Palladium-catalyzed asymmetric synthesis of allylic fluorides

Article abstract of DOI:10.1021/ja109120n

The enantioselective fluorination of readily available cyclic allylic chlorides with AgF has been accomplished using a Pd(0) catalyst and Trost bisphosphine ligand. The reactions proceed with unprecedented ease of operation for Pd-mediated nucleophilic fluorination, allowing access to highly enantioenriched cyclic allylic fluorides that bear diverse functional groups. Evidence that supports a mechanism in which C-F bond formation occurs by an S_N2-type attack of fluoride on a Pd(II)-allyl intermediate is presented.

Full text of DOI:10.1021/ja109120n

Published on Web 11/18/2010
Palladium-Catalyzed Asymmetric Synthesis of Allylic Fluorides
Matthew H. Katcher and Abigail G. Doyle*
Department of Chemistry, Princeton UniVersity, Princeton, New Jersey 08544, United States
Received October 11, 2010; E-mail: agdoyle@princeton.edu
Table 1. Reaction of Palladium Complex 1 with Various Fluoride
Sources
Abstract: The enantioselective fluorination of readily available
cyclic allylic chlorides with AgF has been accomplished using a
Pd(0) catalyst and Trost bisphosphine ligand. The reactions
proceed with unprecedented ease of operation for Pd-mediated
nucleophilic fluorination, allowing access to highly enantioenriched
cyclic allylic fluorides that bear diverse functional groups. Evi-
dence that supports a mechanism in which C-F bond formation
occurs by an SN2-type attack of fluoride on a Pd(II)-allyl inter-
mediate is presented.
conv.
(%)a
yield 2
(%)a
yield 3
(%)a
entry
“F-” source
1
2
3
4
5
KF
CsF
TBAT
AgF
CuF2
77
96
72
100
36
0
0
14
24
19
4
The extension of palladium-catalyzed cross coupling to the
formation of carbon-fluorine bonds is of widespread interest due
to the growing need for versatile, mild, and selective methods in
the preparation of organofluorine compounds.1 While aryl fluoride
bond formation has been intensively investigated in this context,2
the preparation of aliphatic C-F bonds by palladium catalysis
remains relatively unexplored.3 Our lab has initiated a program
aimed at the development of asymmetric catalytic methods for
nucleophilic fluorination.4,5 As part of this program, we sought a
method for the enantioselective synthesis of allylic fluorides by
palladium-catalyzed nucleophilic substitution. Although allylic
fluorides are key components of bioactive molecules and serve as
versatile synthetic intermediates,6,7 only one direct asymmetric
catalytic method has been identified for their preparation.8 Herein,
we report that a chiral bisphosphine-ligated palladium complex
catalyzes the enantioselective fluorination of allylic chlorides with
AgF. We propose that a unique mechanism for C-F bond formation
is operative, which enables allylic fluorination to take place at room
temperature with high enantioselectivity and broad functional group
compatibility.
10b
49b
0
1
a Determined by 1H NMR using methyl benzoate as a quantitative
internal standard; reactions were conducted in the presence of dimethyl
fumarate (1.2 equiv) to complex residual Pd(0). b >20:1 dr as determined
by 19F NMR.
attack of fluoride on an electrophilic Pd(II)-allyl intermediate
(Scheme 1). This elementary step is well-established for Pd-
catalyzed allylic alkylations with various carbon- and heteroatom-
centered nucleophiles10 but, to the best of our knowledge, has no
precedent for fluoride anion.11,12 We reasoned that accessing allylic
fluorides in this manner would provide a platform for the design
of a broadly useful catalyst-controlled enantio-, diastereo-, and
regioselective methodology that uses readily available and inex-
pensive fluoride sources.13
To evaluate this proposal, we investigated the reaction of known
Pd(II)-allyl complex 114 with representative “F-” sources (Table
1). Consistent with the documented Brønsted basicity of alkali metal
fluorides, reactions with KF and CsF led predominantly to elimina-
tion, providing diene 3 as the major observable product (entries 1
and 2).11 With the less basic fluoride source tetrabutylammonium
difluorotriphenylsilicate (TBAT),15 allylic fluoride trans-2 was
formed, albeit in 10% yield (entry 3). A more promising result was
obtained from the reaction of complex 1 with AgF, which gave
the desired product in almost 50% yield with minimal (<5%) diene
production (entry 4). The selective formation of trans-2 in this
reaction indicates that reductive elimination occurs with inversion
of configuration, consistent with nucleophilic attack of fluoride on
the allyl ligand of 1.10,16 Notably, the efficacy of AgF does not
appear to be tied solely to its potential redox activity since other
transition metal fluorides tested, such as CuF2, were ineffective as
fluoride sources (entry 5).17
Scheme 1. Mechanisms for C-F Bond Formation from Pd(II)
2
Two strategies for palladium-mediated Csp -F bond formation
have been described previously, both involving discrete palladium
fluoride intermediates. In the most widespread approach, methods
for C-H and C-B fluorination have been identified that proceed
through high-valent palladium fluoride complexes, accessed by
oxidation of Pd(II) with electrophilic “F+” sources.2a-d,f In contrast,
C-F bond formation by reductive elimination from Pd(II)ArF
intermediates, generated from Pd(II) precursors with nucleophilic
“F-” reagents, has proven more challenging.9 The recent identifica-
tion of a Pd(0)-catalyzed methodology for the fluorination of aryl
triflates with CsF at elevated temperature (80-120 °C) is the only
successful example to date.2e
After establishing the feasibility of the critical C-F bond-forming
event, we pursued a catalytic variant of the allylic fluorination.
Remarkably, substrates possessing traditional leaving groups for
Pd-catalyzed allylic alkylation, such as allylic acetates and carbon-
ates, were unreactive in the presence of a Pd(0) catalyst and AgF
(Table 2, entries 1 and 2). Instead, efficient C-F bond formation
We hypothesized that efficient and mild allylic C-F bond
formation could proceed by a distinct mechanism: the nucleophilic
9
17402 J. AM. CHEM. SOC. 2010, 132, 17402–17404
10.1021/ja109120n  2010 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Identification of a Catalytic Allylic Fluorination Protocol
Table 3. Palladium-Catalyzed Fluorination of Allylic Chlorides
yield 5
(%)a
ee 5
(%)b
yield 6
(%)a
entry
X
ligand
1
2
OAc (4a)
OCO2Et (4b)
Cl (4c)
Cl (4c)
Cl (4c)
PPh3
PPh3
PPh3
none
L1
0
1
64
8
85
70
-
-
-
-
88
74
1
0
15
1
4
5
3
4c
5d
6d
Cl (4c)
L2
a Determined by GC using decane as a quantitative internal standard.
b Determined by chiral GC using
a commercial chiral column.
c Reaction conducted with AgF (1.1 equiv) in the absence of a Pd
catalyst. d Using Pd2(dba)3 (5 mol %), L1 or L2 (10 mol %), AgF (1.1
equiv).
was observed when allylic chloride 4c was employed as a substrate
(entry 3).18,19 We propose that 4c and Pd(0) combine to generate
an intermediate analogous to 1, from which precipitation of AgCl
provides a driving force for C-F bond formation. Although 4c
reacts with AgF in the absence of palladium,20 background reaction
is suppressed at room temperature and in nonpolar solvents (entry
4).21 Evaluation of chiral ligands under these conditions revealed
that the commercially available Trost ligand L122 imparts high
levels of enantioinduction in the production of allylic fluoride 5
(85% yield, 88% ee, entry 5). Control experiments demonstrated
that the enantioenriched allylic fluoride product does not decompose
or epimerize under the reaction conditions, despite reports docu-
menting the instability of certain allylic fluorides toward Pd(0).11,23
We have found that the identified procedure provides a versatile
protocol for enantioselective allylic fluorination of six-membered
cyclic allylic chlorides.24 The transformation is tolerant of substrates
bearing common functional groups, including ethers, amines, esters,
and amides (Table 3, entries 2-6). Less than 10% diene is observed
in most cases, and the nonvolatile allylic fluorides can be isolated
in moderate to good yields with excellent enantioselectivities.25
Moreover, good levels of efficiency and enantioselectivity are
maintained in the presence of traditionally fluoride-sensitive silyl
ethers (entry 7) and unprotected alcohols (entry 8). These two
examples underscore the mildness of the reaction conditions and
the synthetic utility of the process; by contrast, the synthesis of
enantioenriched allylic fluorides by dehydroxyfluorination with
(diethylamino)sulfur trifluoride (DAST) is typically incompatible
with these substituents and can suffer from poor stereoselectivity.6,26
Moreover,unlikeAr-FbondformationfromPd(II)Fintermediates,2e,27
the allylic fluorinations are conducted under ambient conditions
and without rigorous exclusion of moisture or air.28
a Absolute configuration of 8 assigned by X-ray crystallography;
other products assigned by analogy. b Isolated yield for reactions carried
out on a 0.5 mmol scale. c Determined by chiral GC or HPLC using
commercial chiral columns. d Determined by 1H NMR; dr’s were
unchanged from starting material to product. e Determined by GC using
octane as an internal quantitative standard. f Using commercial L2
instead of L1. g Reaction conducted in toluene with 2.0 equiv of AgF.
h Reaction conducted with [η3-C3H5PdCl]2 (5 mol %) and L1 (15 mol
%). i ee of minor diastereomer in parentheses. j dr of the allylic chloride
was 5:1.
fluorination accesses a product primed for diversification by
sequential Pd-catalyzed substitutions.
Since allylic carbonates and acetates were unreactive under the
conditions described in Table 2, we anticipated that fluorination of
bifunctional substrate cis-12 would proceed in a chemoselective
manner. In the event, fluoroallylic carbonate cis-13 was formed in
80% yield as the only detectable fluorine-containing product (eq
1). Since preferential Pd-catalyzed allylic substitution of allylic
carbonates over allylic fluorides has been reported,23b the catalytic
Notably, the allylic fluorinations in Table 3 and eq 1 deliver
fluoride to the same diastereotopic face as that of the chloride
leaving group in the substrate. To account for this outcome, we
propose that Pd(0) oxidatively adds to the allylic chlorides by an
SN2-type reaction with inversion of configuration.29 Chloride
abstraction by Ag+ prior to or concurrent with an outer-sphere attack
of fluoride on the allyl-Pd intermediate would deliver fluoride with
9
J. AM. CHEM. SOC. VOL. 132, NO. 49, 2010 17403

C O M M U N I C A T I O N S
(8) (a) Ishimaru, T.; Shibata, N.; Horikawa, T.; Yasuda, N.; Nakamura, S.;
Toru, T.; Shiro, M. Angew. Chem., Int. Ed. 2008, 47, 4157–4161. For an
indirect method to acyclic allylic fluorides by asymmetric organocatalytic
R-fluorination, followed by homologation, see: (b) Jiang, H.; Falcicchio,
A.; Jensen, K. L.; Paixa˜o, M. W.; Bertelsen, S.; Jørgensen, K. A. J. Am.
Chem. Soc. 2009, 131, 7153–7157.
(9) (a) Grushin, V. V. Chem.sEur. J. 2002, 8, 1006–1014. (b) Yandulov, D. V.;
Tran, N. T. J. Am. Chem. Soc. 2007, 129, 1342–1358. (c) Grushin, V. V.
Acc. Chem. Res. 2010, 43, 160–171.
(10) For reviews, see: (a) Trost, B. M.; Van Vranken, D. L. Chem. ReV. 1996,
96, 395–422. (b) Lu, Z.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 258–
297. (c) Trost, B. M.; Zhang, T.; Sieber, J. D. Chem. Sci. 2010, 1, 427–
440.
(11) For an unsuccessful investigation, see: Hintermann, L.; La¨ng, F.; Maire,
P.; Togni, A. Eur. J. Inorg. Chem. 2006, 1397–1412.
(12) A computational investigation has been reported that supports the feasibility
of allylic C-F bond formation from a π-allyl palladium complex: Hagelin,
H.; Åkermark, B.; Norrby, P.-O. Chem.sEur. J. 1999, 5, 902–909.
(13) For a discussion of the relative advantages and disadvantages of nucleophilic
and electrophilic fluorine sources, see: Furuya, T.; Kuttruff, C. A.; Ritter,
T. Curr. Opin. Drug DiscoVery DeV. 2008, 11, 803–819, and references
therein.
an overall retention of configuration. As support for this proposal,
and in line with our stoichiometric studies in Table 1, fluorination
of an L1-ligated analog of 1 provided trans-2 in 53% yield and
88% ee (compare to Table 3, entry 4; see Supporting Information
for more details). In addition, the observation that the absolute
configuration of the allylic fluorides is in accord with the predictive
model developed by Trost for Pd-catalyzed asymmetric allylic
alkylations with soft nucleophiles provides circumstantial evidence
for an analogous mechanism.30
In summary, we have described a new methodology for the
enantioselective synthesis of allylic fluorides by Pd(0)-catalyzed
C-F bond formation with AgF. Our future efforts will be directed
toward elucidating the full synthetic scope of the reaction. Ad-
ditionally, studies to understand how a classically hard nucleophile
plays the role of a soft nucleophile in this process are underway.31
(14) Kurosawa, H.; Kajimaru, H.; Ogoshi, S.; Yoneda, H.; Miki, K.; Kasai, N.;
Murai, S.; Ikeda, I. J. Am. Chem. Soc. 1992, 114, 8417–8424.
(15) Pilcher, A. S.; Ammon, H. L.; DeShong, P. J. Am. Chem. Soc. 1995, 117,
5166–5167.
(16) Trost, B. M.; Verhoeven, T. R. J. Am. Chem. Soc. 1980, 102, 4730–4743.
(17) CoF2, MnF2, AgF2, and several other transition metal fluorides were found
to be ineffective as fluoride sources under catalytic conditions similar to
those described in Table 2.
(18) Allylic chlorides can be prepared from allylic alcohols in one step. See
Supporting Information for details.
Acknowledgment. We thank Dr. Pat Carroll of the University
of Pennsylvania for X-ray crystallographic structural determination.
Financial support provided by Princeton University, kind gifts from
Eli Lilly and Sanofi-Aventis, and a fellowship from Eli Lilly to
M.H.K. are gratefully acknowledged. The authors also thank the
MacMillan and Sorensen groups for use of their chemicals and
instruments.
(19) Togni and Mezzetti have reported a ruthenium complex that promotes Br/F
exchange between alkyl, allyl, and benzyl bromides and TlF; see: (a)
Barthazy, P.; Hintermann, L.; Stoop, R.; Wo¨rle, M.; Mezzetti, A.; Togni,
A. HelV. Chim. Acta 1999, 82, 2448–2453. (b) Barthazy, P.; Togni, A.;
Mezzetti, A. Organometallics 2001, 20, 3472–3477.
Supporting Information Available: Experimental procedures,
details of mechanistic experiments and optimization studies, and
spectroscopic data for all new compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(20) Finkelstein reactions between allylic halides and silver fluoride have been
j
reported. For an example, see: Bannai, K.; Toru, T.; Oba, T.; Tanaka, T.;
Okamura, N.; Watanabe, K.; Hazato, A.; Kurozumi, S. Tetrahedron 1986,
42, 6735–6746.
(21) Among the solvents tested, toluene and THF were optimal. Reactions
conducted above room temperature or in more polar solvents showed
increased background reactivity, presumably due in part to the enhanced
solubility of AgF under these conditions.
References
(1) For a recent review, see: Furuya, T.; Klein, J. E. M. N.; Ritter, T. Synthesis
2010, 1804–1821.
(2) (a) Hull, K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc. 2006,
128, 7134–7135. (b) Furuya, T.; Kaiser, H. M.; Ritter, T. Angew. Chem.,
Int. Ed. 2008, 47, 5993–5996. (c) Furuya, T.; Ritter, T. J. Am. Chem. Soc.
2008, 130, 10060–10061. (d) Wang, X.; Mei, T.-S.; Yu, J.-Q. J. Am. Chem.
Soc. 2009, 131, 7520–7521. (e) Watson, D. A.; Su, M.; Teverovskiy, G.;
Zhang, Y.; Garc´ıa-Fortanet, J.; Kinzel, T.; Buchwald, S. L. Science 2009,
325, 1661–1664. (f) Furuya, T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.;
Tang, P.; Goddard, W. A., III; Ritter, T. J. Am. Chem. Soc. 2010, 132,
3793–3807.
(22) Trost, B. M.; Van Vranken, D. L.; Bingel, C. J. Am. Chem. Soc. 1992,
114, 9327–9343.
(23) (a) Narumi, T.; Tomita, K.; Inokuchi, E.; Kobayashi, K.; Oishi, S.; Ohno,
H.; Fujii, N. Org. Lett. 2007, 9, 3465–3468. (b) Hazari, A.; Gouverneur,
V.; Brown, J. M. Angew. Chem., Int. Ed. 2009, 48, 1296–1299. (c) Pigeon,
X.; Bergeron, M.; Barabe´, F.; Dube´, P.; Frost, H. N.; Paquin, J.-F. Angew.
Chem., Int. Ed. 2010, 49, 1123–1127.
(24) We have found that five- and seven-membered allylic chlorides undergo
fluorination with reduced efficiency and enantioselectivity. An acyclic allylic
chloride, (E)-6-chloronon-4-ene, also reacted with low enantioselectivity.
See Supporting Information for details.
(25) Allylic fluorides 2 and 5 are unstable to borosilicate glass in neat form.
See: Lee, E.; Yandulov, D. V. J. Fluorine Chem. 2009, 130, 474–483.
(26) (a) Singh, R. P.; Shreeve, J. M. Synthesis 2002, 2561–2578. (b) Bresciani,
S.; Slawin, A. M.; O’Hagan, D. J. Fluorine Chem. 2009, 130, 537–543.
(27) Pilon, M. C.; Grushin, V. V. Organometallics 1998, 17, 1774–1781.
(28) Allylic fluorinations set up under an inert atmosphere, with rigorous
exclusion of air and moisture, proceed with comparable % yield and % ee.
(29) Oxidative addition of Pd(0) to allylic chlorides with inversion of config-
uration has been documented. See ref 14.
(3) (a) Hamashima, Y.; Sodeoka, M. Synlett 2006, 1467–1478. (b) Wu, T.;
Yin, G.; Liu, G. J. Am. Chem. Soc. 2009, 131, 16354–16355. (c) Qiu, S.;
Xu, T.; Zhou, J.; Guo, Y.; Liu, G. J. Am. Chem. Soc. 2010, 132, 2856–
2857.
(4) Kalow, J. A.; Doyle, A. G. J. Am. Chem. Soc. 2010, 132, 3268–3269.
(5) For recent reviews of asymmetric catalytic C-F bond formation using
electrophilic and nucleophilic fluorine sources, see: (a) Cahard, D.; Xu,
X.; Couve-Bonnaire, S.; Pannecoucke, X. Chem. Soc. ReV. 2010, 39, 558–
568. (b) Ma, J.-A.; Cahard, D. Chem. ReV. 2008, 108, PR1–PR43.
(6) For a review, see: Pacheco, M. C.; Purser, S.; Gouverneur, V. Chem. ReV.
2008, 108, 1943–1981.
(7) For examples of allylic fluoride motifs in pharmaceutical or imaging agents,
see: (a) Khan, M. O. F.; Lee, H. J. Chem. ReV. 2008, 108, 5131–5145. (b)
Stehouwer, J. S.; Daniel, L. M.; Chen, P.; Voll, R. J.; Williams, L.; Plott,
S. J.; Votaw, J. R.; Owens, M. J.; Howell, L.; Goodman, M. M. J. Med.
Chem. 2010, 53, 5549–5557.
(30) Trost, B. M.; Machacek, M. R.; Aponick, A. Acc. Chem. Res. 2006, 39,
747–760.
(31) Parr, R. G.; Pearson, R. G. J. Am. Chem. Soc. 1983, 105, 7512–7516.
JA109120N
9
17404 J. AM. CHEM. SOC. VOL. 132, NO. 49, 2010