Fluorination of nonactivated haloarenes via arynes under mild conditions, resulting from further studies toward Ar-F reductive elimination from palladium(II)

Article abstract of DOI:10.1021/om800520e

Attempted preparation of N,N- and S,S-chelatestabilized arylpalladium fluorides has led to the finding ofarynemediated fluorination of nonactivated haloarenes with Me₄NF in DMSO under mild conditions.

Full text of DOI:10.1021/om800520e

Organometallics 2008, 27, 4825–4828
4825
Fluorination of Nonactivated Haloarenes via Arynes under Mild
Conditions, Resulting from Further Studies toward Ar-F Reductive
Elimination from Palladium(II)
Vladimir V. Grushin* and William J. Marshall
Central Research & DeVelopment, E. I. DuPont de Nemours & Co., Inc., Experimental Station,
Wilmington, Delaware 19880
ReceiVed June 5, 2008
Scheme 1
Summary: Attempted preparation of N,N- and S,S-chelate-
stabilized arylpalladium fluorides has led to the finding of aryne-
mediated fluorination of nonactiVated haloarenes with Me4NF
in DMSO under mild conditions.
Fluoroaromatic compounds often exhibit biological activity and
have been extensively used as building blocks and monomers in
the synthesis of pharmaceuticals, agrochemicals, dyes, specialty
polymers, and various advanced materials.1,2 Despite the widely
recognized usefulness of fluoroarenes, selective monofluorination
of aromatic compounds still represents one of the most significant
practical and intellectual challenges of modern synthetic methodol-
ogy. The Balz-Schiemann reaction3 dealing with costly, toxic, and
often explosive diazonium compounds still remains the main
method for the synthesis of monofluorinated aromatics.
left for several hours at room temperature prior to workup, a different,
brown crystalline material 7 was isolated and identified as a product
of dba insertion into the Pd-Ph bond of 5 (Scheme 2). This type of
migratory insertion of dba into a Pd-Ar bond has been previously
proposed on the basis of the observed formation of arylated dba
products.14 Apparently, 5 is remarkably capable of undergoing
New routes to aromatic C-F bond formation have been highly
sought. For a number of years, we have been working on one such
route, the Pd-catalyzed fluorination of nonactivated haloarenes.4
The problematic step for the proposed catalytic loop (Scheme 1)
is Ar-F reductive elimination from Pd.5 An overview of this
problem has been recently reported,5a emphasizing the preference
of tertiary phosphine stabilized arylpalladium(II) fluorides to take
the undesired reaction path leading to P-F rather than Ar-F bond
formation.4a,5a,6,7 Such P-F bond forming reactions appear to be
general and have also been observed for complexes of Ru,8 Rh,9
Ir,10 and Pt.11 This prompted us to seek Ar-F reductive elimination
from Pd aryls that either were phosphine-free or were stabilized
by phosphines that could be more reluctant to P-F bond formation.
Herein we describe some of these attempts and how they have led
to the finding of aromatic fluorination proceeding via aryne
intermediates under mild conditions.
(2) (a) Hagmann, W. K. J. Med. Chem. 2008, 51, 4359. (b) Müller, K.;
Faeh, C.; Diederich, F. Science 2007, 317, 1881. (c) Thayer, A. M. Chem.
Eng. News 2006, 84, 15.
(3) Balz, G.; Schiemann, G. Chem. Ber. 1927, 60, 1186.
(4) (a) Grushin, V. V. Chem. Eur. J. 2002, 8, 1006. (b) A reviewer’s
comment prompted us to reemphasize (see ref 5a) that Pd-mediated fluorination
reactions reported by Sanford (aromatic),4c Sodeoka (aliphatic),4d and most
recently Ritter (aromatic)4e all employ electrophilic, so-called “positive fluorine”
reagents for the C-F bond formation. This electrophilic fluorination
methodology4c–e cannot be used for the highly sought displacement of X in
nonactivated ArX (X ) I, Br, Cl, OTf, etc.) with fluoride and is therefore rather
irrelevant to our approach to metal-catalyzed aromatic nucleophilic fluorination
with the F- (Scheme 1). Ritter’s recent report4e describes the synthesis of
fluoroarenes ArF by reacting Selectfluor with stoichiometric quantities of (σ-
aryl)palladium reagents. The latter are prepared in five steps, with the overall
yield of the ArF products for the entire reaction sequence being in the range of
ca. 15-40% (usually around 30%), as calculated on the Pd(OAc)2 starting
material. In the step preceding the final fluorination with Selectfluor, the Pd
center is arylated with arylboronic acids. It is noteworthy that in most instances
ArB(OH)2 reagents are made from ArMgX which may be fluorinated directly
with the same Selectfluor reagent to produce the same product in one step and
in higher yield (e.g., 61% for Ar ) Ph).4f While Ritter4e anticipates his method
to find applicaitons in 18F-PET, practical routes to suffieciently active [F-18]-
Selectfluor are nonexistent4g and may never be developed, as suggested by a
number of fundamental scientific and engineering considerations.4h (c) Hull,
K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc. 2006, 128, 7134. (d)
Hamashima, Y.; Sodeoka, M. Synlett 2006, 1467, and references cited
therein. (e) Furuya, T.; Kaiser, M. H.; Ritter, T. Angew. Chem., Int. Ed.,
2008, 47, 5993. (f) Lal, G. S. J. Org. Chem. 1993, 58, 2791. (g) See, for
example: Bolton, R. J. Label Compd. Radiopharm. 2002, 45, 485. Nyffeler,
P. T.; Duron, S. G.; Burkart, M. D.; Vincent, S. P.; Wong, C.-H. Angew.
Chem. Int. Ed. 2005, 44, 192. Ametamey, S. M.; Honer, M.; Schubiger,
P. A. Chem. ReV. 2008, 108, 1501. (h) In a recent highlight, the fluorination
reaction4e is misinterpreted as catalytic in Pd, performed at room temper-
ature, and important for 18F-PET: Nature 2008,454,471.
A series of iodo aryl Pd N,N-chelates, 1-4,12 and the two S,S-
chelates 513 and 6 were synthesized by reacting Pd2(dba)3 with
PhI in the presence of the corresponding ligand.
The newly isolated 5 and 6 were fully characterized, including
X-ray analysis (see the Supporting Information). Interestingly,
working up the reaction with PhSCH2CH2SPh without delay
produced yellow 5, whereas after the reaction solution had been
(5) (a) Grushin, V. V.; Marshall, W. J. Organometallics 2007, 26, 4997.
(b) While being of some academic interest, the most recently reported5c
Ar–F reductive elemination from Pd(IV) cannot provide foundation for the
highly sought catalytic reaction (Scheme 1) because unlike Pd(0), Pd(II)
does not oxidatively add haloarenes. The most closely related Ar–Cl
reductive eleimation from a similar Pd(IV) complex was earlier reported
by Whitfield and Sanford.5d(c) Furuya, T.; Ritter, T. J. Am. Chem. Soc.
2008, 130, 10060. (d) Whitfield, S. R.; Sanford, M. S. J. Am. Chem. Soc.
2007, 129, 15142.
* To whom correspondence should be addressed. E-mail: vlad.grushin-1
@usa.dupont.com.
(1) Clark, J. H.; Wails, D.; Bastock, T. W. Aromatic Fluorination; CRC
Press: Boca Raton, FL, 1996.
10.1021/om800520e CCC: $40.75
2008 American Chemical Society
Publication on Web 09/03/2008

4826 Organometallics, Vol. 27, No. 19, 2008
Communications
Scheme 2
Figure 1. ORTEP drawings of 8 (left) and PPNF (right) with
thermal ellipsoids drawn to the 50% probability level and all H
atoms omitted for clarity.
migratory insertion with olefins, likely due to the labile Pd-S
bonds. Upon Pd-S dissociation a vacant coordination site is created
on Pd, which is occupied by dba prior to its insertion into the
Pd-Ph bond. Note that 6, bearing a more electron-donating t-Bu
group on the sulfur, did not undergo dba insertion under similar
conditions.
Complexes 1-6 were treated with AgF in benzene under
sonication, using our original method15 for the synthesis of R3P-
stabilized arylpalladium fluorides.4,6c–e,7 While the virtually in-
soluble 1 remained intact, all other iodopalladium phenyls readily
reacted. The reactions produced either dark precipitates (2 and 3)
or microcrystalline off-white solids which were not amenable to
study due to insolubility (4-6). In all cases, neither were isolable
palladium fluorides produced nor was C-F bond formation
observed (19F NMR).
Attempted preparation of [(PTA)2Pd(Ph)F] from 8 and AgF under
sonication15 failed, due to the insufficient solubility of 8 in benzene
and toluene. The ultrasound-free reaction of 8 with AgF in CH2Cl2
produced a complex reaction mixture of decomposition products.19
Hence, we explored the fluorination of bromobenzene with 8 or 9
as catalysts. With anhydrous CsF or KF in the presence or in the
absence of 18-crown-6 in MeCN, glyme, diglyme, dioxane, or
DMSO at 80-130 °C, no PhF formation was observed. To promote
the in situ Pd-F bond formation, we then considered sources of
most active fluoride,20–22 often referred to as “naked” fluoride,
including anhydrous [Me4N]F20 and [Ph3PdNdPPh3]F (PPNF).22,23
24
Although the latter did form on treatment of [PPN]CN with C6F6
(Figure 1),25 its isolation in pure, colorless form in bulk appeared
problematic. Therefore, [Me4N]F was chosen.
We were excited to observe the formation of fluorobenzene (19F
NMR) from PhBr upon treatment with [Me4N]F in DMSO in the
presence of 5-10% of 8 or 9 at 80-120 °C. The yield of Ph-F
was modest, corresponding to only 3-5 equiv per Pd complex
used. Attempted optimization was not successful but led to a critical
observation: when 2-bromonaphthalene was used as the substrate
in the presence of 9 at 105 °C, a ca. 3:2 mixture of 2-fluoronaph-
thalene and 1-fluoronaphthalene was obtained, immediately sug-
gesting aryne intermediacy.The same result was obtained when
the experiment was repeated in the absence of 9 or any other source
of palladium (eq 1). Furthermore, the yield of PhF from PhBr was
higher when the Pd complex was omitted from the reaction. It
became clear that the fluorination occurred via arynes and was not
catalyzed by the Pd species.
The outcome of the reactions of 2-6 with AgF indicated that
N and S ligands are not as capable of stabilizing the Pd-F bond
as tertiary phosphines.16 Hence, we turned our attention back to
R3P ligands, in search of ones that would be resistant to the
undesired P-F bond formation. The latter likely occurs via
intramolecular attack of the fluoride on the phosphorus to produce
a metallophosphorane intermediate,17 a process that requires
significant geometry changes around the P atom. This mechanistic
consideration pointed to 1,3,5-triaza-7-phosphaadamantane (PTA)18
as a viable candidate, a cage phosphine which, upon coordination
to a metal, was expected to be less prone to metallophosphorane
formation due to its rigidity.
The complexes [(PTA)2Pd(Ph)X], where X ) I (8), Br (9), were
prepared in >90% yield by reacting [(Ph3P)2Pd(Ph)X] with PTA
and found to be trans in solution and in the solid state (Figure 1).
(6) (a) Grushin, V. V. Organometallics 2000, 19, 1888. (b) Roe, D. C.;
Marshall, W. J.; Davidson, F.; Soper, P. D.; Grushin, V. V. Organometallics
2000, 19, 4575. (c) Grushin, V. V.; Marshall, W. J. Angew. Chem., Int. Ed.
2002, 41, 4476. (d) Marshall, W. J.; Grushin, V. V. Organometallics 2003,
22, 555. (e) Grushin, V. V.; Marshall, W. J. J. Am. Chem. Soc. 2006, 128,
12644.
(7) Yandulov, D. V.; Tran, N. T. J. Am. Chem. Soc. 2007, 129, 1342.
(8) (a) den Reijer, C. J.; Woerle, M.; Pregosin, P. S. Organometallics
2000, 19, 309. (b) den Reijer, C. J.; Dotta, P.; Pregosin, P. S.; Albinati, A.
Can. J. Chem. 2001, 79, 693. (c) Geldbach, T. J.; Pregosin, P. S. Eur.
J. Inorg. Chem. 2002, 1907.
(9) (a) Grushin, V. V.; Marshall, V. J. J. Am. Chem. Soc. 2004, 126,
3068. (b) Macgregor, S. A.; Roe, D. C.; Marshall, W. J.; Bloch, K. M.;
Bakhmutov, V. I.; Grushin, V. V. J. Am. Chem. Soc. 2005, 127, 15304. (c)
Macgregor, S. A.; Wondimagegn, T. Organometallics 2007, 26, 1143.
(10) Blum, O.; Frolow, F.; Milstein, D. J. Chem. Soc., Chem. Commun.
1991, 258.
(11) Jasim, N. A.; Perutz, R. N.; Whitwood, A. C.; Braun, T.; Izundu, J.;
Neumann, B.; Rothfeld, S.; Stammler, H.-G. Organometallics 2004, 23, 6140.
(12) (a) Markies, B. A.; Canty, A. J.; de Graaf, W.; Boersma, J.; Janssen,
M. D.; Hogerheide, M. P.; Smeets, W. J. J.; Spek, A. L.; van Koten, G. J.
Organomet. Chem. 1994, 482, 191. (b) Marshall, W. J.; Grushin, V. V.
Can. J. Chem. 2005, 83, 640.
While the century-old chemistry of arynes is widely used for
aryl-carbon and aryl-heteroatom bond formation,26 information
on aryne-mediated fluorination is scarce and lacking in detail. We
believe that carbon-fluorine bond formation via arynes was
previously observed by Schwesinger’s group. However, in their
otherwise meticulous and highly detailed recent account of “naked”
fluoride basicity and nucleophilicity, Schwesinger et al.21b mention
this type of fluorination only in one brief sentence in the Con-
clusion, without providing any specifics or references to conven-
tionally accessible literature sources,27 except for one patent.28
(15) (a) Fraser, S. L.; Antipin, M. Y.; Khroustalyov, V. N.; Grushin,
V. V. J. Am. Chem. Soc. 1997, 119, 4769. (b) Pilon, M. C.; Grushin, V. V.
Organometallics 1998, 17, 1774.
(13) Complex 5 has been previously detected but not isolated: Paladino,
G.; Madec, D.; Prestat, G.; Maitro, G.; Poli, G.; Jutand, A. Organometallics
2007, 26, 455.
(14) Stambuli, J. P.; Incarvito, C. D.; Bu¨hl, M.; Hartwig, J. F. J. Am.
Chem. Soc. 2004, 126, 1184.
(16) Marshall, W. J.; Grushin, V. V. Organometallics 2004, 23, 3343.
(17) Macgregor, S. A. Chem. Soc. ReV. 2007, 36, 67.
(18) (a) Daigle, D. J. Inorg. Synth. 1998, 32, 40. (b) Phillips, A. D.;
Gonsalvi, L.; Romerosa, A.; Vizza, F.; Peruzzini, M. Coord. Chem. ReV.
2004, 248, 955, and references cited therein.

Communications
Organometallics, Vol. 27, No. 19, 2008 4827
Table 1. Aromatic Fluorination of Nonactivated Haloarenes (3
While an aryne mechanism might intuitively be most plausible for
the patented fluorinations,28 it does not seem to be consistent with
some of the results claimed, such as the exclusive formation of
1-chloro-2-fluorobenzene from 1,2-dichlorobenzene28 (see below).
Our results are summarized in Table 1. All reactions were run
in glass vessels with agitation. The reaction mixtures were analyzed
by GC-MS and 19F NMR with an internal standard, and the
isomeric structures of the fluoroarene products were unambiguously
assigned on the basis of their distinct and characteristic 19F NMR
chemical shifts.29Bromoarenes gave rise to the fluorinated products
in higher yields than analogous chloroarenes. In full accord with
this, the reaction of 1-chloro-3-bromobenzene led to the formation
of 1-chloro-3-fluorobenzene and 1-bromo-3-fluorobenzene in a ca.
100:1 ratio (Table 1, entry 2), along with traces of the ortho isomers.
This indicated that, after deprotonation of the most acidic C-H
bond between the Cl and Br atoms, the next step involves
preferential elimination of the Br-, a better leaving group than
chloride in anhydrous DMSO. Similarly, the reaction of 1-bromo-
3-iodobenzene with [Me4N]F produced 1-bromo-3-fluorobenzene
as the main fluoroarene product. The isomer distribution patterns
for the bromotoluene and bromoanisole series (Table 1, entries
Equiv) with [Me4N]F (1 Equiv) in DMSO at Initial [F-] ≈ 10%
(19) (a) Since this modification of our original method15 was proposed,19b
it has reportedly failed in the synthesis of other Pd(II) fluorides6e,7 but appeared
useful for the preparation of certain Ir(III) fluoro complexes.19c(b) Yahav, A.;
Goldberg, I.; Vigalok, A. J. Am. Chem. Soc. 2003, 125, 13634. (c) Bourgeois,
C. J.; Garratt, S. A.; Hughes, R. P.; Larichev, R. B.; Smith, J. M.; Ward,
A. J.; Willemsen, S.; Zhang, D.; DiPasquale, A. G.; Zakharov, L. N.;
Rheingold, A. L. Organometallics 2006, 25, 3474.
(20) (a) Christe, K. O.; Wilson, W. W. J. Fluorine Chem. 1990, 47,
117. (b) Christe, K. O.; Wilson, W. W.; Wilson, R. D.; Bau, R.; Feng,
J. A. J. Am. Chem. Soc. 1990, 112, 7619.
(21) (a) Schwesinger, R.; Link, R.; Thiele, G.; Rotter, H.; Honert, D.;
Limbach, H. H.; Maennle, F. Angew. Chem., Int. Ed. Engl. 1991, 30, 1372.
(b) Schwesinger, R.; Link, R.; Wenzl, P.; Kossek, S. Chem. Eur. J. 2006, 12,
438.
(22) Grushin, V. V. Angew. Chem., Int. Ed. 1998, 37, 994.
(23) (a) There have been a few reports describing the synthesis of PPNF
in methanol,23b water,23c and aqueous acetonitrile.23d The preparation
methods for these forms of PPNF, along with their reported characterization
and reactivity, suggest that none of them is a source of highly active fluoride
such as the PPNF generated in situ under rigorously anhydrous conditions.22
(b) Douglas, W.; Ruff, J. K. J. Organomet. Chem. 1974, 65, 65. (c)
Martinsen, A.; Songstad, J. Acta Chem. Scand., Ser. A 1977, 31, 645. (d)
Berkessel, A.; Brandenburg, M. Org. Lett. 2006, 8, 4401.
3-8) are remarkably similar to those reported in the classical papers
by Roberts and co-workers30 for haloarene aminations via arynes.
Absolute yields of the ArF products (Table 1) can be calculated
only if the stoichiometry of the reaction is known. Mechanistic
considerations suggest that 3 equiV of Me4NF might be needed
for the introduction of one fluorine atom into the aromatic ring
via the aryne mechanism (Scheme 3). In the first step of the
reaction, a CH bond ortho to the halogen is deprotonated by the
first equivalent of fluoride that is highly basic,20–22 despite its high
solvation energy in DMSO.31 As a result, HF is formed that
instantly consumes the second equivalent of F- to form stable
bifluoride, FHF-. Indeed, large quantities of bifluoride are always
produced in the reaction (19F NMR). The third equivalent of
fluoride is needed for nucleophilic addition to the aryne in the C-F
bond forming step.32 Thus, the fluoride likely plays a triple role in
the reaction, namely C-H deprotonation, neutralization of the HF
produced, and nucleophilic addition across the formally triple bond
of the aryne electrophile. One might argue that the reaction is
stoichiometric in the fluoride as the nucleophile but only catalytic
in the F- as the base. If the FHF- produced in the first step
(Scheme 3) added to the aryne and/or if the fluoroaryl anion
deprotonated the bifluoride, then the F- that originally acted as
the base would be regenerated. It is unlikely, however, that
bifluoride can successfully compete with much more nucleophilic
fluoride for addition to the aryne electrophile. Furthermore, as has
been observed in this work and reported by others,33 anhydrous
(24) As described for the synthesis of anhydrous [Bu4N]F: Sun, H.;
DiMagno, S. G. J. Am. Chem. Soc. 2005, 127, 2050.
(25) The covalent form of PPNF (Figure 1) was obtained by reacting
[PPN]CN with C6F6 in MeCN, followed by treatment with THF and
precipitation with ether. When the reaction was repeated in MeCN or EtCN at
a high concentration, treatment with THF produced other crystalline forms of
PPNF, in which no covalent P-F bond was present (X-ray). Numerous attempts
to obtain an accurate enough structure of those ionic forms failed. The P-F
bond in the covalent form of PPNF (Figure 1) is uncommonly long in the
crystalline state (1.8196(9) Å) and is ionized in solution (MeCN, DMSO), as
judged by the lack of P-F coupling in the 19F and 31P NMR spectra.
(26) (a) Hoffmann, R. W. Dehydrobenzene and Cycloalkynes; Academic
Press: New York, 1967. (b) Miller, J. Aromatic Nucleophilic Substitution;
Elsevier: London, 1968. (c) March, J. AdVanced Organic Chemistry, 4th
ed.; Wiley: New York, 1992. (d) Pellissier, H.; Santelli, M. Tetrahedron
2003, 59, 701, 2575. (e) Wenk, H. H.; Winkler, M.; Sander, W. Angew.
Chem. Int. Ed. 2003, 42, 502. (f) Winkler, M.; Wenk, H. H.; Sander, W. In
ReactiVe Intermediate Chemistry; Moss, R. A., Platz, M. S., Jones, M., Jr.,
Eds.; Wiley: Hoboken, NJ, 2004; pp 741-794. (g) Pen˜a, D.; Pe´rez, D.;
Guitia´n, E. Angew. Chem., Int. Ed. 2006, 45, 3579.
(27) Two Ph.D. theses, one by R. Link and the other by P. Wenzl, as
cited in refs 58 and 60 of the account.21b We have not been able to find
either thesis and/or its abstract in the STN International Dissertation
Database, Scopus, and Scifinder.
(28) Nobori, T.; Fujiyoshi, S.; Hara, I.; Hayashi, T.; Shibahara, A.;
Funaki, K.; Mizutani, K.; Kiyono, S. U.S. Patent 6,469,224, 2002.
(29) (a) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165.
(b) Sterk, H.; Fabian, W. Org. Magn. Reson. 1975, 7, 274. (c) Arunima;
Kurur, N. D. Magn. Reson. Chem. 2005, 43, 132.
(30) (a) Roberts, J. D.; Semenow, D. A.; Simmons, H. E., Jr.; Carlsmith,
L. A. J. Am. Chem. Soc. 1956, 78, 601. (b) Roberts, J. D.; Vaughan, C. W.,
Jr.; Carlsmith, L. A.; Semenow, D. A. J. Am. Chem. Soc. 1956, 78, 611.

4828 Organometallics, Vol. 27, No. 19, 2008
Communications
Scheme 3
Figure 2. 19F NMR signals from CH3F, CH2DF, CHD2F, and CD3F
(JH-F ) 46 Hz; JD-F ) 7 Hz) formed upon heating of Me4NF in
DMSO-d6 at 90 °C for 24 h.
[Me4N]F does deprotonate DMSO to produce bifluoride in the
temperature range used for the fluorination reactions (Table 1). For
instance, we found that, at 90 °C, a 10% solution of [Me4N]F in
anhydrous DMSO produced bifluoride in ca. 5% and 10% yield
(19F NMR) after 1 and 2.2 h, respectively (for more data, see
below). Furthermore, a phosphazenium fluoride has been demon-
strated to deprotonate such weak acids as dibenzocycloheptane (pKa
in DMSO 32.8) and 4-methylbiphenyl (extrapolated pKa in DMSO
37.6).21b For comparison, pKa values of DMSO have been reported
at 31.334 and 35.1.35
On the basis of the carefully proposed stoichiometry (Scheme
3), the total fluorination yields in the reactions employing ArX
substrates in excess (Table 1) were in the range of 10-65%, as
calculated on the amount of Me4NF used. The NMR yield of PhF
from PhBr after 12 h at 110 °C was calculated at 60%. All
bromoanisoles and bromotoluenes, being less reactive, gave rise to
the corresponding fluoroarenes in only 10-30% yield. The two
isomeric fluoronaphthalenes (Table 1, entry 9) were produced in
65% total yield. The yield of 9-fluorophenanthrene (entry 10) was
55%.
to the DMSO solvent being the primary proton (deuteron) donor
in the final step of the fluoroarene formation.
As follows from the above, it would be desirable to identify an
aprotic solvent which, while providing good solubility and stability
to Me4NF, was even less reactive to the active fluoride than DMSO
in the required temperature range. This appears to be nontrivial,
as DMF, NMP, MeCN, and ethereal solvents are considerably less
stable toward strong bases, Me4NF included. When the 9-bro-
mophenanthrene reaction (Table 1, entry 10) was repeated in
HMPA, less than 10% conversion was observed and only trace
amounts of 9-fluorophenanthrene were formed, along with com-
parable small quantities of phenanthrene. DMSO seems to be
especially suitable as a medium for the reaction. We carefully
propose that the CH acidity of DMSO is in the right range for the
fluorination reaction, being low enough for the active fluoride to
survive yet sufficient for serving as a proton donor to the carbanion
emerging upon F- addition to the aryne electrophile. It is also
noteworthy that Me4NF was found to be insoluble in pivalonitrile
even at its boiling point of around 107 °C.
The remarkably high reactivity of “naked” fluoride,20–22 both
as a base and as a nucleophile, is the key for the aryne-mediated
aromatic fluorination, yet is also a yield-lowering factor in this
reaction. As a strong base, Me4NF can deprotonate even very
weakly CH-acidic DMSO in the temperature range required for
the aryne generation. As a strong nucleophile, the F- in Me4NF is
known to react with its own counterion to give MeF and Me3N.20
After a 5% solution of Me4NF in DMSO-d6 was kept at 90 °C for
24 h, 1H and 19F NMR analysis indicated that approximately 25%
of the salt had decomposed to produce FDF- (-142.5 ppm, t, JD-F
) 18 Hz) and FHF- (-142.0 ppm, d, JH-F ) 120 Hz) in a 3.7:1
molar ratio, and all four isotopomers of MeF (Figure 2), pointing
to efficient H/D exchange between the solvent and the Me4N+
In conclusion, N,N- and S,S-chelate-stabilized palladium iodo
aryls neither form isolable Pd-F species nor produce fluoro-
arenes upon treatment with AgF under sonication. These studies,
however, have led to the finding of aryne-mediated fluorination
of nonactivated haloarenes with Me4NF in DMSO under mild
conditions (80-110 °C).
Acknowledgment. This is DuPont CR&D Contribution
No. 8826. We thank Dr. David L. Davies for bringing to
our attention rigid cage phosphine ligands, Professor Stuart
A. Macgregor for communicating this idea to us, and Dr.
Alexander A. Kolomeitsev for discussions.
Supporting Information Available: Text giving experimental
details and CIF files giving X-ray analysis data for 5-8 and PPNF.
This material is available free of charge via the Internet at
http://pubs.acs.org.
1
cation. In the H NMR spectrum, the exchange was manifested
by multiplet resonances from the H,D isotopomers of Me3N,
Me4N+, and MeF at 2.1, 3.2, and 4.2 ppm, respectively. Clearly,
both DMSO and Me4N+ undergo deprotonation by the strongly
basic fluoride under these conditions. Note that to deprotonate
DMSO and Me4N+ 36 strong bases are employed, such as NaH
and PhLi, respectively. When the PhBr fluorination with Me4NF
(Table 1) was repeated in DMSO-d6, a significant incorporation
of deuterium into the fluorobenzene product was observed (19F
NMR and GC-MS). This is conceivably due to H/D exchange at
the first deprotonation step that is probably reversible, as well as
OM800520E
(32) A side reaction of fluoride addition to an independently generated
aryne has been documented: Kolomeitsev, A. A.; Vorobyev, M.; Gillandt,
H. Tetrahedron Lett. 2008, 49, 449.
(33) Bastos, M. M.; Barbosa, J. P.; Pinto, A. C.; Kover, W. B.; Takeuchi,
Y.; Boechat, N. J. Braz. Chem. Soc. 2001, 12, 417.
(34) Steiner, E. C.; Gilbert, J. M. J. Am. Chem. Soc. 1965, 87, 382.
(35) Olmstead, W. N.; Margolin, Z.; Bordwell, F. G. J. Org. Chem.
1980, 45, 3295.
(36) (a) Wittig, G.; Wetterling, M.-H. Liebigs Ann. Chem. 1947, 557,
193. (b) Wittig, G.; Krauss, D. Liebigs Ann. Chem. 1964, 679, 34.
(31) Dillow, G. W.; Kebarle, P. J. Am. Chem. Soc. 1988, 110, 4877.