Catalytic Carbon-Fluorine Bond Activation with Monocoordinated Nickel-Carbene Complexes: Reduction of Fluoroarenes

Article abstract of DOI:10.1002/adsc.200390036

Monocoordinate nickel/N-heterocyclic carbene complexes reveal unexpected reactivity towards aryl fluorides. Defluorination reactions were efficiently performed with a β-hydrogen-containing alkoxide (3 equiv.) in the presence of 3 mol % of [1:1] Ni(0)/IMes · HCl catalyst (IMes = 1,3-dimesitylimidazol-2-ylidene).

Full text of DOI:10.1002/adsc.200390036

COMMUNICATIONS
Catalytic Carbon-Fluorine Bond Activation with
Monocoordinated Nickel-Carbene Complexes: Reduction of
Fluoroarenes
¬
»
Sebastien Kuhl, Raphael Schneider, Yves Fort*
¡
¬
¬
¬
¬
Laboratoire de Synthese Organique et Reactivite, UMR CNRS-UHP 7565, Universite Henri Poincare, Nancy I, BP 239,
54506 Vandoeuvre les Nancy Cedex, France
Fax: ( ‡ 33)-38-3684785, e-mail: Yves.Fort@sor.uhp-nancy.fr
Received: October 11, 2002; Accepted: December 12, 2002
of C F bonds by reaction at transition metal centres,
Abstract: Monocoordinate nickel/N-heterocyclic
only few defluorination methods are currently available.
carbene complexes reveal unexpected reactivity
Heterogeneous rhodium-catalysed hydrogenolyses re-
towards aryl fluorides. Defluorination reactions
quire high temperatures and/or pressure.[10] An im-
were efficiently performed with a b-hydrogen-con-
proved process was recently described using rhodium
taining alkoxide (3 equiv.) in the presence of
complexes tethered on a silica-supported palladium
3 mol % of [1:1] Ni(0)/IMes ¥ HCl catalyst (IMes ˆ
heterogeneous catalyst.[11] Homogeneous rhodium[12] or
zirconium[13] catalysts are also efficient as is also the
decamethylferrocene-photoinduced defluorination of
perfluoroalkanes.[14] Catalytic reductions of fluorocar-
bons can finally be achieved using YbCp2(dme)[15] and
Group 4 metallocenes such as Cp2TiF2 or CpZrCl2
associated to terminal reductants such as magnesium
or aluminium metals.[16] However, these methods often
1,3-dimesitylimidazol-2-ylidene).
Keywords: catalysis; fluoroarenes; nickel(0)/N-he-
terocyclic carbene complex; reduction
The activation of C F bonds of fluoroaromatic com- yielded partial defluorination and new protocols are still
pounds by reaction at transition metal centres is an of interest.
exciting challenge in organometallic chemistry and
From our part, we have embarked upon a program of
catalyst development due to the remarkable thermal work which has been directed toward the preparation of
and chemical stability of organofluorine compounds. a range of Ni(0)/N-heterocyclic carbene (NHC) com-
Although the stoichiometric activation and functional- plexes and examination of their reactivity. We have
isation of C F bonds by transition metal complexes has recently demonstrated that Ni(0) clusters, in situ gen-
been shown to proceed under mild conditions,[1] cata- erated by reduction of Ni(acac)2 with alkoxide-activated
lytic transformations of C F bonds are still poorly sodium hydride, associated to strong electron-donating
developed.
ligands such as NHCs[17] can be used to mediate
Murai studied Si F exchange reactions between aminations of aryl chlorides.[18] We have also prepared
fluorobenzenes and Me3SiSiMe3 catalysed by complex 1 via a solution-phase method and we demon-
[Rh(cod)2]BF4.[2] Kumada[3] and Herrmann[4] have dem- strated that 1, combined with i-PrONa was an efficient
onstrated the catalytic cross-coupling of aryl fluorides catalyst for the dehalogenation of functionalised aryl
with Grignard reagents mediated respectively by chlorides, bromides, iodides and polyhalogenated hy-
Ni(dmpe)Cl2
[dmpe ˆ 1,2-bis(dimethylphosphanyl)- drocarbons.[19] With aryl fluorides, the procedure was,
ethane] and Ni(cod)2 associated to an N-heterocyclic however, less efficient and only partial defluorination,
carbene (NHC), 1,3-di(2,6-di-isopropylphenyl)imida- not exceeding 40%, was observed.
zolin-2-ylidene. The nickel-catalysed activation of the
C F bond of 2-fluoropyridines and pyrimidines has also
been reported.[5 8]
R
R
R
Dehalogenation of fluorocarbons is also an important
process since the chemical inertness of the C F bond
translates into environmental persistence and such
compounds are quite difficult to degrade. Fluorine-
hydrogen exchange is also of interest in multistep
syntheses in which the fluorine atom is used as a
N
N
N
Ni
Ni
N
N
N
R
R
R
2
1
R = 2,4,6-Me3C6H2
directing group.[9] Despite recent advances in activation Figure 1. Di- and monocoordinated nickel complexes 1 and 2.
Adv. Synth. Catal. 2003, 345, No. 3
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim 1615-4150/03/34503-341 344 $ 17.50+.50/0
341

¬
Sebastien Kuhl et al.
COMMUNICATIONS
Recent reports of Hartwig,[20] Nolan[21] and Caddick[22]
on cross-coupling reactions of amines with aryl halides
catalysed by Pd/NHC complexes suggested that initially
formed dicoordinated Pd/NHC species dissociate to
form mono(carbene) palladium complexes which un-
dergo oxidative addition of the aryl halide. Since
activation of C F bonds requires highly reactive and
electron-rich metal centres, we wondered if complex 2,
in situ generated using Ni(0) and the imidazolium salt 3
in a 1:1 ratio, could yield a catalyst with enhanced
performances in defluorination reactions. In this com-
munication, we wish to report the first example of a
reduction of the C F bond in monofluoroarenes using a
catalytic 1:1 combination of Ni(0)/NHC associated to a
b-hydrogen-containing alkoxide.
N
N
N
N
, Cl
, Cl
R
R
R
R
3, IMes·HCl : R = 2,4,6-Me3C6H2
4, IPr·HCl : R = 2,6-i-Pr2C6H3
5, R = (CH2)2Ni-Pr2
6, SIMes·HCl : R = 2,4,6-Me3C6H2
7, SIPr·HCl : R = R = 2,6-i-Pr2C6H3
N
N
N
N
N
N
N
2 Br
N
N
2 Br
Me
Me
9
8
Figure 2. Carbene precursors.
The reduction of 1-fluoronaphthalene performed with
complex 2, i-PrONa in refluxing THF was undertaken
As expected, dehalogenation run without ligand
first. Compared to the modest yield (36%) observed with resulted in no detectable reaction product (entry 1).
the two-coordinate nickel catalyst 1, we were pleased to The imidazolium IMes ¥ HCl 3 was found to be the most
find out that mono-(carbene)-complex 2 afforded an effective NHC precursor leading to naphthalene in 92%
improved (57%) defluorination yield.
yield (entry 2). The NHC generated from the sterically
A study aimed at determining the optimum Ni(0)/ demanding IPr¥ HCl 4 can also be used (entry 3). All
NHC system and the reaction conditions for the other imidazolium salts that were investigated led to the
catalytic dehalogenation of aryl fluorides was then formation of moderately active catalysts (entries 4 8).
undertaken.
Table 2 shows the results of the defluorination of
Using complex 2 as defluorination catalyst, either various aromatic fluorides using the optimal conditions
dioxane or THFappears to be a suitable solvent whereas previously determined. Fluoronaphthalenes were re-
toluene led to partial catalyst decomposition. Reaction duced to naphthalene within 3 h (entries 1 and 2). Using
rates are however lower in dioxane compared to THF. 3- or 4-fluorotoluene (entries 3 and 4), the reaction was
Curiously, the sodium alkoxide Et2CHONa gave sig- much slower and afforded the reduction product in a
nificantly greater rates of reduction than i-PrONa. modest 30% yield. In the series of fluoroanisoles
Other alkoxides (e.g., MeONa) were ineffective.
(entries 5 7), a considerable dependency of the deha-
A survey of the catalytic efficiency of various imida- logenation yield on the position of the methoxy group
zolium salts was finally undertaken. Results are sum- was observed. 2-Fluoroanisole was quantitatively con-
marised in Table 1.
verted into anisole after 2.5 h. GC analysis of the
reaction mixture revealed that the rate of formation of
anisole was slower with 3- and, especially, 4-fluoroani-
sole. Only 53% and 28% yields, respectively, of the
reduction product were observed after 2.5 h. This effect
has already been observed in carbon-nitrogen couplings
mediated by Ni(0)/NHC complexes[18b] and might result
from a coordination of the nickel centre at oxygen atom
of 2-fluoroanisole. It is noteworthy that the reduction of
3-fluoro-N-methylaniline required an additional equi-
molar amount of alkoxide due to the partial deproto-
nation of the amine (entry 8). Dehalogenation was
finally not observed with 2-fluoropyridine (entry 9).
Only 1-ethylpropyl (2-pyridyl) ether was produced in
97% yield by a classic SNAr reaction.
At the end of the dehalogenation, pentan-3-one was
clearly identified in the reaction products by GC. More-
over, clean formation of 1-deuteronaphthalene was ob-
served when the dehalogenation was conducted with the
deuterium-labelled sodium alkoxide Et2CDONa.
Mercury tests also confirmed that defluorination
reactions performed with the Ni(0)/IMes complex are
homogeneously catalysed.[23] Two types of experiments
Table 1. Assessment of carbene precursors in the Ni(0)-
catalysed defluorination of 1-fluoronaphthalene.[a]
F
Ni(0) (3 mol %)
Carbene (3 mol %)
Et2CHONa (3 equiv.)
Dioxane, 100 °C, 3 h
Entry
Imidazolium salt
Yield [%][b]
1
2
3
4
5
6
7
8
À
0
92
87
51
68
53
7
IMes ¥ HCl, 3
IPr¥ HCl, 4
5
SIMes ¥ HCl, 6
SIPr¥ HCl, 7
8[c]
9[c]
8
[a]
Reaction conditions: 1-fluoronaphthalene (10 mmol),
Et2CHONa (30 mmol), Ni(0) (0.3 mmol), imidazolium
salt (0.3 mmol), dioxane, 100 8C.
Determined by H NMR on the crude reaction product.
1.5 mol % imidazolium salt.
[b]
[c]
1
342
Adv. Synth. Catal. 2003, 345, 341 344

Catalytic Carbon-Fluorine Bond Activation with Ni-Carbene Complexes
COMMUNICATIONS
Table 2. Results of the Ni(0)/IMes ¥ HCl defluorination reactions.[a]
Entry
Aromatic fluoride
Time [h][b]
Yield [%][c]
Recovered aromatic fluoride [%][b]
1
2
3
4
5
6
7
8
9
1-Fluoronaphthalene
2-Fluoronaphthalene
3-Fluorotoluene
4-Fluorotoluene
2-Fluoroanisole
3-Fluoroanisole
4-Fluoroanisole
3-Fluoro-N-methylaniline
2-Fluoropyridine
3-Fluoropyridine
3
3
92
100
30
30
100
75
5
0
70
70
0
19
54
9
15
15
2.5
3.5
12
23
3.5
2
46
89[d]
0[e]
100
0
0
10
[a]
The typical reactions conditions: aromatic fluoride (10 mmol), Et2CHONa (30 mmol), Ni(0) (0.3 mmol), IMes ¥ HCl
(0.3 mmol), dioxane, 100 8C.
Determined by GC.
Yield was determined by GC.
Et2CHONa (40 mmol).
[b]
[c]
[d]
[e]
1-Ethylpropyl (2-pyridyl) ether was produced in quantitative yield by an SNAr reaction.
were performed to evaluate the effect of Hg on the ligand (L ˆ IMes) to the metal centre. Oxidative addi-
catalyst. First, the reduction of 3-fluoroanisole was tion of the aryl fluoride to the Ni(0) complex is followed
stopped after 1.5 h (45% conversion) and mercury by Et2CHONa attack at nickel and displacement of the
(321 equiv.) was added. No further catalytic activity fluorine atom. Finally, reductive elimination of the
was observed after 2 hours heating at 100 8C (46% arene from an intermediate nickel-hydride complex
conversion). Second, when Hg was added to a freshly obtained by b-elimination regenerates the Ni(0)-car-
prepared Ni(0)/IMes complex just before 3-fluoroani- bene catalyst.
sole, a complete poisoning of the catalyst is observed
(<4% conversion).
In conclusion, we have shown that the monocoordi-
nate nickel-carbene complex 2 efficiently promotes the
Taking into account these observations, a general activation of aromatic C F bonds and demonstrated
catalytic cycle for the defluorination of aryl fluorides in that this catalyst, associated to Et2CHONa, can be used
the presence of the Ni(0) complex 2 is presented in in defluorination reactions. These findings may be useful
Scheme 1. The first step involves the formation of Ni(0) for the design of new nickel-carbene catalysts for
by reduction of Ni(acac)2 with Et2CHONa activated organic synthesis.
sodium hydride followed by coordination of the carbene
Ni(acac)2 + L
Experimental Section
2 NaH/Et2CHONa
H2 + 2 AcacNa
General
LNi(0)
ArH
ArF
All manipulations were carried out using standard Schlenk
techniques under an atmosphere of nitrogen. Gas chromato-
graphic analyses were performed on a Shimadzu GC-17
capillary gas chromatograph fitted with an ™Optima 5∫ column
(22 m  0.25 mm, ID  0.25 mm). All quantifications of reac-
tion constituents were achieved by gas chromatography using
a known quantity of dodecane as reference standard. The 1H
and 13C NMR spectra were recorded on a Bruker AM 400
spectrometer, respectively at 400.13 and 100.40 MHz using
CDCl3 as solvent. Aryl fluorides, arenes and nickel(II)
acetylacetonate were purchased from Acros and used as
received. Sodium hydride (65% in mineral oil) was purchased
from Fluka and used after two washings with THF under
nitrogen. Dioxane was distilled over sodium/benzophenone
and stored over sodium wires. Pentan-3-ol was purchased from
Acros and distilled over sodium before use. All imidazolium
salts were synthesised according to the literature procedures.[24]
H
F
LNi(II)
LNi(II)
Ar
Ar
Et
Et
Et
Et2CHONa
NaF
O
H
Et
O
LNi(II)
Ar
Scheme 1. Proposed mechanism for the defluorination reac-
tion.
Adv. Synth. Catal. 2003, 345, 341 344
343

¬
Sebastien Kuhl et al.
COMMUNICATIONS
[12] a) M. Aizenberg, D. Milstein, Science 1994, 265, 359;
b) M. Aizenberg, D. Milstein, J. Am. Chem. Soc. 1995,
117, 8674.
[13] J. L. Kiplinger, T. G. Richmond, Chem. Commun. 1996,
1115.
[14] a) J. Burdeniuc, R. H. Crabtree, J. Am. Chem. Soc. 1996,
118, 2525; b) J. Burdeniuc, R. H. Crabtree, Organo-
metallics 1998, 17, 1582.
[15] G. B. Deacon, C. M. Forsyth, J. Sun, Tetrahedron Lett.
1994, 35, 1095.
Typical Procedure for the Catalytic Reduction of Aryl
Fluorides
A Schlenk tube was loaded with degreased NaH (30.6 mmol),
Ni(acac)2 (0.3 mmol), IMes ¥ HCl (0.3 mmol) and 8 mL of
dioxane and the mixture was heated to 100 8C. A solution of
Et2CHOH (30 mmol) in 2 mL dioxane was then added
dropwise and the mixture was further heated for 1 h. A
solution of aryl fluoride (10 mmol) in dioxane (3 mL) was then
added. The progress of reactions was monitored by GC. All
quantifications of reaction constituents were achieved by GC.
[16] a) J. L. Kiplinger, T. G. Richmond, J. Am. Chem. Soc.
1996, 118, 1805; b) B. L. Edelbach, B. M. Kraft, W. D.
Jones, J. Am. Chem. Soc. 1999, 121, 10327; c) B. M. Kraft,
R. J. Lachicotte, W. D. Jones, J. Am. Chem. Soc. 2001,
123, 10973.
[17] W. A. Herrmann, Angew. Chem. Int. Ed. 2002, 41, 1290.
[18] a) B. Gradel, E. Brenner, R. Schneider, Y. Fort, Tetrahe-
dron Lett. 2001, 42, 5689; b) C. Desmarets, R. Schneider,
Y. Fort, J. Org. Chem. 2002, 67, 3029.
[19] C. Desmarets, S. Kuhl, R. Schneider, Y. Fort, Organo-
metallics 2002, 21, 1554.
[20] S. R. Stauffer, S. Lee, J. P. Stambuli, S. I. Hauck, J. F.
Hartwig, Org. Lett. 2000, 2, 1423.
References
[1] a) G. Guillaumet, L. Mordenti, P. Caubere, J. Organo-
met. Chem. 1975, 92, 43; b) J. Burdeniuc, B. Jedlicka,
R. H. Crabtree, Chem. Ber. 1997, 130, 145.
[2] Y. Ishii, N. Chatani, S. Yorimitsu, S. Murai, Chem. Lett.
1998, 157.
[3] Y. Kiso, K. Tamao, M. Kumada, J. Organomet. Chem.
1973, 50, C12-C14.
[4] V. P. W. Bˆhm, C. W. K. Gstˆttmayr, T. Weskamp, W. A.
Herrmann, Angew. Chem. Int. Ed. 2001, 40, 3387.
[5] L. Cronin, C. L. Higgitt, R. Karch, R. N. Perutz, Organo-
metallics 1997, 16, 4920.
[6] T. Braun, S. P. Foxon, R. N. Perutz, P. H. Walton, Angew.
Chem. Int. Ed. 1999, 38, 3326.
[7] T. Braun, R. N. Perutz, M. I. Sladek, Chem. Commun.
2001, 2254.
[8] M. I. Sladek, T. Braun, B. Neumann, H.-G. Stammler, J.
Chem. Soc. Dalton Trans, 2002, 297.
[9] See, for example: M. Schlosser, Eur. J. Org. Chem. 2001,
3975.
[21] G. A. Grasa, M. S. Viciu, J. Huang, S. P. Nolan, J. Org.
Chem. 2001, 66, 7729.
[22] L. R. Titcomb, S. Caddick, F. G. N. Cloke, D. J. Wilson,
D. Mc Kerrecher, Chem. Commun. 2001, 1388.
[23] a) D. R. Anton, R. H. Crabtree, Organometallics 1983, 2,
855; b) K. S. Weddle, J. D. Aiken III, R. G. Finke, J. Am.
Chem. Soc. 1998, 120, 5653.
[24] a) A. J. Arduengo III, U. S. Patent 5,077,414, 1991;
b) A. J. Arduengo III, R. Krafczyk, R. Schmutzler,
Tetrahedron 1999, 55, 14523; c) A. M. Maggill, D. S.
McGuiness, K. J. Cavell, G. J. P. Britovsek, V. C. Gibson,
A. J. P. White, D. J. Williams, A. H. White, B. W. Skelton,
J. Organomet. Chem. 2001, 617-618, 546.
[10] R. J. Young Jr, V. V. Grushin, Organometallics 1999, 18,
294.
[11] H. Yang, H. Gao, R .J. Angelici, Organometallics 1999,
18, 2285.
344
Adv. Synth. Catal. 2003, 345, 341 344