Influence of the transmetalating agent in difficult coupling reactions: Control in the selectivity of C-F bond activation by Ni(0) complexes in the presence of AlMe3

Article abstract of DOI:10.1021/acs.organomet.7b00165

Although Ni(PEt₃)₄ does not react with di- or trifluoroarenes at room temperature, upon the addition of aluminum hydrocarbons such as AlMe₃ an immediate reaction occurs, to give AlMe₂F and Ni(II) complexes from C-F bond activation and transmetalation. The influence of additional Lewis basic compounds, such as pyridine, on selectivity in these systems provides insight into how selectivity in a cross-coupling reaction is controlled by the transmetalating agent and how the oxidative addition and transmetalation steps are not necessarily distinct.

Full text of DOI:10.1021/acs.organomet.7b00165

Article
pubs.acs.org/Organometallics
Influence of the Transmetalating Agent in Difficult Coupling
Reactions: Control in the Selectivity of C−F Bond Activation by Ni(0)
Complexes in the Presence of AlMe3
Drake A. Baird, S. Jamal, and Samuel A. Johnson*
Department of Chemistry and Biochemistry, University of Windsor, 401 Sunset Avenue, Windsor, Ontario N9B 3P4, Canada
S
* Supporting Information
ABSTRACT: Although Ni(PEt3)4 does not react with di- or
trifluoroarenes at room temperature, upon the addition of aluminum
hydrocarbons such as AlMe3 an immediate reaction occurs, to give
AlMe2F and Ni(II) complexes from C−F bond activation and
transmetalation. The influence of additional Lewis basic compounds,
such as pyridine, on selectivity in these systems provides insight into
how selectivity in a cross-coupling reaction is controlled by the
transmetalating agent and how the oxidative addition and trans-
metalation steps are not necessarily distinct.
Cross-coupling of C−F bonds using Ni catalysts has seen
considerable recent progress.4d,11 Although cross-coupling
involving fluorobenzene was demonstrated by Kumada as
early as 1972,12 yields and selectivity were poor. A 2001 report
showed that N-heterocyclic carbene complexes of Ni act as
catalysts for C−F functionalization of fluorobenzene and
supported the hypothesis that a requirement for the activation
of C−F bonds in weakly electrophilic substrates was the use of
a highly electron rich metal center;11g however, since then
numerous examples of Ni-catalyzed C−F bond cross-coupling
reactions have emerged that feature only modestly electron
donating ligands, with diphenylphosphines13 and even
triphenylphosphine11a utilized as the supporting ligands.
These ligands are not featured in the stoichiometric oxidative
addition chemistry of Ni with C−F bonds, raising questions as
to the exact nature of the C−F bond cleavage step. The
mechanism of C−F cleavage has been controversial even in
stoichiometric oxidative addition transformations, with exper-
imental evidence supporting traditional concerted and radical
mechanisms for different substrates, whereas phosphine-
assisted mechanisms have been proposed from a computational
study.7a,b
The catalytic functionalization of di-, tri-, and tetrafluor-
obenzenes presents problems beyond that of the facile, but
poorly understood, C−F cross-coupling of hexa- and
pentafluorobenzenes. Most attempts at using di- and
trifluorobenzenes as substrates yield a mixture of products
from the unselective substitution of the multiple C−F bonds,
thus yielding undesired di- and trisubstitution products.9,11d,d,14
The di- and trisubstitution products have been attributed to π-
bound intermediates after the C−C bond forming reduction
INTRODUCTION
■
Fluorinated organics have unique properties which have led to
their use in a wide variety of applications. Examples include
pharmaceuticals,1 Lewis acid catalysts,2 and materials applica-
tions such as n-type semiconductors.3 The low cost and
availability of simple partially fluorinated arene precursors, such
as the di-, tri-, and tetrafluorobenzenes, renders these as
attractive starting materials to partially fluorinated organics. A
combination of both C−H and C−F activation could provide a
versatile synthetic pathway to convert these substrates to
complex organics with a wide array of fluorination patterns;
however, their use mandates the development of selectivity in
these difficult bond activation reactions.4
The continued interest in the use of nickel in C−H and C−F
catalytic functionalization stems from the availability and low
cost of nickel relative to its heavier congeners, which are more
commonly used in catalysis.5 The ability of Ni to facilitate C−F
bond oxidative addition was first described in a 1977 report
which showed that Ni(PEt3)4 undergoes the oxidative addition
of C6F6 to yield trans-(Et3P)2NiF(C6F5).6 In general, the ability
of substrates to undergo C−F bond oxidative addition with
Ni(PEt3)4 mirrors their propensity to react with nucleophiles.6
Whereas C6F6 reacts slowly with Ni(PEt3)4 under ambient
conditions, the more electrophilic nitrogen-containing substrate
pentafluoropyridine reacts rapidly.7 With less electrophilic
substrates such as pentafluorobenzene, slightly more reactive
sources of the (Et3P)2Ni moiety such as (Et3P)2Ni(η2-C14H10)
are required for a clean conversion.8 In contrast to these
electrophilic substrates, the di- and trifluorobenzenes do not
undergo C−F activation at any appreciable rate with sources of
the (Et3P)2Ni moiety. The activation of less activated arenes
such as the tetra-, tri-, and difluorobenzenes has required the
use of more electron rich donors, such as N-heterocyclic
carbenes,9 or electron-rich nitrogen donors.10
Received: March 3, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
elimination step. These π-bound intermediates are capable of
ring whizzing8,15 and successive C−F bond activations and
functionalizations of the substrate. An example of cross-
coupling with 1,2-difluorobenzene is shown in Scheme 1; the
reaction yields as much disubstituted product as monosub-
stituted product.11b
adducts that undergo C−F activation after the addition of
Lewis acids as weak as LiI.16,17
The addition of 1 equiv of AlMe3 to solutions of Ni(PEt3)4
and 1,2-, 1,3-, or 1,4-difluorobenzene caused an immediate
color change from purple to yellow. Analysis of each crude
reaction mixture by 1H, 31P{1H}, 19F, and 13C{1H} NMR
confirmed instantaneous complete conversion of the reagents
to the three isomers of trans-(Et3P)2Ni(C6FH4)(Me) (112,13,14),
as shown in Scheme 2. The production of AlMe2F was
Scheme 1. (A) Example of C−F Bond Activation of 1,2-
Difluorobenzene and (B) Mono- vs Disubstitution of the
Difluorobenzenes Using a More Selective Catalytic System13
18
1
confirmed from H and 19F NMR spectra.
Scheme 2. Activation of the Difluorobenzenes by Ni(PEt3)4
in the Presence of AlMe3
a
b
Percent yield based on 19F NMR spectra. Isolated yield.
A solution to this problem is the introduction of directing
groups, although this requires functionalized substrates and
limits the range of fluorination patterns accessible.4c,11a,c Very
recently ligand designs involving chelating phosphines and
pendant alkoxide donors have emerged that are capable of
monofunctionalization of substrates such as the di- and
trifluorobenzenes, as shown in Scheme 1B.13,14b These systems
have been suggested to work by the binding of the
transmetalating agent to the oxygen donor. Despite this
breakthrough, very little is known about the fundamental
mechanistic issues involved in the design of successful catalytic
systems for the functionalization of partially fluorinated
aromatics. Even in the more selective system shown in Scheme
1B, varying amounts of disubstitution are observed, depending
upon the substrate.
Similarly to the difluorobenzenes, the reaction of the
trifluorobenzenes with Ni(PEt3)4 and AlMe3 reacted to give
isomers of trans-(Et3P)2Ni(C6F2H3) (Me) (1123a,123b,124,135), as
shown in Scheme 3. The activation of 1,2,3-trifluorobenzene
proceeded with good regioselectivity; the 19F NMR spectra of
Scheme 3. Activation of the Trifluorobenzenes by Ni(PEt3)4
in the Presence of AlMe3
This paper describes a system that provides a significant
acceleration of the C−F activation step by using a Lewis acidic
Al-based transmetalating agent. This system provides funda-
mental mechanistic insight into the key issues that need to be
addressed to advance the design of catalysts for the selective
mono-functionalization of polyfluoroarenes.
RESULTS AND DISCUSSION
■
The attempted reaction of Ni(PEt3)4 with the di- and
trifluorobenzenes at room temperature does not provide
conversion to C−F bond activation products, unlike the
reactions with highly electron deficient substrates such as
hexafluorobenzene and pentafluoropyridine.6,7,15a The addition
of the Lewis acids DIBAL, Ph3B, Ph2Zn, SnCl2, and FeCl2 failed
to facilitate activation. This is in contrast to fluoroalkene
substrates such as tetrafluoroethylene5b and hexafluoroprop-
ene,4d which have been reported to form nickel diphosphine
a
b
c
Percent yield based on 19F NMR spectra. Isolated yield. Selectivity
between 1-site and 2-site activation.
B
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4. Activation of the Tetrafluorobenzenes by Ni(PEt3)4 in the Presence of AlMe3
the reaction mixtures showed a 94:6 mix of 1123a from
activation at the 1-site and the isomer 1123b from activation at
the 2-site, the latter of which could not be isolated. The
assignment of 1123b is based on its NMR spectra, which features
spectroscopy to be the reductive elimination products
1245,1235,1234a,1234b, as well as the second C−F bond activation
2
products 31245,1235,1234a,1234b, as shown in Scheme 4. The second
activation products 31245,1235,1234a,1234b were observed even in
the presence of excess tetrafluorobenzene, suggesting that these
products arise from reductive elimination from
11245,1235,1234a,1234b rapid ring whizzing, and a second C−F
bond activation. Reaction conditions that provide better yields
of 11245,1235,1234a,1234b, without contamination by the second
activation byproducts, are discussed later using alternate
conditions.
a
19F shift similar to that observed in the C−F bond oxidative
addition of 1,2,3-trifluorobenzene by a Ni carbene complex.9
The carbene complex was reported to be slightly less selective,
with an 85:15 conversion to C−F activation products at the 1-
and 2-sites.9 The activation of 1,2,4-tetrafluorobenzene with
Ni(PEt3)4 and AlMe3 occurred at the 2-site selectively to give
1124, and 1,3,5-trifluorobenzene reacted to give 1135, as shown
Select 19F, 31P{1H}, 1H, and 31P NMR parameters for
in Scheme 3.
The difluorobenzene activation products 112,13,14 and
trifluorobenzene activation products 1123a,124,135 were isolated
by crystallization from pentane after removal of the AlMe2F
byproduct by filtration through silica. The modest isolated
yields after recrystallization from pentane (63−77%) reflect the
high solubility of these compounds, not the selectivity of the
reactions; the NMR spectra of the crude reaction mixtures
indicate quantitative conversion.
The activation of the more reactive tetrafluorobenzenes by
Ni(PEt3)4 in the presence of AlMe3 produced the compounds
11245,1235,1234a,1234b, but in modest NMR yields. The activation
of 1,2,3,4-tetrafluorobenzene occurred with little selectivity,
with activation at the 1-site vs 2-site producing 11234a and 11234b
in a 46:54 ratio. Isolation of the tetrafluorobenzene activation
products was complicated by the presence of significant
impurities. These impurities were assigned by 19F NMR
112,13,14, 1123a,123b,124,135, and 11245,1235,1234a,1234b are provided in
Table 1. The 19F NMR shifts associated with the aryl
substituents are similar to those of the structurally related C−
F activation products.5a The H NMR gives a triplet for the
1
3
Ni−Me substituent in the range of δ −0.9 to −0.6, with a JPH
coupling of about 9.0 Hz. Complexes 113, 114, and 2135, which
lack o-F substituents, feature 13C{1H} NMR shifts for the Ni−
CH3 group from δ −10.4 to −10.6, whereas the remaining
complexes feature shifts from δ −9.8 to −9.9. The 31P{1H}
NMR shifts for the species without o-F substituents are all near
δ 18.8, whereas complexes with o-F substituents feature shifts
from δ 19.6 to 20.2.
Single crystals suitable for characterization by X-ray
crystallography were obtained for all the products except the
minor regioisomer 1123b and the mixture of isomers 11234a,1234b
.
All featured nearly square planar geometries at the Ni center
C
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
with trans-disposed phosphines. Crystallographic details are
provided in Table 2 and in the Supporting Information.
Second Activation Products from Ring Whizzing. The
mechanism by which AlMe3 promotes C−F bond activation at
room temperature is of interest both for fundamental reasons
and for the development of new C−F coupling reactions under
mild conditions. The addition of varying concentrations of
AlMe3 to 1,4-difluorobenzene in the absence of Ni(PEt3)4
yielded no notable change in the 19F NMR spectrum, suggestive
of a negligible Lewis acid/Lewis base interaction between the
two reagents. Additionally, this mixture did not undergo
reaction without the addition of Ni(PEt3)4. The addition of
AlMe3 to Ni(PEt3)4 in the absence of a fluorinated substrate led
to rapid decomposition at room temperature, with the
elimination of a black powder and free PEt3. This mixture
did not react with added 1,4-difluorobenzene. These experi-
ments cannot rule out two different mechanistic paradigms: the
first where AlMe3 and Ni(PEt3)4 react to create an unstable
metal complex capable of rapid C−F activation, and the second
more traditional mechanism where an equilibrium π adduct8,15a
of the type (Et3P)2Ni(η2-C6H4F2) interacts with AlMe3 to
undergo oxidative addition and transmetalation, albeit perhaps
in a single step rather than the more traditional two-step
mechanism commonly proposed for these reaction steps.
Attempts to perform the reaction between 1,4-difluoroben-
zene and AlMe3 in the presence of catalytic Ni(PEt3)4 provides
some insight and supports π-bound complexes as intermediates
in these C−F bond activations. For example, if the reaction
mixture of 1,4-difluorobenzene with Ni(PEt3)4 and AlMe3 that
generates 114 is not immediately filtered through alumina to
remove AlMe2F or any residual AlMe3, the reaction continues
to produce trans-(Et3P)2Ni(4-MeC6H4)(Me) (314). The
reductive elimination from 114 most likely produces the π-
bound intermediate (Et3P)2Ni(η2-C6H4F-4-Me) (414), which
then undergoes ring whizzing and a rapid second C−F
activation prior to the dissociation of 4-fluorotoluene (214).
This is suggested by the fact that 314 accumulates prior to the
observation of a significant amount of 4-fluorotoluene (214) in
solution, and 314 is produced even in the presence of an excess
of 1,4-difluorobenzene. Attempts to isolate 314 have failed due
to its conversion of p-xylene via reductive elimination at a rate
comparable to the rate of its production; thus, its character-
ization is based on its 1H and 31P{1H} NMR spectrum and the
absence of a 19F resonance for this species. Second activation
compounds (3) were observed for all the di-, tri-, and
tetrafluorobenzene substrates studied (Scheme 5), and 1H
and 19F NMR parameters for these species are as given in
Tables 3 and 4.
Di- vs Trifluorobenzene Competition. The clean and
rapid C−F activation of the di- and trifluorobenzenes in the
presence of AlMe3 and Ni(PEt3)4 allows for a variety of studies
to be performed to gain a better understanding of what factors
are important for catalytic monofunctionalization of polyfluori-
nated arenes. In general, the faster a polyfluorinated aromatic
undergoes nucleophilic aromatic substitution, the more reactive
it is to C−F bond oxidative addition; the order of reactivity is
expected to be C6F6 > C6F5H > C6F4H2 > C6F3H3 > C6F2H4 >
C6FH5 with Ni(0) complexes.5a Thus, it might be expected
that, in the catalytic functionalization of perfluorinated arenes,
di- or trisubstitution products should arise only from ring
whizzing, and not from competition between the substrate and
monofunctionalized products after dissociation from the metal
center. We tested this assumption under stoichiometric
D
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 2. Crystallographic Data for Compounds 112,13,14,123a,124,135,1235,1245
112
113
114
1123a
1124
1135
formula
C19H37FNiP2
405.13
C19H37FNiP2
405.13
C19H37FNiP2
405.14
C19H36F2NiP2
423.13
C19H36F2NiP2
423.13
C19H36F2NiP2
423.13
fw
cryst size (mm3)
0.21 × 0.18 × 0.16
0.25 × 0.21 × 0.15
0.45 × 0.40 × 0.24
0.36 × 0.2 × 0.14 0.38 × 0.35 ×
0.38 × 0.35 × 0.15
0.15
color
yellow
yellow
yellow
yellow
orthorhombic
Pna21
yellow
yellow
cryst syst
space group
a (Å)
triclinic
triclinic
triclinic
triclinic
triclinic
P1
̅
P1
P1
̅
P1
̅
P1
̅
̅
8.9294(5)
9.0260(5)
15.6529(8)
104.318(2)
95.468(2)
112.444(2)
1104.15(11)
2
8.8535(4)
9.1739(4)
14.9537(7)
95.4440(10)
97.7340(10)
111.0600(10)
1109.56(9)
2
8.7209(4)
9.1999(5)
14.9603(7)
96.243(2)
96.832(2)
111.361(2)
1094.78(9)
2
23.868(2)
8.7008(7)
10.7163(9)
90
9.0188(5)
9.0954(5)
15.6547(8)
105.4480(10)
95.214(2)
112.3410(10)
1117.95(10)
2
12.4069(5)
13.6769(6)
14.8874(6)
107.1400(10)
110.7570(10)
90.2580(10)
2239.99(16)
4
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
ρcalcd (g cm−3
μ (mm−1
90
90
2225.5(3)
4
)
1.219
1.213
1.229
1.263
1.257
1.255
)
1.029
1.024
1.038
1.031
1.026
1.024
F(000)
436
436
436
904
452
904
θ range (deg)
θ comp (deg)
no. of rflns collected
R(int)
2.95−30.00
99.7
3.04−38.49
99.9
2.97−35.00
99.6
3.02−30.00
99.9
2.92−34.998
99.9
2.93/33.26
99.9
26804
72918
59543
19024
52494
132123
0.0745
0.0507
0.0315
0.0341
6421/1/225
0.0392
0.0548
no. of data/restraints/
params
6424/0/215
11686/0/225
9621/0/215
9858/18/261
19701/0/447
GOF
1.058
1.044
1.118
1.027
1.025
1.043
a
R1, wR2 (I > 2σ(I))
0.0549/0.1250
0.1149/0.1439
1.21/−0.39
0.0444/0.1067
0.0591/0.1153
0.95/−0.94
0.0410/0.0760
0.0615/0.0879
0.83/−0.68
11235
0.0345/0.0730
0.0511/0.0788
0.61/−0.28
0.0412/0.0837
0.0724/0.0945
0.51/−0.34
0.0457/0.0816
0.0873/0.0928
0.42/−0.34
a
R1, wR2 (all data)
diff peak/hole (e Å−3
)
11245
formula
fw
C19H35F3NiP2
441.12
C19H37FNiP2
405.13
cryst size (mm3)
0.17 × 0.16 × 0.14
yellow
0.25 × 0.21 × 0.15
yellow
color
cryst syst
space group
a (Å)
triclinic
triclinic
P1
̅
P1
̅
12.3705(13)
13.6600(13)
14.8513(14)
106.971(3)
110.937(3)
90.913(3)
2221.4(4)
4
8.8535(4)
9.1739(4)
14.9537(7)
95.4440(10)
97.7340(10)
111.0600(10)
1109.56(9)
2
b (Å)
c (Å)
α (deg)
β (deg)
γ (deg)
V (Å3)
Z
ρcalcd (g cm−3
)
1.319
1.213
μ (mm−1
)
1.042
1.024
F(000)
936
436
θ range (deg)
θ comp (deg)
2.93−29.99
99.9
3.04−38.49
99.9
no. of rflns collected
105448
40493
R(int)
0.0995
0.0434
no. of data/restraints/params
GOF
12942/0/485
1.04
11686/0/225
1.044
a
R1, wR2 (I > 2σ(I))
0.0452/0.0784
0.0795/0.0879
0.571/−0.439
0.0444/0.1067
0.0591/0.1153
0.95/−0.94
a
R1, wR2 (all data)
diff peak/hole (e Å−3
)
a
2
2
R1 = ∑||Fo| − |Fc||/[∑|Fo|]; wR2 = {[∑w(Fo − Fc)2]/[∑w(Fo )2]}1/2
.
activation conditions by performing a competition reaction
where a 5-fold excess of both 1,3-difluorobenzene and 1,3,5-
trifluorobenzene were reacted with Ni(PEt3)4 and AlMe3, as
shown in Scheme 6. These substrates were chosen because they
are the most similar among fluorobenzenes with different
degrees of fluorination, lacking both o-F substituents and p-F
substituents. Also, in a catalytic system, the monofunctionaliza-
tion of 1,3,5-trifluorobenzene would generate a functionalized
E
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
donors were also examined. Stoichiometric Et3N provided
nearly the same selectivity as stoichiometric THF. The best
stoichiometric donor additives were found to be 1,4-
diazabicyclo[2.2.2]octane (DABCO) and pyridine, the latter
of which provided a 95% selectivity for 1,3,5-C6F3H3 activation.
A similar selectivity could be obtained by using isolated
crystalline AlMe3·(pyridine) in combination with Ni(PEt3)4 in
hydrocarbon solvents.
Scheme 5. Second Activation Products of the Di-, Tri-, and
Tetrafluorobenzenes
Improved Monoactivation of the Tetrafluoroben-
zenes. Since the addition of pyridine to AlMe3 showed
improved selectivity, 1,2,3,5- 1,2,4,5-, and 1,2,3,4-tetrafluor-
obenzenes were all reacted with Ni(PEt3)4 and AlMe3·
(pyridine) complex to give the activation products
1
1245,1235,1234a,1234b, without the generation of impurities from
reductive elimination or the second activation products
31245,1235,1234a,1234b observed with AlMe3. Unlike the nearly
unselective AlMe3 activation of 1,2,3,4-tetrafluorobenzene, the
activation with AlMe3·(pyridine) occurred with improved
regioselectivity, as shown in Scheme 7. Similarly, improved
regioselectivity was observed in the activation of 1,2,3-
trifluorobenzene by AlMe3·(pyridine), which occurred to give
1123a and 1123b in a 99:1 ratio, as shown in Scheme 8; with
AlMe3 as the transmetalating agent a 94:6 ratio was obtained.
Di-, Tri-, and Tetrafluorobenzene Competition Re-
actions. Improved selectivity with the AlMe3·(pyridine)
adduct was also observed in competition reactions between
the difluorobenzenes. Reactions were carried out with a 5-fold
excess of each of the three difluorobenzenes, and the results are
summarized in Scheme 9. The reaction using AlMe3 showed no
selectivity among the difluorobenzenes, whereas when AlMe3·
(pyridine) was used, there was greater selectivity for 1,2-
difluorobenzene activation.
Competition reactions with all of the di-, tri-, and
tetrafluorobenzenes were also performed. The relative rates
span 4 orders of magnitude and are summarized in Figure 1.
These rates could be affected by many factors, including the
thermodynamic stability of the Ni(PEt3)2 adducts of the
arenes15a and their relative rate of formation, as well as the
barrier to aluminum-aided C−F activation. There are a few
predictions that can be made from these results. First, certain
substrates should be easier to monofunctionalize than others,
due to their reactivity relative to their functionalization
products, assuming the substituent added does not influence
activity. For example, 1,2-difluorobenzene should be easiest of
the difluorobenzenes to monofunctionalize because it should be
the most reactive relative to a monofluorinated product. Steric
effects are likely to even further favor monosubstitution in the
1,3-difluoroarene that should have reactivity similar to that of
1,3-difluorobenzene; thus, this is a practical comparison.
The activation of 1,3,5-C6F3H3 and 1,3-C6F2H4 occurred
with little selectivity and provided a 2.2:1 ratio of 1135 and 113.
This unexpected result suggests that, in catalytic C−F
functionalization, rapid ring whizzing is not the only
mechanism by which second activation products could be
observed; rather, the direct competition between functionaliza-
tion products and substrates is a potential problem.
Although the traditional approach to improving selectivity
would involve ligand design, the strongly Lewis acidic
transmetalating reagent is clearly involved directly in the C−
F activation step and thus provides an additional means to tune
reaction conditions. The effect oxygen and nitrogen donors
have on selectivity was investigated, and the results are
summarized in Table 6. The use of 1,4-dioxane, Et2O, or
THF as the reaction solvent all produced improvements in the
selectivity of the reaction. THF provided the greatest selectivity,
with 95% of the C−F activation products arising from the C−F
activation of 1,3,5-C6F3H3 vs 1,3-C6F2H4. These reactions took
30−60 min to go to completion at room temperature, unlike
the instantaneous reactions with AlMe3 in the absence of an
added donor. As a result, the reductive elimination products
from the decomposition of thermally sensitive 114 and 1135 are
observed in significant amounts in solution. The second
activation products from ring whizzing, 314 and 3135, were
not observed, due to the decreased rate of activation.
When used as a neat solvent, THF undergoes Lewis acid
catalyzed polymerization, which limits its utility. As a
stoichiometric additive, THF in C6D6 provided selectivity that
was intermediate between the reactions performed in neat
C6D6 and neat THF. Given that donors which formed strong
adducts with AlMe3 had better selectivity, a variety of nitrogen
Table 3. Summary of NMR Data for C−F Bond Activation Products 212,13,14,123a,123b,124,135,1234a,1234b,1245,1235
19F
1H arene−CH3
212
−117.6 (m)
−114.1 (m)
−118.4 (m)
−139.8 (m), −143.3 (m)
−112.8 (m)
2.04 (s)
1.95 (s)
1.96 (s)
213
214
4
2123a
2123b
2124
2.0 (d, JFH = 2.2 Hz)
2.08 (s)
4
−120.2 (m), −124.0 (m)
1.94 (d, JFH = 1.8 Hz)
3
2135
−111.4 (t, JFH = 8.8 Hz)
1.88 (s)
4
21235
21245
21234(a)
21234(b)
−116.0 (m), −134.2 (m), −147.1 (m)
−119.3 (m), −138.4 (m), −144.4 (m)
−138.5 (m), −139.2 (m), −162.4 (m)
−120.4 (m), −138.3 (m), −143.6 (m)
1.90 (d, JFH = 2.2 Hz)
1.90 (s)
1.91 (s)
1.93 (s)
F
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Table 4. Summary of NMR Data for C−F Bond Activation Products 312,13,14,123a,123b,124,135,1234a,1234b,1245,1235
19F
31P{1H}
17.9 (s)
18.0 (s)
1H Ni−CH3
1H arene−CH3
3
312
N/A
−0.78 (t) JPH = 8.8 Hz
overlapped by 113 (−0.80)
−0.83 (t) JPH = 9.0 Hz
−0.72 (t) JPH = 8.8 Hz
2.69 (s)
313
N/A
2.20 (s)
1.96 (s)
6
3
314
N/A
18.8 (d) JPF = 2.8 Hz
19.8 (s)
3
4
3123a
3123b
3124
−115.9 (s)
−87.5 (s)
2.0 (d) JFH = 2.2 Hz
2.08 (s)
3
20.2 (s)
−0.81 (t) JPH = 8.9 Hz
3
−122.9 (m)
18.1 (s)
−0.80 (overlapped t) JPH = 8.0 Hz
2.21 (s)
3
3
3135
−119.0 (t, JFH = 9.3 Hz)
18.1 (s)
−0.80 (overlapped t) JPH = 9.1 Hz
1.96 (s)
3
4
31235
31245
31234(a)
31234(b)
−99.4 (m), −134.2 (m)
−97.4 (m), −129.6 (m)
−138.5 (m), −139.2 (m)
−120.4 (m), −138.3 (m)
18.7 (s)
−0.74 (t) JPH = 9.0 Hz
Overlapped by 11245 (−0.69)
−0.80 (t) JPH = 9.0 Hz
−0.72 (t) JPH = 9.3 Hz
1.96 (d) JFH = 2.2 Hz
18.4 (s)
1.94 (s)
1.91 (s)
1.93 (s)
3
20.3 (s)
3
20.5 (s)
Scheme 6. Competition of 1,3-Difluorobenzene vs 1,3,5-
Trifluorobenzene Activation by Ni(PEt3)4 in the Presence of
AlMe3 and Donor Solvents or Stoichiometric Additives
Scheme 7. Tetrafluorobenzene Activation with AlMe3·
(pyridine)
Table 6. Effects of Oxygen and Nitrogen Donors on the
Competition Shown in Scheme 6
1,3,5-activation
acd
1,3-activation
bcd
, ,
,
,
solvent
products
products
C6D6
69 (3)
82 (4)
31 (2)
18 (4)
13 (8)
5 (2)
a
b
c
Percent yield based on 19F NMR spectra. Isolated yield. Selectivity
1,4-dioxane (neat)
Et2O (neat)
between 1-site and 2-site activation.
87 (35)
95 (47)
THF (neat)
Scheme 8. Improved Regioselectivity in 1,2,3-
Trifluorobenzene Activation with AlMe3·(pyridine)
e
acd
, ,
bcd
, ,
donor
1,3,5-activation products
1,3-activation products
THF (1 equiv)
Et3N
85 (4)
84 (4)
92 (6)
95 (22)
15 (3)
16 (2)
8 (1)
1/2 DABCO
pyridine
5 (0)
a
Percentage includes both the activation product 1135 and its reductive
b
elimination product 3,5-difluorotoluene. Percentage includes both the
Scheme 9. Difluorobenzene Competition using AlMe3 and
AlMe3·(pyridine)
activation product 113 and its reductive elimination product 3-
c
d
fluorotoluene. Percentages based upon 19F NMR. Percentage of
e
reductive elimination product in parentheses. 1 equiv vs AlMe3, in
C6D6.
1,2-substituted product. The least reactive difluoroarene, 1,4-
difluorobenzene, should be the hardest to monofunctionalize
since it is the least reactive of the difluorinated arenes; thus, the
monofunctionalized product can compete most effectively with
this substrate. This result provides a rationale for the result
shown in Scheme 1B,13 where the disubstitution products in
the catalytic functionalization of the 1,2-, 1,3-, and 1,4-
difluorobenzenes accounted for 0, 12, and 25% of the product
mixture, respectively. For the tri- and tetrafluorobenzenes the
difficulty in achieving monofunctionalization should be related
to not only the fluorination pattern of the starting material but
also the product. For example, the monofunctionalization of
1,2,3,5-tetrafluorobenzene should be relatively facile, since it
should be around 32 times more reactive than its mono-
functionalization product, which would be 1,2,4-trifluoroarene;
once again this ignores the electronic effect and steric effects of
the added substituent. The selective monofunctionalization of
1,2,3,4-tetrafluorobenzene at the 1-site should be among the
G
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
EXPERIMENTAL SECTION
■
General Procedure. Unless otherwise stated, all reactions were
performed under an inert nitrogen atmosphere using an Innovative
Technology glovebox. Dry, oxygen-free solvents were used for all
experiments. Anhydrous THF and pentane packed under argon were
purchased from Alfa Aesar and dried with molecular sieves. Benzene-d6
was freeze−pump−thawed three times and dried by passing through 5
g of Brockmann I activated aluminum oxide. All fluorinated aromatics,
phosphines, and Lewis acids were purchased from Sigma-Aldrich and
degassed prior to use. Literature procedures were used for the
synthesis of Ni(COD)2,20 and Ni(PEt3)4.21 1H, 31P{1H}, 19F, 19F{1H},
and 13C{1H} NMR were recorded on a Bruker AMX spectrometer
operating at 500 or 300 MHz with respect to the proton nuclei. All
chemical shifts are reported in parts per million, and all coupling
1
constants are reported in hertz. H NMR spectra were referenced to
residual protons (C6D5H, δ 7.15; THF, δ 3.62) with respect to
tetramethylsilane at δ 0.00. 31P{1H} NMR spectra were referenced to
external 85% H3PO4 at δ 0.0. 19F and 19F{1H} NMR were referenced
to external α,α,α-trifluorotoluene at δ −63.7. 13C{1H} NMR spectra
were referenced to the solvent resonance (C6D6, δ 128.0). Elemental
analyses were performed at the Centre for Catalysis and Materials
Research, Windsor, Ontario, Canada.
Figure 1. Rates of reaction with Ni(PEt3)4 and AlMe3·(pyridine),
relative to 1,4-difluorobenzene, with sites of selective activation circled
in red. Based on 19F NMR integrals of competition reaction between
the two reactants and includes all reductive elimination products. a)
for activation at both the 1- and 2-site.
Synthesis of CH3(PEt3)2Ni(2-FC6H4) (112). Crystalline Ni(PEt3)4
(1.0 g, 1.88 mmol), 1 mL of pentane, and 1,2-difluorobenzene (0.236
mg, 2.07 mmol) were placed in a 50 mL flask. A solution of AlMe3
(149 mg, 2.07 mmol) in pentane (2.08 mL) was added dropwise at
room temperature. This solution was then passed through 1 cm of
dried 100 mesh silica to remove any excess AlMe3 and aluminum
byproducts. The solvent was removed under vacuum until the product
began to crystallize. This solution was put into a−35 °C refrigerator
overnight to finalize crystallization. The orange-yellow solid was
isolated by filtration and dried under vacuum (total yield 0.528 g,
69.4%). Crystals suitable for characterization by X-ray crystallography
were obtained directly from the recrystallized product. 1H NMR
(C6D6, 500 MHz, 298 K): δ −0.65 (t, 3JPH = 9.0 Hz, 3H, NiCH3), 0.95
(coincident t of vt 3JHH = 7.5 Hz, 3JPH + 5JPH = 15.5 Hz, and 2JPP > 50
Hz, 18H, NiPCH2CH3), 1.25 (m, 12H), 6.83 (m, 1H), 6.88 (m, 1H),
6.94 (m, 1H), 7.48 (m, 1H). 31P{1H} NMR (C6D6, 202.46 MHz, 298
K): 19.7 (s, 2P). 19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −89.4
(s, 1F). 13C{1H} NMR (C6D6, 125.77 MHz, 298 K): −9.9 (t of d 2JPC
= 24.4 Hz, 4JFC = 3.65 Hz, 1C), 8.4 (s, 6C), 14.7 (vt, 1JPC + 3JPC = 11.7
hardest to accomplish. This substrate would be expected to
react only ∼3 times as fast as the 1,2,3-trifluoroarene product
that such a monofunctionalization should afford, and the
substituent added does not sterically protect the predicted site
of functionalization of this trifluoroarene.
CONCLUSIONS
■
Although many nickel-based systems have been designed for
C−F bond functionalization, little has been reported to date
about the design principles necessary for monofunctionalization
of the polysubstituted aromatics, despite some ligand design
based success. Even the nature of the C−F activation steps
remained unclear in these processes; many of the ligand designs
that support catalysis use ligands that yield relatively electron
poor Ni(0) moieties that have not been observed to give
stoichiometric C−F bond activation. The ability of AlMe3 to
provide instant C−F bond activation of even the least reactive
polyfluorinated aromatics using Ni(PEt3)4 implies that, in many
catalytic cases, the C−F oxidative addition step and trans-
metalation step are intimately mixed. The nickel(II) fluoride
intermediates commonly proposed in catalytic cycles for
coupling reactions are unlikely to exist as such in these
systems. The reduction in the activity of the Lewis acidic AlMe3
by the addition of donor solvents had a strong impact on
selectivity, providing an additional means to tune reactivity
without necessitating novel ligand designs. Catalyst designs
must also take into consideration the competition between
functionalized products and the starting polyfluorinated
aromatic.
2
Hz, and JPP > 150 Hz, 6C, NiPCH2CH3), 112.5 (s, 1C), 122.6 (m,
2
4
1C), 122.9 (m, 1C), 139.5 (d of t JFC = 23.5 Hz JPC = 2.9 Hz, 1C),
153.6 (m, 1C), 168.3 (d of t 3JPC = 3.6 Hz, 1JFC = 221.7 Hz, 1C). Anal.
Calcd for CH3(PEt3)2Ni(2-FC6H4): C, 56.33; H, 9.2. Found: C, 56.33;
H, 9.57.
Synthesis of CH3(PEt3)2Ni(3-FC6H4) (113). Synthesis and
crystallization were performed using the same procedure as for 112,
but with 1,3-difluorobenzene in lieu of 1,2-difluorobenzene. The
orange-yellow solid was isolated by filtration and dried under vacuum
(total yield 0.508 g, 66.7%). Crystals suitable for characterization by X-
ray crystallography were obtained directly from the recrystallized
1
3
product. H NMR (C6D6, 500 MHz, 298 K): δ −0.80 (t, JPH = 8.5
Hz, 3H, NiCH3), 0.88 (coincident t of vt, 3JHH = 8.0 Hz, 3JPH + 5JPH
14.5 Hz, JPP
=
2
> 50 Hz, 18H, NiPCH2CH3), 1.19 (m,
12H,NiPCH2CH3), 6.60 (m, 1H, Ar−H), 6.94 (m, 1H, Ar−H), 7.27
3
(broad d JFH = 7.0 Hz, 1H, Ar−H), 7.36 (m, 1H, Ar−H). 31P{1H}
5
NMR (C6D6, 202.46 MHz, 298 K): 18.8 (d, JPF = 1.82 Hz, 2P).
19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −116.9 (s, 1F). 13C{1H}
2
The conclusions regarding the intimate involvement of
transmetalating reagents in the activation of unreactive C−F
bonds may be true for other difficult oxidative addition
reactions such as C−O bonds, which have previously been
shown to undergo catalytic functionalization using Lewis acids
and Ni-based catalysts.19 Further work is underway in our
laboratory to explore the scope of this effect, as well as to better
understand the influence of donor solvents on the selectivities
observed in the activation of fluorinated aromatics.
NMR (C6D6, 125.77 MHz, 298 K): −10.5 (t, JPC = 23.5 Hz, 1C,
NiCH3), 8.4 (s, 6C, NiPCH2CH3), 14.5 (vt, 1JPC + 3JPC = 11.8 Hz, 2JPP
> 150 Hz, 6C, NiPCH2CH3), 107.0 (d of t, 2JFC = 21.5 Hz, 3JPC = 2.1
2
5
Hz, 1C), 123.9 (d of t, JFC = 12.8 Hz, JPC = 2.6 Hz, 1C), 126.3 (m,
1C), 133.8 (m,1C), 162.2 (d of t 1JFC = 248.9 Hz, 4JPC = 2.8 Hz, 1C),
178.3 (t, JPC = 28.4 Hz, 1C). Anal. Calcd for CH3(PEt3)2Ni(3-
2
FC6H4); C, 56.33; H, 9.2. Found: C, 55.95; H, 9.16.
Synthesis of CH3(PEt3)2Ni(4-FC6H4) (114). Synthesis and
crystallization were performed using the same procedure as for 112,
but with 1,4-difluorobenzene in lieu of 1,2-difluorobenzene. The
H
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3
orange-yellow solid was isolated by filtration and dried under vacuum
(total yield 0.588 g, 77.3%). Crystals suitable for characterization by X-
ray crystallography were obtained directly from the recrystallized
(d, JHF = 8.5 Hz). 31P{1H} NMR (C6D6, 202.46 MHz, 298 K): 19.6
(s, 2P). 19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −124.7 (d, 4JFF
=
21.6 Hz, 1F), −97.1 (d, JFF = 21.6 Hz, 1F). 19F NMR (C6D6, 470.55
MHz, 298 K): −124.7 (m, 1F), −97.1 (d, 4JFF = 21.6 Hz, 1F). 13C{1H}
4
1
3
product. H NMR (C6D6, 500 MHz, 298 K): δ −0.79 (t, JPH = 8.5
Hz, 3H, NiCH3), 0.91 (coincident t of vt, 3JHH = 7.5 Hz, 3JP3H + 5JPH
=
NMR (C6D6, 125.77 MHz, 298 K): −9.9 (t of d 2JCP = 24.3 Hz, 4JCF
=
2
3.6 Hz, 1C), 8.3 (s, 1C), 14.6 (vt, 1JCP + 3JCP = 12.2 Hz, 2JPP > 150 Hz,
6C), 108.6 (d of d, 3JCF = 8.9 Hz, 2JCF = 24.5 Hz, 1C), 112.3 (m, 1C),
15.0 Hz, JPP > 50 Hz, 18H, NiPCH2CH3), 1.19 (q of vt, JHH = 7.5
Hz, 2JPH + 4JPH = 6.5 Hz, 2JPP > 50 Hz, 12H, NiPCH2CH3), 6.90 (2nd
order AA′BB′X, 2H, Ar-H), 7.32 (2nd order AA′BB′X m, 2H, Ar-H).
112.6 (d, 2JCF = 8.2 Hz, 1C), 124.2 (d of d of t, 3JCF = 17.2 Hz, 2JCF
=
6
31P{1H} NMR (C6D6, 202.46 MHz, 298 K): 18.8 (d, JPF = 2.8 Hz).
25.5 Hz, 3JCP = 2.8 Hz, 1C), 158.6 (m, 1C), 163.9 (d of t, 1JCF = 215.8
3
19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −126.5 (t, 6JPF = 2.8 Hz).
19F NMR (C6D6, 470.55 MHz, 298 K): −126.5 (t of t of t, 3JFH = 10.0
Hz, JCP = 3.8 Hz, 1C). Anal. Calcd for CH3(PEt3)2Ni(2,5-F2C6H3):
C, 53.93; H, 8.58. Found: C, 54.05; H, 8.91
Synthesis of CH3(PEt3)2Ni(3,5-F2C6H3) (1135). Synthesis and
crystallization were performed using the same procedure as for 1123a
but with 1,3,5-trifluorobenzene in lieu of 1,2,3-trifluorobenzene. The
orange-yellow solid was isolated by filtration and dried under vacuum
(total yield 0.557 g, 70.06%). Crystals suitable for characterization by
X-ray crystallography were obtained directly from the recrystallized
6
Hz 4JFH = 7.0 Hz JPF = 2.8 Hz). 13C{1H} NMR (C6D6, 125.77 MHz,
2
,
298 K): −10.6 (t, JPC1 = 22.6 Hz, 1C, NiCH3), 8.5 (s, 6C,
3
2
NiPCH2CH3), 14.6 (vt, JPC + JPC = 11.9 Hz, JPP > 150 Hz, 6C,
2
4
NiPCH2CH3), 113.0 (d of t JFC = 16.8 Hz, JPC = 2.0 Hz, 2C, Ar),
138.3 (d of t 3JFC = 3.8 Hz, 3JPC = 2.6 Hz, 2C, Ar), 160.6 (d of t, 1JFC
=
=
5
4
2
237.2 Hz, JPC = 1.8 Hz, 1C, Ar), 164.5 (d of t, JFC = 3.6 Hz, JPC
1
3
product. H NMR (C6D6, 500 MHz, 298 K): δ −0.85 (t, JHP = 9.0
29.1 Hz, 1C, Ar). Anal. Calcd for CH3(PEt3)2Ni(4-FC6H4): C, 56.33;
H, 9.2. Found: C, 56.30; H, 9.40.
Hz, 3H, NiCH3), 0.84 (coincident t of vt, 3JHH = 7.5 Hz, 3JHP + 5JHP
=
15.0 Hz, 2JPP > 50 Hz, 18H, NiPCH2CH3), 1.14 (m, 12H), 6.36 (2nd
order m, 1H), 7.15 (m, 2H). 31P{1H} NMR (C6D6, 202.46 MHz, 298
K): 18.8 (s,2P). 19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −116.8
(s, 2F). 19F NMR (C6D6, 470.55 MHz, 298 K): −116.8 (m, 2F). 13C
CH3(PEt3)2Ni(C6H4CH3) (314). Complex 314 was identified as a
1
minor side product in the synthesis of 114 by H and 31P{1H} NMR.
Conversion to 314 increased if the reaction mixture was not worked up
immediately to remove excess aluminum methyl species. Attempts to
isolate 314 failed due to concomitant conversion to p-xylene. 1H NMR
(C6D6, 500 MHz, 298 K): δ −0.75 (t overlapped by main product,
3JPH = 8.5 Hz, 3H, NiCH3), δ 2.28 (s, 3H, CH3), 6.99 (m, 2H, Ar-H),
7.46 (m, 2H, Ar-H). 31P{1H} NMR (C6D6, 202.46 MHz, 298 K): 18.7
(s, 2P).
Synthesis of CH3(PEt3)2Ni(2,3-F2C6H3) (1123a). Crystalline Ni-
(PEt3)4 (1.0 g, 1.88 mmol), 1 mL of pentane, and 1,2,3-
trifluorobenzene (273 mg, 2.07 mmol) were placed in a 50 mL
flask. A solution of AlMe3 (149 mg, 2.07 mmol) in pentane (2.08 mL)
was added dropwise at room temperature. This solution was then
passed through 1 cm of dried 100 mesh silica to remove any excess
AlMe3 and aluminum byproducts. The solvent was removed under
vacuum until the product began to crystallize. This solution was put
into a −35 °C refrigerator overnight to finalize crystallization. The
orange-yellow solid was isolated by filtration and dried under vacuum
(total yield 0.520 g, 65.4%). Crystals suitable for characterization by X-
ray crystallography were obtained directly from the recrystallized
3
NMR (C6D6, 125.77 MHz, 298 K) −10.4 (t, JCP = 23.8 Hz, 1C), 8.4
(s, 6C), 14.4 (vt, 1JCP + 3JCP = 12.2 Hz, and 2JPP > 150 Hz, 6C), 95.4 (t
of t, 2JCF = 25.7 Hz, 5JCP = 2.0 Hz, 2C), 119.0 (m, 2C), 161.6 (d of d of
3
1
4
2
t, JCF = 9.9 Hz, JCF = 252.3 Hz, JCP = 3.5 Hz, 1C), 182.5 (t, JCF
28.3 Hz, 1C). Anal. Calcd for CH3(PEt3)2Ni(3,5-F2C6H3): C, 53.93;
H, 8.58. Found: C, 54.02; H, 8.89.
=
Synthesis of CH3(PEt3)2Ni(2,3,5-F3C6H2) (11235). Crystalline
Ni(PEt3)4 (1.0 g, 1.88 mmol), 1 mL of pentane, and 1,2,3,5-
tetrafluorobenzene (310 mg, 2.07 mmol) were placed in a 50 mL flask
(A). Crystalline AlMe3·(pyridine) (312 mg, 2.07 mmol) and pentane
(1.0 mL) were placed in a 10 mL flask (B). Solution B was added to
solution A dropwise at room temperature with stirring. This solution
was allowed to sit at room temperature under an inert atmosphere for
30 min. Next it was passed through 1 cm of dried 100 mesh silica to
remove any excess AlMe3·(pyridine) and aluminum byproducts. The
solvent was removed under vacuum until the product began to
crystallize. This solution was put into a −35 °C refrigerator overnight
to finalize crystallization. The orange-yellow solid was isolated by
filtration and dried under vacuum (total yield 0.597 g, 72.0%). Crystals
suitable for characterization by X-ray crystallography were obtained
1
3
product. H NMR (C6D6, 500 MHz, 298 K): δ −0.68 (t, JPH = 9.0
Hz, 3H, NiCH3), 0.91 (coincident t of vt, 3JHH = 7.5 Hz, 3JPH + 5JPH
15.0 Hz, JPP > 150 Hz, 18H, NiPCH2CH3), 1.21 (m, 12H,
NiPCH2CH3), 6.67 (m, 1H), 6.75 (m, 1H), 7.12 (m, 1H). 31P{1H}
NMR (C6D6, 202.46 MHz, 298 K): 19.8 (s, 2P). 19F{1H} NMR
=
2
1
directly from the recrystallized product. H NMR (C6D6, 300 MHz,
3
298 K): δ −0.69 (t, JHP = 9.0 Hz, 3H, NiCH3), 0.85 (coincident t of
3
3
2
vt, JHH = 7.5 Hz, JHP
+
5JHP = 15.0 Hz, JPP > 50 Hz, 18H,
3
(C6D6, 470.55 MHz, 295 K): −142.1 (d, JFF = 36.2 Hz, 1F),−115.9
(d, 3JFF = 36.2 Hz, 1F). 19F NMR (C6D6, 470.55 MHz, 294 K): −142.1
(d of d of t, 3JFF = 36.2 Hz, 5JFP = 2.3 Hz, 3JFH = 11.3 Hz, 1F). −115.9
NiPCH2CH3), 1.14 (m, 12H), 6.45 (2nd order m, 1H), 7.01 (m, 2H).
31P{1H} NMR (C6D6, 121.51 MHz, 298 K): 19.6 (overlapping d of d,
5JPF = 3.8 Hz, JPF = 3.8 Hz, 2P). 19F NMR (C6D6, 282.40 MHz, 298
5
3
4
(d of d, JFF = 36.2 Hz, JFH = 8.0 Hz, 1F). 13C{1H} NMR (C6D6,
125.77 MHz, 298 K): −9.8 (t of d, 2JCP = 24.1 Hz, 3JCF = 3.5 Hz, 1C),
8.4 (s, 6C), 14.6 (vt, 1JCP+3JCP = 12.7 Hz, 2JPP > 150 Hz, 6C), 109.8 (d,
K): −120.9 (d of d of d, 3JFF = 35.8 Hz, 4JFF = 19.6 Hz, 5JFH = 6.2 Hz,
4
5
1F), −121.2 (overlapping d of t, JFF = 20.3 Hz, JPF = 7.7 Hz, 1F),
−138.1 (d of d, 3JFF = 35.8 Hz,, 3JFH = 10.4 Hz, 1F) 13C NMR (C6D6,
75.48 MHz, 298 K) −9.6 (t, 3JCP = 24.3 Hz, 1C), 8.2 (s, 6C), 14.4 (vt,
1JCP + 3JCP = 12.3 Hz, and 2JPP > 150 Hz, 6C), 97.8 (overlapping d of
2JCF = 13.0 Hz, 1C), 122.6 (d, JCF = 2.0 Hz, 1C), 133.4 (m, 1C),
3
150.3 (d of d of t, 3JCP = 2.5 Hz, 2JCF = 21.8 Hz, 1JCF = 251.2 Hz, 1C),
152.7 (m, 1C), 159.1 (m, 1C). Anal. Calcd for CH3(PEt3)2Ni(2,3-
F2C6H3): C, 53.93; H, 8.58. Found: C, 53.97; H, 8.65
2
5
d, JCF = 25.7 Hz, JCP = 2.0 Hz, 1C), 117.7 (m, 1C), 149.0 (d of m,
1
CH3(PEt3)2Ni(2,6-F2C6H3) (1123b). Activation at the 2-site of 1,2,3-
1JCF = 235.7 Hz, 1C), 149.5 (d of m, JCF = 215.6 Hz, 1C), 155.9 (m,
trifluorobenzene. 1H NMR (C6D6, 500 MHz, 298 K): −0.56 (t, 3JHP
=
1C), 159.2 (m, 1C). Anal. Calcd for CH3(PEt3)2Ni(2,3,5-F3C6H2): C,
51.73; H, 8.00. Found: C, 51.52; H, 8.06.
Synthesis of CH3(PEt3)2Ni(2,4,5-F3C6H2) (11245). Synthesis and
9.5 Hz). 31P{1H} NMR (C6D6, 202.46 MHz, 298 K): 19.7 (s).
19F{1H} NMR (C6D6, 470.55 MHz, 298 K): −87.5 (s). 19F NMR
3
crystallization were performed using the same procedure as for 11235
,
(C6D6, 470.55 MHz, 298 K): −87.5 (t, JFH = 6.1 Hz).
Synthesis of CH3(PEt3)2Ni(2,5-F2C6H3) (1124). Synthesis and
but with 1,2,4,5-tetrafluorobenzene in lieu of 1,2,3,5-tetrafluoroben-
zene. The orange-yellow solid was isolated by filtration and dried
under vacuum (total yield 0.563 g, 67.89%). Crystals suitable for
characterization by X-ray crystallography were obtained directly from
the recrystallized product. 1H NMR (C6D6, 300 MHz, 298 K): δ
crystallization were performed using the same procedure as for 1123a
,
but with 1,2,4-trifluorobenzene in lieu of 1,2,3-trifluorobenzene. The
orange-yellow solid was isolated by filtration and dried under vacuum
(total yield 0.500 g, 62.89%). Crystals suitable for characterization by
X-ray crystallography were obtained directly from the recrystallized
3
3
−0.71 (t, JHP = 9.0 Hz, 3H, NiCH3), 0.87 (coincident t of vt, JHH
=
1
3
7.5 Hz, 3JHP + 5JHP = 15.0 Hz, 2JPP > 50 Hz, 18H, NiPCH2CH3), 1.15
product. H NMR (C6D6, 500 MHz, 298 K): δ −0.70 (t, JHP = 9.5
Hz, 3H), 0.90 (coincident t of vt 3JHH = 7.5 Hz, 3JHP + 5JHP = 15.0 Hz,
2JPP > 50 Hz, 18H), 1.21 (m, 12H), 6.49 (m, 1H), 6.60 (m, 1H), 7.24
(m, 12H), 6.58 (2nd order m, 1H), 7.19 (m, 2H). 31P{1H} NMR
4
(C6D6, 121.51 MHz, 298 K): 19.6 (coincident p (d of d of d), JPF
=
I
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
5
6
1.6 Hz, JPF = 1.6 Hz, JPF = 1.6 Hz, 2P). 19F NMR (C6D6, 282.40
MHz, 298 K): −93.2 (d, 4JFF = 18.1 Hz, 1F), −146.9 (m, 1F), −148.7
(m, 1F) 13C NMR (C6D6, 75.48 MHz, 298 K) −10.0 (d of t, 4JCF = 3.0
Et2O. Percentage of activation on the basis of 19F NMR integrals: for
1,4-dioxane, 78% 1135, 4% 2135, 14% 113, and 4% 213; for THF, 48%
1135, 47% 2135, 3% 113 ,and 2% 213; for Et2O, 52% 1135, 35% 2135, 5%
113, and 8% 213.
3
1
3
Hz, JCP = 24.9 Hz, 1C), 8.3 (s, 6C), 14.5 (vt, JCP + JCP = 12.2 Hz,
and 2JPP > 150 Hz, 6C), 102.1 (d of d, 3JCF = 18.2 Hz, 1JCF = 39.6 Hz,
1C), 124.4 (d of d, 3JCF = 12.4 Hz, 1JCF = 27.7 Hz, 1C), 144.7 (m, 1C),
149.7 (d of m, JCF = 215.6 Hz, 1C), 159.8 (m, 1C), 162.8 (m, 1C).
Anal. Calcd for CH3(PEt3)2Ni(2,3,5-F3C6H2): C, 51.73; H, 8.00.
Found: C, 51.79; H, 8.37.
Synthesis of CH3(PEt3)2Ni(2,3,4-F3C6H2) (11234a). Synthesis and
crystallization were performed using the same procedure as for 11235
Crystalline Ni(PEt3)4 (0.050 g, 0.0941 mmol), 1,3-difluorobenzene
(0.054 g, 0.471 mmol), 1,3,5-trifluorobenzene (0.062 g, 0.471 mmol),
and 0.21 mL of C6D6 were placed in a 2 dram vial (A). Pure AlMe3
(0.007 g, 0.0941 mmol) was added to 0.30 mL of C6D6. Next THF
(0.007 g, 0.104 mmol) was added dropwise (B). Solution B was added
1
1
dropwise to solution A with stirring. H, 19F, and 31P NMR spectra
1
were acquired immediately after reaction was prepared. H, 19F, and
,
31P NMR shifts for 1135 and 113 are the same as those reported above.
1H, 19F, and 31P NMR shifts for 2135 and 213 are reported in Table 3.
but with 1,2,3,4-tetrafluorobenzene in lieu of 1,2,3,5-tetrafluoroben-
zene. The orange-yellow solid was isolated by filtration and dried
under vacuum (total yield 0.563 g, 67.89%). X-ray crystallography was
Percentage of activation on the basis of 19F NMR integrals: 81% 1135
,
not obtained due to the presence of the isomer 11234b
.
1H NMR
4% 2135, 12% 113, and 3% 213. The same reaction was carried out in the
same manner but with substitution of THF with pyridine, trimethyl-
amine, and 1,4-diazabicuclo[2,2,2]octane (DABCO). Percentage of
(C6D6, 300 MHz, 298 K): δ −0.69 (t, 3JHP = 9.0 Hz, 3H, NiCH3), 0.87
(coincident t of vt, 3JHH = 7.5 Hz, 3JHP + 5JHP = 15.0 Hz, 2JPP > 50 Hz,
18H, NiPCH2CH3), 1.15 (m, 12H), 6.68 (2nd order m, 1H), 6.86 (m,
2H). 31P{1H} NMR (C6D6, 121.51 MHz, 298 K): 19.7 (s, 2P). 19F
activation on the basis of 19F NMR integrals: for pyridine, 73% 1135
22% 2135, 5% 113, and 0% 213; for trimethylamine, 80% 1135, 4% 2135
,
,
3
14% 113, and 2% 213; for DABCO, 86% 1135, 6% 2135, 7% 113, and 1%
213.
NMR (C6D6, 282.40 MHz, 298 K): −113.0 (d, JFF = 21.4 Hz, 1F),
3
3
−147.9 (d of m, JFF = 11.2 Hz, 1F), −165.7 (d of d, JFF = 11.2 Hz,
3JFF = 21.4 Hz, 1F) 13C NMR (C6D6, 75.48 MHz, 298 K) −10.0 (t,
Difluorobenzene Competition. Crystalline Ni(PEt3)4 (0.050 g,
0.0941 mmol), 1,2-difluorobenzene (0.054 g, 0.471 mmol), 1,3-
difluorobenzene (0.054 g, 0.471 mmol), 1,4-difluorobenzene (0.054 g,
0.471 mmol), and 0.16 mL of C6D6 were placed in a 2 dram vial (A).
Pure AlMe3 (0.034 g, 0.471 mmol), was added to 0.3 mL of C6D6 (B).
Solution B was added dropwise to solution A with stirring at room
temperature. 1H, 19F, and 31P NMR spectra were acquired immediately
after the reaction mixture was prepared. Percentage of activation
products: 33% 112, 33% 113, and 33% 114. The same reaction was
carried out in the same manner but with substitution of AlMe3 with
crystalline AlMe3·(pyridine). Percentage of activation on the basis of
19F NMR integrals: for AlMe3·(pyridine), 76% 112, 21% 113, and 3%
114.
3JCP = 24.9 Hz, 1C), 8.3 (s, 6C), 14.5 (vt, JCP + JCP = 12.2 Hz, and
1
3
2JPP > 150 Hz, 6C), 102.1 (d of d, 3JCF = 18.2 Hz, 1JCF = 39.6 Hz, 1C),
3
1
124.4 (d of d, JCF = 12.4 Hz, JCF = 27.7 Hz, 1C), 144.7 (m, 1C),
149.7 (d of m, JCF = 215.6 Hz, 1C), 159.8 (m, 1C), 162.8 (m, 1C).
Anal. Calcd for CH3(PEt3)2Ni(2,3,5-F3C6H2): C, 51.73; H, 8.00.
Found: C, 51.79; H, 8.37.
Synthesis of CH3(PEt3)2Ni(2,3,4-F3C6H2) (11234b). Synthesis and
crystallization were performed using the same procedure as for 11235
but with 1,2,3,4-tetrafluorobenzene in lieu of 1,2,3,5-tetrafluoroben-
zene. The orange-yellow solid was isolated by filtration and dried
under vacuum (total yield 0.563 g, 67.89%). X-ray crystallography was
1
,
1
not carried out due to the presence of the isomer 11234a. H NMR
Fluorinated Aromatic Competition Reactions. Crystalline
Ni(PEt3)4 (1.35 g, 2.541 mmol) was dissolved in 8.1 mL of C6D6 in
a 2 dram vial (A). Solution B was made 27 different times, as described
in Table S1 in the Supporting Information. In general solution B
consists of 0.471 mmol of two different fluorinated aromatics and
0.471 mmol of crystalline AlMe3·(pyridine) dissolved in benzene
topped up to 0.3 mL. A 0.3 mL portion of solution A was placed in a 2
dram vial. Solution B was added dropwise to solution A with stirring at
room temperature. Reactions took anywhere from 1 to 30 min to go to
(C6D6, 300 MHz, 298 K): δ −0.56 (t, 3JHP = 9.0 Hz, 3H, NiCH3), 0.94
(coincident t of vt, 3JHH = 7.5 Hz, 3JHP + 5JHP = 15.0 Hz, 2JPP > 50 Hz,
18H, NiPCH2CH3), 1.21 (m, 12H), 6.42 (2nd order m, 1H), 6.90 (m,
2H). 31P{1H} NMR (C6D6, 121.51 MHz, 298 K): 20.1 (s, 2P). 19F
5
NMR (C6D6, 282.40 MHz, 298 K): −93.1 (d, JFF = 11.2 Hz, 1F),
−111.5 (d, 3JFF = 20.2 Hz, 1F), −146.3 (d of d, 5JFF = 11.2 Hz, 3JFF
=
=
>
20.2 Hz, 1F). 13C NMR (C6D6, 75.48 MHz, 298 K): −10.0 (t, JCP
3
1
3
2
24.9 Hz, 1C), 8.3 (s, 6C), 14.5 (vt, JCP + JCP = 12.2 Hz, and JPP
150 Hz, 6C), 102.1 (d of d, 3JCF = 18.2 Hz, 1JCF = 39.6 Hz, 1C), 124.4
(d of d, 3JCF = 12.4 Hz, 1JCF = 27.7 Hz, 1C), 144.7 (m, 1C), 149.7 (d
of m, 1JCF = 215.6 Hz, 1C), 159.8 (m, 1C), 162.8 (m, 1C). Anal. Calcd
for CH3(PEt3)2Ni(2,3,5-F3C6H2): C, 51.73; H, 8.00. Found: C, 51.79;
H, 8.37.
1
completion. H, 19F, and 31P{1H} NMR spectra were obtained at first
and up until completion of the reaction. 19F NMR resonances were
used to determine relative rates.
ASSOCIATED CONTENT
Activation Attempts with Different Lewis Acids. Crystalline
Ni(PEt3)4 (0.079 g, 0.150 mmol), 0.60 mL of tetrahydrofuran, and 1,4-
difluorobenzene (0.017 g, 0.150 mmol), were placed in a 2 dram vial.
Solid BPh3 (0.0363 g, 0.150 mmol) was slowly added to the solution at
room temperature. The resultant 19F{1H} NMR shows a majority of
starting materials, along with many unassigned peaks. None of the
peaks present represented the desired product for these reactions. This
procedure, with the same results, was done with the following
compounds in place of BPh3: ZnEt2, FeCl3, FeCl2, DIBAL, SnCl2, and
SnCl3.
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.7b00165.
Spectra for compounds 112,13,14,123a,123b,124,135,-
and 212,13,14,123a,123b,124,135,-
1234a,1234b,1245,1235
(PDF)
1234a,1234b,1245,1235
Crystallographic data for 112,13,14,123a,123b,124,135, 1245,1235
(CIF)
Di- and Trifluorobenzene Competition. Crystalline Ni(PEt3)4
(0.050 g, 0.0941 mmol), 1,3-difluorobenzene (0.054 g, 0.471 mmol),
1,3,5-trifluorobenzene (0.062 g, 0.471 mmol), and 0.21 mL of C6D6
were placed in a 2 dram vial (A). Pure AlMe3 (0.007 g, 0.0941 mmol)
was added to 0.30 mL of C6D6 (B). Solution B was added dropwise to
AUTHOR INFORMATION
■
solution A with stirring. H, 19F, and 31P NMR spectra were acquired
1
Corresponding Author
*E-mail for S.A.J.: sjohnson@uwindsor.ca.
immediately after the reaction mixture was prepared. H, 19F, and 31P
1
NMR shifts for 1135 and 113 are the same as those reported above. 1H,
19F, and 31P NMR shifts for 2135 and 213 are reported in Table 3.
ORCID
Samuel A. Johnson: 0000-0001-6856-0416
Notes
Percentage of activation on the basis of 19F NMR integrals: 66% 1135
,
3% 2135, 29% 113, and 2% 213. The same reaction was carried out in the
same manner but with substitution of C6D6 with 1,4-dioxane, THF, or
The authors declare no competing financial interest.
J
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
(14) (a) Guo, W.-J.; Wang, Z.-X. J. Org. Chem. 2013, 78 (3), 1054−
1061. (b) Yoshikai, N.; Matsuda, H.; Nakamura, E. J. Am. Chem. Soc.
2009, 131 (27), 9590−9599.
(15) (a) Johnson, S. A.; Taylor, E. T.; Cruise, S. J. Organometallics
2009, 28 (13), 3842−3855. (b) Silvestre, J.; Albright, T. A. J. Am.
Chem. Soc. 1985, 107 (24), 6829−6841. (c) Li, T.; García, J. J.;
Brennessel, W. W.; Jones, W. D. Organometallics 2010, 29 (11), 2430−
2445.
(16) Ohashi, M.; Kambara, T.; Hatanaka, T.; Saijo, H.; Doi, R.;
Ogoshi, S. J. Am. Chem. Soc. 2011, 133 (10), 3256−3259.
(17) Lewis acids are also known to assist C−CN bond activation with
Ni(0) compounds: (a) Nakao, Y.; Ebata, S; Yada, A.; Hiyama, T.;
Ikawa, M.; Ogoshi, S. J. Am. Chem. Soc. 2008, 130 (39), 12874−12875.
(b) Minami, Y.; Yoshiyasu, H.; Nakao, Y.; Hiyama, T. Angew. Chem.,
Int. Ed. 2013, 52 (3), 883−887.
(18) Rennekamp, C.; Stasch, A.; Muller, P.; Roesky, H. W.;
Noltemeyer, M.; Schmidt, H. G.; Uson, I. J. Fluorine Chem. 2000,
102 (1−2), 17−20.
ACKNOWLEDGMENTS
■
We thank the Natural Sciences and Engineering Research
Council (NSERC) of Canada for their support of this research.
REFERENCES
■
(1) (a) Isanbor, C.; O’Hagan, D. J. Fluorine Chem. 2006, 127 (3),
303−319. (b) Muller, K.; Faeh, C.; Diederich, F. Science 2007, 317
̈
(5846), 1881−1886.
(2) (a) Keess, S.; Simonneau, A.; Oestreich, M. Organometallics 2015,
́
́
34 (4), 790−799. (b) Gyomore, A.; Bakos, M.; Foldes, T.; Papai, I.;
̈
̈
̈
́ ́
Domjan, A.; Soos, T. ACS Catal. 2015, 5 (9), 5366−5372.
(3) (a) Babudri, F.; Farinola, G. M.; Naso, F.; Ragni, R. Chem.
Commun. 2007, No. 10, 1003−1022. (b) Liang, Z.; Tang, Q.; Liu, J.;
Li, J.; Yan, F.; Miao, Q. Chem. Mater. 2010, 22 (23), 6438−6443.
(c) Shibata, Y.; Tsutsumi, J. y.; Matsuoka, S.; Matsubara, K.; Yoshida,
Y.; Chikamatsu, M.; Hasegawa, T. Appl. Phys. Lett. 2015, 106 (14),
143303.
(19) (a) Tobisu, M.; Chatani, N. Acc. Chem. Res. 2015, 48 (6), 1717−
1726. (b) Tobisu, M.; Takahira, T.; Morioka, T.; Chatani, N. J. Am.
Chem. Soc. 2016, 138 (21), 6711−6714. (c) Sergeev, A. G.; Hartwig, J.
F. Science 2011, 332 (6028), 439−443. (d) Guo, L.; Liu, X.; Baumann,
C.; Rueping, M. Angew. Chem., Int. Ed. 2016, 55 (49), 15415−15419.
(20) Krysan, D. J.; Mackenzie, P. B. J. Org. Chem. 1990, 55 (13),
4229−4230.
(4) (a) Ahrens, T.; Kohlmann, J.; Ahrens, M.; Braun, T. Chem. Rev.
2015, 115 (2), 931−972. (b) Weaver, J.; Senaweera, S. Tetrahedron
2014, 70 (41), 7413−7428. (c) Yu, D.; Wang, C.-S.; Yao, C.; Shen, Q.;
Lu, L. Org. Lett. 2014, 16 (21), 5544−5547. (d) Xu, W.; Sun, H.;
Xiong, Z.; Li, X. Organometallics 2013, 32 (23), 7122−7132.
(e) Keyes, L.; Love, J. A. RSC Catalysis Series 2013, 11, 159−192.
(f) Fischer, P.; Goetz, K.; Eichhorn, A.; Radius, U. Organometallics
2012, 31 (4), 1374−1383. (g) Breyer, D.; Berger, J.; Braun, T.; Mebs,
S. J. Fluorine Chem. 2012, 143, 263−271. (h) Sun, A. D.; Love, J. A.
Dalton Trans. 2010, 39 (43), 10362−10374. (i) Johnson, S. A.; Huff,
C. W.; Mustafa, F.; Saliba, M. J. Am. Chem. Soc. 2008, 130 (51),
17278−17280. (j) Schaub, T.; Backes, M.; Radius, U. J. Am. Chem. Soc.
2006, 128, 15964. (k) Vela, J.; Smith, J. M.; Yu, Y.; Ketterer, N. A.;
Fachenriem, C. J.; Lachicotte, R. J.; Holland, P. L. J. Am. Chem. Soc.
2005, 127, 7857−7870. (l) Schaub, T.; Radius, U. Chem. - Eur. J. 2005,
11, 5024−5030. (m) Dankwardt, J. W. J. Organomet. Chem. 2005, 690
(4), 932−938. (n) Braun, T.; Perutz, R. N. Chem. Commun. 2002,
No. 23, 2749−2757.
(21) Cundy, C. J. Organomet. Chem. 1974, 69 (2), 305−310.
(5) (a) Johnson, S. A.; Hatnean, J. A.; Doster, M. E. Prog. Inorg.
Chem. 2011, 255, 255. (b) Ohashi, M.; Shibata, M.; Saijo, H.;
Kambara, T.; Ogoshi, S. Organometallics 2013, 32 (13), 3631−3639.
(6) Fahey, D. R.; Mahan, J. E. J. Am. Chem. Soc. 1977, 99 (8), 2501−
8.
(7) (a) Nova, A.; Reinhold, M.; Perutz, R. N.; MacGregor, S. A.;
McGrady, J. E. Organometallics 2010, 29 (7), 1824−1831. (b) Hatnean,
J. A.; Johnson, S. A. Organometallics 2012, 31 (4), 1361−1373.
(c) Cronin, L.; Higgitt, C. L.; Karch, R.; Perutz, R. N. Organometallics
1997, 16 (22), 4920−4928.
(8) Johnson, S. A.; Mroz, N. M.; Valdizon, R. V.; Murray, S.
Organometallics 2011, 30 (3), 441−457.
(9) Schaub, T.; Fischer, P.; Steffen, A.; Braun, T.; Radius, U.; Mix, A.
J. Am. Chem. Soc. 2008, 130 (29), 9304−9317.
(10) Doster, M. E.; Johnson, S. A. Angew. Chem., Int. Ed. 2009, 48
(12), 2185−2187.
(11) (a) Sun, A. D.; Leung, K.; Restivo, A. D.; La Berge, N. A.;
Takasaki, H.; Love, J. A. Chem. - Eur. J. 2014, 20 (11), 3162−3168.
(b) Zhu, F.; Wang, Z.-X. J. Org. Chem. 2014, 79 (10), 4285−4292.
(c) Yang, X.; Sun, H.; Zhang, S.; Li, X. J. Organomet. Chem. 2013, 723,
36−42. (d) Yoshikai, N.; Mashima, H.; Nakamura, E. J. Am. Chem. Soc.
2005, 127 (51), 17978−17979. (e) Steffen, A.; Sladek, M. I.; Braun,
T.; Neumann, B.; Stammler, H.-G. Organometallics 2005, 24, 4057.
́ ́
(f) Mongin, F.; Mojovic, L.; Guillamet, B.; Trecourt, F.; Queguiner, G.
J. Org. Chem. 2002, 67 (25), 8991−8994. (g) Bohm, V. P. W.;
Gstottmayr, C. W. K.; Weskamp, T.; Herrmann, W. A. Angew. Chem.,
Int. Ed. 2001, 40 (18), 3387−3389.
(12) Tamao, K.; Sumitani, K.; Kumada, M. J. Am. Chem. Soc. 1972,
94 (12), 4374−6.
(13) Nakamura, Y.; Yoshikai, N.; Ilies, L.; Nakamura, E. Org. Lett.
2012, 14 (13), 3316−3319.
K
DOI: 10.1021/acs.organomet.7b00165
Organometallics XXXX, XXX, XXX−XXX