Ar-F reductive elimination from palladium(II) revisited

Article abstract of DOI:10.1021/om700469k

In contrast with [(o-Tol₃P)₂Pd₂(p-C ₆H₄X)₂(μ-F)₂] (1; X = NO ₂; J. Am. Chem. Soc. 2007, 129, 1342), its nonactivated congeners (X = H, Me, and MeO) do not produce ¹⁹F NMR-detectable quantities of p-FC₆H₄X upon thermal decomposition in the presence of Buchwald's ligand t-Bu₂P-2-C₆H₄C ₆H₂(i-Pr)₃-2′,4′,6′ (BL). These results do not support the previously asserted net Ar-F reductive elimination , given that for X = NO₂ some quantities of p-FC₆H₄X might conceivably be formed from 1 via a variety of other, S_NAr-type paths.

Full text of DOI:10.1021/om700469k

Organometallics 2007, 26, 4997-5002
4997
Ar-F Reductive Elimination from Palladium(II) Revisited†
Vladimir V. Grushin* and William J. Marshall
Central Research and DeVelopment, E. I. DuPont de Nemours & Co., Inc., Experimental Station,
Wilmington, Delaware 19880-0328
ReceiVed May 11, 2007
In contrast with [(o-Tol3P)2Pd2(p-C6H4X)2(µ-F)2] (1; X ) NO2; J. Am. Chem. Soc. 2007, 129, 1342),
its nonactivated congeners (X ) H, Me, and MeO) do not produce 19F NMR-detectable quantities of
p-FC6H4X upon thermal decomposition in the presence of Buchwald’s ligand t-Bu2P-2-C6H4C6H2(i-Pr)3-
2′,4′,6′ (BL). These results do not support the previously asserted “net Ar-F reductive elimination”,
given that for X ) NO2 some quantities of p-FC6H4X might conceivably be formed from 1 via a variety
of other, SNAr-type paths.
Scheme 1
Introduction
Over the past 15 years, remarkable progress has been made
in the area of Pd-catalyzed C-N and C-O bond forming
reactions of nonactivated haloarenes.1 An efficient similar
process leading to aryl fluorides (eq 1) is highly sought but has
not been reported, thus far.2 Thus, the Balz-Schiemann reaction
of costly and hazardous dizaonium compounds, first published
in 1927,3 remains the only practical method to selectively
introduce fluorine into the aromatic ring.4
solution, and demonstration that the problematic step of the
desired catalytic loop (Scheme 1) is Ar-F reductive elimination
from the Pd(II) center (step 3). Thermal enforcement of
reductive elimination from aryl palladium fluorides led to no
ArF formation, and only P-C and P-F bond forming reactions
occurred instead.11
In 2002, a concept article5 was published on potential use of
palladium catalysis for nucleophilic fluorination of haloarenes
(eq 1 and Scheme 1). This paper5 summarized our early work
in the area,6-12 including the synthesis of the first aryl palladium
fluorides, their full characterization in the solid state and in
Since then, we have made a number of attempts to suppress
the P-F bond formation and promote the desired Ar-F
reductive elimination. Mono- and dinuclear phenyl palladium
fluorides stabilized by trialkylphosphines, including Me3P and
bulky Cy3P and i-Pr3P have been prepared13 and found to
produce no PhF upon thermal decomposition. A novel phenyl
palladium fluoride has been designed, [(BINAP(O))Pd(Ph)F],
in which the F ligand is cis to the Ph and trans to the only
P-donor atom on the molecule.14 Nonetheless, the thermal
decomposition of this complex did not result in Ph-F bond
formation and only P-F and P-C bonds were formed.14 Having
recently solved the problem of very difficult15,16 Ar-CF3
reductive elimination from Pd(II) by using Xantphos,17 we
attempted the thermal decomposition of [(Xantphos)Pd(F)Ph].
Again, however, no C-F reductive elimination took place and
only P-F bond formation was observed.17
† DuPont CR&D contribution No. 8783.
* Corresponding author. E-mail: vlad.grushin-1@usa.dupont.com.
(1) For selected reviews, see: Muci, A. R.; Buchwald, S. L. Top. Curr.
Chem. 2002, 219, 131. Hartwig, J. F. Synlett 2006, 1283. Schlummer, B.;
Scholz, U. AdV. Synth. Catal. 2004, 346, 1599.
(2) The high demand for new reactions leading cleanly to Ar-F bond
formation is largely driven by biological activity of selectively fluorinated
organic molecules. “As many as 30-40% of agrochemicals and 20% of
pharmaceuticals on the market are estimated to contain fluorine, including
half of the top 10 drugs sold in 2005.” Thayer, A. M. Chem. Eng. News
2006, 84, No. 23 (June 5), 15.
(3) Balz, G.; Schiemann, G. Chem. Ber. 1927, 60, 1186.
(4) Aromatic substrates containing strong electron-withdrawing groups
(e.g., NO2 and CN) on the ring are well-known to undergo SNAr fluorination
via a Meisenheimer complex.4a Similarly, polyfluorinated benzenes can be
made from the corresponding polychlorinated compounds and alkali
fluorides in the so-called “Halex” process.4b With highly reactive fluoride
sources employed, aromatic nucleophilic fluorination of activated, electron-
deficient substrates can easily occur at as low as room temperature.4c (a)
Adams, D. J.; Clark, J. H. Chem. Soc. ReV. 1999, 28, 225. (b) Langlois, B.;
Gilbert, L.; Forat, G. Ind. Chem. Libr. 1996, 8, 244. (c) Sun, H.; Dimagno,
S. G. Angew. Chem., Int. Ed. 2006, 45, 2720.
Unlike reductive elimination from Pd(II), which commonly
occurs from a tricoordinate species and thus may be somewhat
controlled by adding or scavenging extra phosphine, the P-F
bond forming reactions at Pd11,13,14,17 and Rh18,19 do not require
(5) Grushin, V. V. Chem.-Eur. J. 2002, 8, 1006.
(6) Fraser, S. L.; Antipin, M. Yu.; Khroustalyov, V. N.; Grushin, V. V.
J. Am. Chem. Soc. 1997, 119, 4769.
(11) Grushin, V. V. Organometallics 2000, 19, 1888.
(12) Roe, D. C.; Marshall, W. J.; Davidson, F.; Soper, P. D.; Grushin,
V. V. Organometallics 2000, 19, 4575.
(7) Pilon, M. C.; Grushin, V. V. Organometallics 1998, 17, 1774.
(8) Marshall, W. J.; Thorn, D. L.; Grushin, V. V. Organometallics 1998,
17, 5427.
(9) Flemming, J. P.; Pilon, M. C.; Borbulevitch, O. Ya.; Antipin, M.
Yu.; Grushin, V. V. Inorg. Chim. Acta 1998, 280, 87.
(13) Grushin, V. V.; Marshall, W. J. Angew. Chem., Int. Ed. 2002, 41,
4476.
(14) Marshall, W. J.; Grushin, V. V. Organometallics 2003, 22, 555.
(15) Culkin, D. A.; Hartwig, J. F. Organometallics 2004, 23, 3398.
(16) Grushin, V. V.; Marshall, W. J. J. Am. Chem. Soc. 2006, 128, 4632.
(17) Grushin, V. V.; Marshall, W. J. J. Am. Chem. Soc. 2006, 128, 12644.
(10) Grushin, V. V. Angew. Chem., Int. Ed. 1998, 37, 994.
10.1021/om700469k CCC: $37.00 © 2007 American Chemical Society
Publication on Web 08/28/2007

4998 Organometallics, Vol. 26, No. 20, 2007
Grushin and Marshall
coordinative unsaturation (see below). These reactions are
governed by a different mechanism that involves intramolecular
nucleophilic attack of the coordinated fluoride on the phosphine
ligand to produce a metallophosphorane intermediate.20,21
Avoiding P-ligands altogether was obvious to consider, even
though tertiary phosphines seem to play an important role in
stabilization of the late transition metal-fluorine bond.22 Our
brief exposure to N-heterocyclic carbenes (NHCs)23 made it
clear that NHC-Ar reductive elimination from Pd(II) can be
remarkably facile and hence kinetically favored over the desired
Ar-F bond formation. That prompted us to discontinue our
work with NHCs as suitable ligands for reaction 1. The first
examples of Ar-F bond formation from ArX (Ar ) phenyl,
2-naphthyl; X ) I, Br) in the presence of CuF2 and N,N-ligands
were disclosed by us in 2005 and since then have received patent
protection.24 We are not aware of any other reported metal-
induced transformations leading to fluoroarenes from the
corresponding nonactiVated aryl iodides, bromides, chlorides,
triflates, etc.25
nonactivated aryls (Y * a strong electron acceptor, such as NO2
or CN in eq 1). Strangely, the Yandulov-Tran account, while
being vast and apparently complete, did not mention Ar-F bond
formation from similar fluoro palladium aryls devoid of an
electron-withdrawing activator on the benzene ring. Because
knowing if the Ar-F bond can be formed without this activator
is critical for further developments in the area, we have now
carried out such experiments and describe them herein.
Results
Analogues of 1 with H (3), Me (4), and MeO (5) in place of
the nitro group were synthesized by our original method6,7,13
from the corresponding iodides31 and AgF under sonication (eq
4). All three complexes were isolated as slightly yellowish
crystalline solids and characterized by elemental analysis, NMR,
and single-crystal X-ray diffraction data.
Most recently, Yandulov and Tran26 reported their compu-
tational study of Ar-F reductive elimination from Pd(II) to
reconfirm our conclusions above. They also described some
experimental work, including a new example of Ar-F bond
formation leading to p-FC6H4NO2 in ca. 10% yield upon thermal
decomposition of [L2Pd2(p-C6H4NO2)2(µ-F)2], where L )
o-Tol3P (1) or t-Bu3P (2), in the presence of Buchwald’s ligand
(BL), t-Bu2P-2-C6H4C6H2(i-Pr)3-2′,4′,6′ (eq 2).26,27 The forma-
tion of Ar-F was reported26 only for Ar ) p-C6H4NO2, a
Meisenheimer-type aryl whose para- (and ortho-) halide deriva-
tives readily undergo SNAr reactions,28,29 nucleophilic fluorina-
tion included (eq 3),4,30 in the absence of any catalyst. For this
As seen from the ORTEP drawings in Figures 1-4, all
fluoride dimers 3, 4, 5, and also 1 are syn in the crystalline
state. This is in contrast with our earlier X-ray structures of
anti-[(R3P)2Pd2(Ph)2(µ-F)2] (R ) i-Pr, Cy)13 and similar to the
syn geometry found for crystalline 2.26 Selected geometry
parameters for the structures are collected in Table 1.
The values in Table 1 point to an increase in both Pd-F and
Pd-C bond lengths when going from the more electron-deficient
p-nitrophenyl complex (1) to the electroneutral phenyl (3) and
further to electron-enriched p-tolyl (4) and p-anisyl (5) dimers.
As expected,8 the trend is more pronounced for the Pd-F bond
distances trans to the aryls, in accord with the structural data
previously reported for trans-[(Ph3P)2Pd(F)(Ph)]6 and trans-
[(Ph3P)2Pd(F)(p-C6H4NO2)].8 The effect may be rationalized in
terms of alleviation of pπ(F)-dπ(Pd) filled/filled repulsion via
push-pull interaction of the lone electron pair on F with π* of
the σ-aryl through filled d orbitals on the metal.8,9 Unlike
[(Cy3P)2Pd2(Ph)2(µ-F)2], which forms a stable adduct with three
reason (compare eqs 2 and 3), in our studies we deal only with
(18) Grushin, V. V.; Marshall, W. J. J. Am. Chem. Soc. 2004, 126, 3068.
(19) 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.
(20) Macgregor, S. A. Chem. Soc. ReV. 2007, 36, 67.
(21) Macgregor, S. A.; Wondimagegn, T. Organometallics 2007, 26,
1143.
(22) Marshall, W. J.; Grushin, V. V. Organometallics 2004, 23, 3343.
(23) Marshall, W. J.; Grushin, V. V. Organometallics 2003, 22, 1591.
(24) Grushin, V. U.S. Patent 7,202,388, 2007.
(25) Hull, Anani, and Sanford have recently reported an interesting exotic
example of fluorination of Ar-H bonds via cyclopalladation, followed by
electrophilic cleavage of the Pd-C bond with N-fluoropyridinium reagents.
This approach cannot allow the highly sought nucleophilic displacement
of Cl, Br, and I of nonactivated haloarenes with fluoride by definition. Hull,
K. L.; Anani, W. Q.; Sanford, M. S. J. Am. Chem. Soc. 2006, 128, 7134.
(26) Yandulov, D. V.; Tran, N. T. J. Am. Chem. Soc. 2007, 129, 1342.
(27) The use26 of BL was apparently inspired by the recent remarkable
development of the synthesis of ArOH from ArX (X ) Br, Cl) and alkali,
catalyzed by Pd2dba3/BL: Anderson, K. W.; Ikawa, T.; Tundel, R. E.;
Buchwald, S. L. J. Am. Chem. Soc. 2006, 128, 10694.
(28) Terrier, F. Nucleophilic Aromatic Displacement: The Influence of
the Nitro Group; VCH: New York, 1991.
(29) Miller, J. Aromatic Nucleophilic Substitution; Elsevier: London
1968.
Figure 1. ORTEP drawing of 1 with thermal ellipsoids drawn at
the 50% probability level and all H atoms omitted for clarity.

Ar-F ReductiVe Elimination from Palladium(II) ReVisited
Organometallics, Vol. 26, No. 20, 2007 4999
Table 1. Selected Geometry Parameters for 1, 3, 4, and 5
geometry param,
Å or deg
1‚5C6H6 3‚3CH2Cl2 4‚2CH2Cl2‚2C6H14 5‚5CH2Cl2
(X ) NO2) (X ) H)
(X ) Me)
(X ) MeO)
Pd(1)-F(1)
Pd(2)-F(1)
Pd(1)-F(2)
Pd(2)-F(2)
Pd(1)-C
2.102(2)
2.103(2)
2.087(2)
2.087(2)
1.972(3)
1.972(3)
2.234(1)
2.234(1)
2.105(2)
2.124(2)
2.094(2)
2.092(2)
1.980(3)
1.976(4)
2.216(1)
2.216(1)
99.6(1)
101.0(1)
79.4(1)
79.1(1)
97.4(1)
98.4(1)
90.2(1)
90.6(1)
2.123(2)
2.123(2)
2.086(2)
2.086(2)
1.984(4)
1.984(4)
2.229(1)
2.229(1)
99.1(1)
101.5(2)
79.8(1)
79.8(1)
97.2(1)
97.2(1)
90.2(1)
90.2(1)
2.125(3)
2.134(3)
2.103(4)
2.101(4)
1.988(6)
1.980(7)
2.220(2)
2.225(2)
99.1(1)
100.9(2)
80.1(1)
79.9(1)
96.9(1)
96.1(1)
90.8(2)
90.3(2)
Pd(2)-C
Pd(1)-P
Pd(2)-P
Pd(1)-F(1)-Pd(2) 100.6(1)
Pd(1)-F(2)-Pd(2) 101.6(1)
F(1)-Pd(1)-F(2)
F(1)-Pd(2)-F(2)
F(1)-Pd(1)-P
F(1)-Pd(2)-P
F(2)-Pd(1)-C
F(2)-Pd(2)-C
78.9(1)
78.9(1)
99.4(1)
99.4(1)
89.4(1)
89.4(1)
Figure 2. ORTEP drawing of 3 with thermal ellipsoids drawn at
the 50% probability level and all H atoms omitted for clarity.
due to the larger cone angle and lower basicity of o-Tol3P as
compared to Cy3P.32
In solution, complexes 3-5 exist as mixtures of syn- and
anti-isomers, much like [(R3P)2Pd2(Ph)2(µ-F)2] (R ) i-Pr, Cy),13
as well as 1 and 2.26 Like 1, 3-5 exhibited broadened NMR
signals, possibly due to syn-anti isomerization13 and hindered
rotation of the bulky phosphine ligands.26 Nonetheless, the syn
and anti structures were reliably established and isomer ratios
calculated at ca. 1.3-2:1 (see the Experimental Section).
The thermal decomposition of 3, 4, and 5 in dry benzene
under N2, in glass or Teflon, in the presence or in the absence
of BL, did not result in 19F NMR-observable quantities of
fluoroaromatics (eq 5). In all cases, only mixtures of P-F bond
Figure 3. ORTEP drawing of 4 with thermal ellipsoids drawn at
the 50% probability level and all H atoms omitted for clarity.
containing products were detected, mainly R2R′PF2 (see the
Experimental Section). This indicated that Pd-Ar/P-Ar′ ex-
change was occurring in parallel with the P-F bond forming
(30) See, for example: (a) Finger, G. C.; Kruse, C. W. J. Am. Chem.
Soc. 1956, 78, 6034. (b) Aksenov, V. V.; Vlasov, V. M.; Moryakina, I. M.;
Rodionov, P. P.; Fadeeva, V. P.; Chertok, V. S.; Yakobson, G. G. J. Fluorine
Chem. 1985, 28, 73.
(31) The starting iodo dimers were prepared by the literature methods:
Paul, F.; Patt, J.; Hartwig, J. F. Organometallics 1995, 14, 3030. Widen-
hoefer, R. A.; Zhong, H. A.; Buchwald, S. L. Organometallics 1996, 15,
2745. Widenhoefer, R. A.; Buchwald, S. L. Organometallics 1996, 15, 2755.
The p-tolyl and p-methoxy complexes are known compounds, whereas the
phenyl analogue has not been reported. Anal. Calcd for C54H52I2P2Pd2: C,
52.7; H, 4.3. Found: C, 52.7; H, 4.3.
(32) (a) However, characteristic32b C-H‚‚‚F contacts to the ligands that
are shorter than the sum of the van der Waals radii of F (1.47 Å) and H
(1.20 Å) are found in all four structures. The fluorine trans to C forms two
C-H‚‚‚F contacts of 2.13 Å (in 1) and 2.36 Å (in 4) to a methyl group of
each of the o-Tol3P ligands. In 3, the F trans to C interacts with one methyl
group of one of the phosphines (2.21 Å) and two methyls of the other (2.53
and 2.54 Å). The same pattern is observed for 5 (2.49, 2.56, and 2.58 Å).
In addition, 3 exhibits an intermolecular contact between the fluorine trans
to P and an aromatic CH of another molecule of 3. In 5, this F atom interacts
intramolecularly with one ortho H of each anisyl ligand (2.56 and 2.67 Å).
(b) Brammer, L.; Bruton, E. A.; Sherwood, P. Cryst. Growth Des. 2001, 1,
277.
Figure 4. ORTEP drawing of 5 with thermal ellipsoids drawn at
the 50% probability level and all H atoms omitted for clarity.
CH2Cl2 molecules H-bonded to the basic F centers,13 3-5 do
not display H-bonding interactions between the fluorine atoms
and cocrystallized molecules of dichloromethane. This is likely

5000 Organometallics, Vol. 26, No. 20, 2007
Grushin and Marshall
processes, as expected.11 We did not carry out a detailed analysis
of the reaction mixtures. As stated above, our goal was only to
determine whether or not congeners of 1 bearing non-Meisen-
heimer σ-aryls would undergo Ar-F reductive elimination.33
Scheme 2
Discussion
Metal-promoted halogen exchange reactions of nonactivated
(non-Meisenheimer; see above) haloarenes are rare,34,35 and
well-defined examples of aryl-halogen reductive elimination
from a transition metal center, particularly Pd(II), are even
scarcer.36,37 Therefore, the formation of p-nitrofluorobenzene
from 1 (eq 2)26 is an interesting observation.
Our results indicate that the nitro group in the “right” para
position of the σ-aryl ligand of 1 is critical for the Ar-F bond
formation. It is understood26,38 that much like, or even more so
than, Ar-OR reductive elimination from Pd(II),39 analogous
Ar-F reductive elimination is probably of considerable SNAr
character. Therefore conditions under which p-XC6H4-F bond
formation from 1/BL was observed (X ) NO2)26 may not be
suitable to give rise to p-XC6H4F (X ) H, Me, and MeO) from
3-5/BL (eq 5). However, no firm evidence has been presented
for Ar-F reductive elimination as the very path leading to
p-FC6H4NO2 (eq 2).26 While apparently facilitating Ar-F
reductive elimination, electron-withdrawing groups, especially
NO2, in the para position also open up the possibility of the
Meisenheimer chemistry to occur, a totally different, “hidden”
path that may lead to the same ArF product.
substitution reactions.28,29 Considering the ability of palladium
fluorides to release highly reactive fluoride,10 one can propose
a variety of transformations yielding some quantities of p-
nitrofluorobenzene from 1/BL (eq 2). For instance (Scheme 2),
a p-nitrophenyl phosphorus electrophile (e.g., a phosphonium
cation)11 formed from 1 might undergo SNAr with fluoride to
give p-NO2C6H4F. Although phosphonium salts were not
detected26 among the final products of the thermolysis reactions
of 1 and 2, their intermediacy in the reactions is manifested11
by the fact that Pd-Ar/P-Ar′ exchange did occur under the
conditions employed for the thermal decomposition of 1, 2,26
and 3-5 (this work). Nitrite would result from ipso substitu-
tion,28,29 which then might produce p-dinitrobenezene upon
Ar-NO2 reductive elimination or SNAr with an active p-
nitrophenyl phosphorus electrophile. Poorly solvated fluoride
has been shown4c to react with p-dinitrobenzene to afford
p-NO2C6H4F in >95% yield in less than 5 min at room
temperature. These are just a few plausible reaction paths other
than Ar-F reductive elimination that might give rise to
p-NO2C6H4F under the conditions employed for reaction 2.26
Therefore the line of reasoning apparently used26 to declare “net
Ar-F reductive elimination”40 being the mechanism for the
C-F bond formation in reaction 2 would be a mere logical
fallacy.41 The entire set of the experimental data reported26 can
neither prove nor disprove that Ar-F reductive elimination is
the mechanism of the C-F bond formation from 1.
A nitro group on the benzene ring is well-known to make
the latter susceptible to classical SNAr as well as ipso- and kine-
(33) The formation of a small quantity of fluorobenzene (ca. 1%) was
observed after a mixture of PhBr, [Me4N]+ F-, Pd2dba3 (3%), and BL (6%)
in toluene was stirred in a Teflon reactor at 110 °C (oil bath) for 2 days.
Considering the low yield of PhF, the high basicity of the fluoride source,
reaction conditions, and the fact that much finely dispersed Pd black was
quickly formed in this reaction, we hesitate to assert Ph-F reductive
elimination from a soluble PhPdF intermediate, although this path should
not be ruled out.
(34) (a) For a recent review, see: Berkenbusch, T. Sci. Synth. 2006, 25,
689. (b) Takagi, K.; Hayama, N.; Okamoto, T. Chem. Lett. 1978, 191. (c)
Takagi, K.; Hayama, N.; Inokawa, S. Bull. Chem. Soc. Jpn 1980, 53, 3691.
(d) Tsou, T. T.; Kochi, J. K. J. Org. Chem. 1980, 45, 1930. (e) Suzuki, H.;
Kondo, A.; Ogawa, T. Chem. Lett. 1985, 411. (f) Meyer, G.; Rollin, Y.;
Perichon, J. Tetrahedron Lett. 1986, 27, 3497. (g) Yang, S. H.; Li, C. S.;
Cheng, C. H. J. Org. Chem. 1987, 52, 691. (h) Bozell, J. J.; Vogt, C. E. J.
Am. Chem. Soc. 1988, 110, 2655. (i) Klapars, A.; Buchwald, S. L. J. Am.
Chem. Soc. 2002, 124, 14844. (j) Zanon, J.; Klapars, A.; Buchwald, S. L.
J. Am. Chem. Soc. 2003, 125, 2890. (k) Arvela, R. K.; Leadbeater, N. E.
Synlett 2003, 1145. (l) Toto, P.; Gesquiere, J.-C.; Cousaert, N.; Deprez, B.;
Willand, N. Tetrahedron Lett. 2006, 47, 4973.
(35) For Pd-promoted halide exchange in otherwise unreactive, pseudo-
aromatic 9-iodo-m-carborane, see: Marshall, W. J.; Young, R. J., Jr.;
Grushin, V. V. Organometallics 2001, 20, 523.
(36) Roy, A. H.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 123, 1232.
Stambuli, J. P.; Bu¨hl, M.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124,
9346. Roy, A. H.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 13944.
Roy, A. H.; Hartwig, J. F. Organometallics 2004, 23, 1533. Stambuli, J.
P.; Incarvito, C. D.; Bu¨hl, M.; Hartwig, J. F. J. Am. Chem. Soc. 2004, 126,
1184.
A comment is due on the mechanism of P-F bond formation
from various previously studied palladium fluoride com-
plexes,5,11,13,14,17 as well as 1,26 2,26 and 3-5 (this work). To
account for the observed products of the thermal decomposition
of [(Ph3P)2Pd(Ph)F] (eq 6), we have earlier proposed5,11
a
(37) (a) Recently, Vigalok et al.37b claimed reductive elimination of
Ar-Br in the reaction of some complexes of the type [(P-P)Pt(Ar)2] with
bromine. It is conceivable, however, that the instantaneous formation of
ArBr upon treatment of [(P-P)Pt(Ar)2] with Br237b might be due to classical
electrophilic cleavage of the Pt-Ar bond with bromine (SEAr), rather than
Br-Br oxidative addition, followed by Ar-Br reductive elimination. In
Vigalok’s report,37b we could not find any proof ruling out SEAr, nor any
unambiguous evidence for the alleged Ar-Br reductive elimination path
for the immediate reaction of [(P-P)Pt(Ar)2] with Br2. (b) Yahav-Levi,
A.; Goldberg, I.; Vigalok, A. J. Am. Chem. Soc. 2006, 128, 8710.
(38) Macgregor, S. A. Unpublished results.
(40) From ref 26: “This is the first indication, to our knowledge, of net
Ar-F reductive elimination operating to a quantifiable extent from a transition
metal aryl fluoride.” “The limited yet measurable success of P(C6H4-2-
Trip)(t-Bu)2 in enabling net Ar-F reductive elimination is therefore doubtless
due to its ability to destabilize sterically the fluoride-bridged Pd dimer
beyond the extent possible with P(t-Bu)3.” “However, use of Buchwald’s
L ) P(C6H4-2-Trip)(t-Bu)2 provided the additional steric pressure on the
[PdArL(µ-F)]2 core needed to enable formation of aryl-fluoride net reductive
elimination product in quantifiable yields (10%) in reactions with both 17
and 18 at 60° over 22 h.”
(41) The logical fallacy, in which it is claimed that a premise is true
only because it has not been proven false or that a premise is false only
because it has not been proven true. See, for example: Copi, I. M.; Cohen,
C. Introduction to Logic, 12th ed.; Prentice Hall: Upper Saddle River, NJ,
2004.
(39) Widenhoefer, R. A.; Zhong, H. A.; Buchwald, S. L. J. Am. Chem.
Soc. 1997, 119, 6787. Widenhoefer, R. A.; Buchwald, S. L. J. Am. Chem.
Soc. 1998, 120, 6504. Mann, G.; Shelby, Q.; Roy, A. H.; Hartwig, J. F.
Organometallics 2003, 22, 2775.

Ar-F ReductiVe Elimination from Palladium(II) ReVisited
Organometallics, Vol. 26, No. 20, 2007 5001
Scheme 3
or in the absence of Buchwald’s ligand (BL). Strong support
would have been provided for the alleged26 “net Ar-F reductive
elimination”40 from 1 if fluoroarenes had been produced from
3, 4, or 5, but they were not (eq 5). While being an important
observation, the formation of p-nitrofluorobenzene from 1/BL
(eq 2) may or may not have resulted from Ar-F reductive
elimination, and further experimental work is needed to
unambiguously elucidate the mechanism of this C-F bond
formation. This task is left to the authors of the report.26
Experimental Section
All chemicals were purchased from Aldrich and Strem chemical
companies and used as received. The solvents were thoroughly dried
using standard techniques and stored over freshly calcined molecular
sieves (4 Å) in a glovebox. All manipulations were carried out under
nitrogen in a glovebox, unless otherwise noted. NMR spectra were
obtained with a Bruker Avance DRX400 spectrometer. A Bruker-
CCD instrument was used for single-crystal X-ray diffraction
studies. Microanalyses were performed by Micro-Analysis, Inc.,
Wilmington, DE.
mechanism that involved the formation of [(Ph3P)Pd(Ph2PF)]
and/or isomeric [(Ph3P)Pd(Ph2P)(F)] upon Ph-Ph reductive
elimination from the corresponding Pd(Ph)2 species. The latter
[(o-Tol3P)2Pd2(Ph)2(µ-F)2] (3). Inside a glovebox, a 20 mL vial
was charged with [(o-Tol3P)2Pd2(Ph)2(µ-I)2]31 (0.30 g; 0.24 mmol),
AgF (0.12 g; 0.94 mmol), o-Tol3P (5 mg),42 and benzene (6 mL).
The vial was sealed under N2, brought out, and sonicated at 20 °C
for 6 h. After that, the vial was brought back in the glovebox and
unsealed. The mixture was diluted with CH2Cl2 (6 mL) and filtered
through Celite. The Celite was then washed with 4 mL of CH2Cl2.
After the combined filtrate and the washings were evaporated to
ca. 1.5 mL, ether (6 mL) was added and the solution was kept at
-25 °C for 3 h. The pale yellow needles of 3 were washed with a
small amount of cold ether, then hexanes, and dried under vacuum.
The yield of 3 was 0.195 g (79%). Anal. Calcd for C54H52F2P2Pd2:
C, 64.0; H, 5.2. Found: C, 64.0; H, 5.4. A solution of 3 for
characterization by NMR was obtained by dissolving the complex
in C6D6 at gentle heating to ca. 50 °C and allowing the solution to
cool to ambient temperature. 19F NMR (20 °C): δ -286 (br dt;
trans-JP-F ) ca. 160 Hz; JF-F ) ca. 50 Hz; 1F; syn-F trans to P),
-289 (br; 1.5 F; anti-F), -313 (br; 1F; syn-F cis to P). 31P NMR
(20 °C): δ 31.2 (br m). Under such conditions, the syn-anti ratio
was ca. 1.3:1 (19F NMR). Single crystals were grown by slow
diffusion of hexanes into a CH2Cl2 solution of 3.
[(o-Tol3P)2Pd2(p-Tol)2(µ-F)2] (4). Inside a glovebox, a 20 mL
vial was charged with [(o-Tol3P)2Pd2(p-Tol)2(µ-I)2]31 (0.304 g; 0.24
mmol), AgF (0.12 g; 0.94 mmol), o-Tol3P (5 mg),42 and benzene
(6 mL). The vial was sealed under N2, brought out, and sonicated
at 20 °C for 6 h. After that, the vial was brought back in the
glovebox and unsealed. The mixture was diluted with CH2Cl2 (6
mL) and filtered through Celite. The Celite was then washed with
5 mL of CH2Cl2. After the combined filtrate and the washings
were evaporated to ca. 1 mL, ether (5 mL) was added and the
solution was kept at -25 °C overnight. The pale yellow crystals
was believed to emerge from R-elimination of a phenyl on P or
a P-F reductive elimination/C-P oxidative addition sequence.
The recent studies18-21 of closely related [(Ph3P)3RhF] and its
facile rearrangement to cis-[(Ph3P)2Rh(Ph2PF)(Ph)] suggest that
palladium fluorides might undergo a similar transformation. In
particular, [(Ph3P)2Pd(Ph)F] would be converted to [(Ph3P)-
(Ph2PF)Pd(Ph)(Ph)], much like the Rh fluoride rearranges to
the Rh-Ph isomer (Scheme 3).
The two reactions shown in Scheme 3 share an important
feature: both are not decelerated by extra phosphine.11,18,19 This
is characteristic of a metallophosphorane19-21 mechanism, rather
than reversible P-C reductive elimination that governs the
slightly more facile, simultaneously occurring Pd-Ar/P-Ar
exchange reaction of [(Ph3P)2Pd(Ph)F].5,11
An updated version of the earlier proposed mechanism5,11
for reaction 6 is presented in Scheme 4. The first step involves
the aforementioned Pd-F/P-Ph to Pd-Ph/P-F rearrange-
ment (Scheme 3), followed by Ph-Ph reductive elimination.
The disproportionation reaction of the resulting [(Ph3P)Pd-
(Ph2P)(F)], possibly via its dimer, accounts for the formation
of all products observed, except for biphenyl, which is formed
in a preceding step (Scheme 4).
Conclusions
Unlike 1,26 its congeners bearing H, Me, or MeO substituents
in place of the nitro group do not produce observable (19F NMR)
quantities of any fluoroarenes upon thermolysis, in the presence
Scheme 4

5002 Organometallics, Vol. 26, No. 20, 2007
Grushin and Marshall
were washed with ether, then hexanes, and dried under vacuum.
The yield of 4‚0.5C6H6 was 0.23 g (88%). Anal. Calcd for
C59H59F2P2Pd2: C, 65.6; H, 5.5. Found: C, 65.3; H, 5.7. A solution
of 4 for characterization by NMR was obtained by dissolving the
complex in C6D6 at gentle heating to ca. 50 °C and allowing the
Thermal Decomposition of 3, 4, and 5 (a representative
procedure). Inside a glovebox, a Teflon tube was charged with 3,
4, or 5 (0.03 mmol), BL (0.12 mmol), benzene (1 mL), and a
Teflon-coated magnetic stirring bar. The tube was sealed and
brought out. After the reaction mixture was agitated at 80-85 °C
(oil bath) for 19 h, a black precipitate (likely Pd metal) was clearly
seen at the bottom of the tube. The liquid phase was analyzed by
19F and 31P NMR (unlocked) to indicate high conversion. Reso-
nances observed were as follows. For decomposed 3, 19F NMR: δ
-25 (d, JP-F ) 633 Hz, minor), -30 (d, JP-F ) 642 Hz, major),
-36 (d, JP-F ) 655 Hz, minor), -47 (d, JP-F ) 723 Hz, minor),
1
solution to cool to ambient temperature. H NMR (20 °C): δ 2.1
(br s; CH3), 2.5 (br s; CH3), 2.65 (br s; CH3), 6.6-8.1 (m; aromatic
H). 19F NMR (20 °C): δ -283 (br dt; trans-JP-F ) ca. 165 Hz;
JF-F ) ca. 55 Hz; 1F; syn-F trans to P), -286 (br; 1 F; anti-F),
-309 (br; 1F; syn-F cis to P). 31P NMR (20 °C): δ 34.1 (br m).
Under such conditions, the syn-anti ratio was 2:1 (19F NMR).
Single crystals were grown by slow diffusion of hexanes into a
CH2Cl2 solution of 3.
-85 (d, JP-F ) 1056 Hz, trace). 31P NMR: δ -39.8 (t, JP-F
)
642 Hz) from the major species observed in the 19F NMR spectrum
and other resonances. For decomposed 4, 19F NMR: δ -25 (d,
JP-F ) 633 Hz, minor), -30 (d, JP-F ) 643 Hz, trace), -31 (d,
JP-F ) 641 Hz, major), -37 (d, JP-F ) 653 Hz, minor), -47 (d,
JP-F ) 718 Hz, minor), -85 (d, JP-F ) 1056 Hz, trace). 31P
NMR: δ -40.3 (t, JP-F ) 641 Hz) from the major species observed
in the 19F NMR spectrum and other resonances. For decomposed
[(o-Tol3P)2Pd2(p-MeOC6H4)2(µ-F)2] (5). Inside a glovebox, a
20 mL vial was charged with [(o-Tol3P)2Pd2(p-MeOC6H4)2(µ-I)2]31
(0.308 g; 0.24 mmol), AgF (0.12 g; 0.94 mmol), o-Tol3P (5 mg),42
and benzene (6 mL). The vial was sealed under N2, brought out,
and sonicated at 20 °C for 5 h. After that, the vial was brought
back in the glovebox and unsealed. The mixture was diluted with
CH2Cl2 (6 mL) and filtered through Celite. The Celite was then
washed with 2 mL of CH2Cl2. After the combined filtrate and the
washings were evaporated, ether (2 mL) was added and the
precipitated yellow oil quickly solidified upon swirling. More ether
(10 mL) was added, and the mixture was kept at -25 °C overnight.
The yellow crystalline solid was separated, washed with hexanes,
and dried under vacuum. The yield of 5‚C6H6 was 0.17 g (62%).
Anal. Calcd for C62H62F2O2P2Pd2: C, 64.6; H, 5.4. Found: C, 64.2;
H, 5.7. Because the complex was found insufficiently soluble even
in warm benzene, its characterization by NMR was performed in
5, 19F NMR: δ -25 (d, JP-F ) 633 Hz, minor), -29 (d, JP-F
645 Hz, trace), -32 (d, JP-F ) 641 Hz, major), -39 (d, JP-F
650 Hz, minor), -43 (d, JP-F ) 660 Hz, trace), -47 (dm, JP-F
)
)
)
714 Hz, trace), and many low-intensity doublets at -70 to -90
ppm with JP-F ) ca. 450-960 Hz. 31P NMR: δ -41.4 (t, JP-F
)
641 Hz) from the major species observed in the 19F NMR spectrum
and other resonances. As seen from the NMR data, the major 19F
NMR-observable species was a mixed triaryldifluorophosphorane
R2R′PF2 in all three cases.
1
CD2Cl2. H NMR (CD2Cl2, 20 °C): δ 2.3 (br s; CH3), 2.5 (br s;
CH3), 3.6 (s; OCH3), 6.1-7.8 (m; aromatic H). 19F NMR (CD2Cl2,
20 °C): δ -284 (br; 1F; syn-F trans to P), -288 (br; 1.5 F; anti-
F), -312 (br; 1F; syn-F cis to P). 31P NMR (CD2Cl2, 20 °C): δ
32.5 (br m). Under such conditions, the syn-anti ratio was ca. 1.3:1
(19F NMR). Single crystals were grown by slow diffusion of hexanes
into a CH2Cl2 solution of 3.
Acknowledgment. We thank Dr. Stuart A. Macgregor for
discussions.
Supporting Information Available: Full details of the crystal-
lographic data (CIF) for 1 and new complexes 3-5. This material
is available free of charge via the Internet at http://pubs.acs.org.
(42) As recommended26 for the synthesis of 1.
OM700469K