Soc. 2008, 130, 12901−12903.
(16) For two notable exceptions, see: (a) Trost, B. M.; Thaisrivongs,
D. A.; Hansmann, M. M. Angew. Chem., Int. Ed. 2012, 51, 11522−
11526. (b) Trost, B. M.; Hansmann, M. M.; Thaisrivongs, D. A.
Angew. Chem., Int. Ed. 2012, 51, 4950−4953.
(17) (a) Hintermann, L.; Lang, F.; Maire, P.; Togni, A. Eur. J. Inorg.
Chem. 2006, 1397−1412. (b) Hazari, A.; Gouverneur, V.; Brown, J. M.
Angew. Chem., Int. Ed. 2009, 48, 1296−1299. (c) Oh, Y. H.; Ahn, D. S.;
Chung, S. Y.; Jeon, J. H.; Park, S. W.; Oh, S. J.; Kim, D. W.; Kil, H. S.;
Chi, D. Y.; Lee, S. J. Phys. Chem. A 2007, 111, 10152−10161.
D
dx.doi.org/10.1021/ja407223g | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX