Mechanistic insight into carbon-fluorine cleavage with a (iPr3P)2Ni source: Characterization of (iPr3P)2NiC6F5 as a significant Ni(I) byproduct in the activation of C6F6

Article abstract of DOI:10.1016/j.ica.2014.07.023

The reaction of (ⁱPr₃P)₂Ni sources such as the anthracene adduct, (ⁱPr₃P)₂Ni(η²-C₁₄H₁₀), with hexafluorobenzene provided the C-F activation product trans-(ⁱPr₃P)₂NiF(C₆F₅) (2) and the unexpected Ni(I) complex (ⁱPr₃P)₂Ni(C₆F₅) (3). The observation of 3 supports a radical pathway for C-F activation, despite the observation of the mononuclear adduct (ⁱPr₃P)₂Ni(η²-C₆F₆) (4) commonly associated with concerted or phosphine-assisted oxidative addition. Carbon-fluorine activation reactionsof pentafluorobenzene and all the isomers of tetrafluorobenzene and trifluorobenzene with (ⁱPr₃P)₂Ni(η²-C₁₄H₁₀) were also examined. These reacted with varying selectivity and yield. It was found that 1,2,3,4-tetrafluorobenzene reacted selectively at the 2-position to provide a fully characterized C-F activation product, whereas 1,2,4,5-tetrafluorobenzene underwent a hydrodefluorination reaction with none of the expected C-F activation product.

Full text of DOI:10.1016/j.ica.2014.07.023

Inorganica Chimica Acta xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Mechanistic insight into carbon–fluorine cleavage with a (iPr3P)2Ni
source: Characterization of (iPr3P)2NiC6F5 as a significant Ni(I) byproduct
in the activation of C6F6
⇑
Jillian A. Hatnean, Manar Shoshani, Samuel A. Johnson
Department of Chemistry & Biochemistry, University of Windsor, Windsor, ON N9B 3P4, Canada
a r t i c l e i n f o
a b s t r a c t
The reaction of (iPr3P)2Ni sources such as the anthracene adduct, (iPr3P)2Ni( 2-C14H10), with hexafluoro-
g
Article history:
benzene provided the CꢀF activation product trans-(iPr3P)2NiF(C6F5) (2) and the unexpected Ni(I) com-
Received 5 May 2014
Received in revised form 3 July 2014
Accepted 7 July 2014
Available online xxxx
plex (iPr3P)2Ni(C6F5) (3). The observation of 3 supports a radical pathway for CꢀF activation, despite
the observation of the mononuclear adduct (iPr3P)2Ni( 2-C6F6) (4) commonly associated with concerted
g
or phosphine-assisted oxidative addition. Carbon–fluorine activation reactions of pentafluorobenzene
and all the isomers of tetrafluorobenzene and trifluorobenzene with (iPr3P)2Ni( 2-C14H10) were also
g
Dedicated to Don Tilley
examined. These reacted with varying selectivity and yield. It was found that 1,2,3,4-tetrafluorobenzene
reacted selectively at the 2-position to provide a fully characterized CꢀF activation product, whereas
1,2,4,5-tetrafluorobenzene underwent a hydrodefluorination reaction with none of the expected CꢀF
activation product.
Keywords:
C–F activation
Nickel complexes
Phosphines
Organometallic Ni(I)
Oxidative addition
Fluorinated arenes
Ó 2014 Elsevier B.V. All rights reserved.
1. Introduction
substrates are of greater interest as substituents in applications
such as pharmaceuticals.
The oxidative addition of CꢀF bonds to transition metal
complexes is a challenging bond activation that has potential
application in catalytic coupling reactions using CꢀF bonds to
produce partially fluorinated organics for applications such as
pharmaceuticals and agrochemicals [1]. Nickel complexes have
been suggested as optimal catalysts for transformations that
involve C–F activation for a variety of reasons, including modest
cost and unique selectivity [2]. Numerous examples of nickel com-
plexes that activate CꢀF bonds in the most reactive substrates,
such as perfluorinated arenes [2b,3], pyridines [2b,3f,4], and
pyrimidines [5], are known [6]. In contrast, the selective activation
of C–F bonds in partially fluorinated arenes such as the tetra-, tri-
and difluoroarenes by nickel complexes has been achieved only
rarely with the use of phosphine [7], N-heterocyclic carbene
[3f,4d], and nitrogen donor ancillary ligands [3c]. With a few
exceptions [8] the catalytic functionalization of aromatic C–F
bonds via mechanisms that conceivably involve oxidative addition
to Ni catalysts has been largely limited to substrates that are highly
fluorinated [1,9]; this is undesirable, given that less fluorinated
If the mechanism of CꢀF bond activation is different for the
more electron-poor aromatics, this could cause difficulties in the
design of general C–F functionalization catalysts, which would
need to be capable of reacting with substrates with differing
degrees of fluorination. Although DFT calculations have been done
to examine both traditional concerted oxidative addition [8a] and
phosphine-assisted pathways of CꢀF activation [10], radical path-
ways involving Ni(I) intermediates are rarely considered computa-
tionally, despite precedent in the oxidative addition of other
carbon–halide bonds to Ni [11]. Experimentally [12], it has been
shown that 2,3,5,6-tetrafluoropyridine reacts with a (Et3P)2Ni
source via a concerted CꢀF oxidative addition without phos-
phine-assistance; however, the more electron-poor pentafluoro-
pyridine reacts via a mechanism that is zeroth order with respect
to the mononuclear (PEt3)2Ni adduct and with an EPR active
species as a likely intermediate. These observations are suggestive
of a radical mechanism. Studies of CꢀF activation have shown
unusual products with highly-fluorinated arenes that may be
indicative of radical pathways, though alternate mechanisms
cannot be ruled out, and the clear identification of radical interme-
diates has not been possible [7,13].
⇑
Corresponding author. Fax: +1 (519) 973 7098.
E-mail address: sjohnson@uwindsor.ca (S.A. Johnson).
http://dx.doi.org/10.1016/j.ica.2014.07.023
0020-1693/Ó 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

2
J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
In previous work [14] we have shown that the anthracene
adduct (iPr3P)2Ni( 2-C14H10) (1) can successfully C–H activate
be present in the reaction of excess C6F6 with (Et3P)4Ni [2b,15].
g
As expected, the exact chemical shifts of paramagnetic 3 are both
strongly temperature and field dependent. The 31P{1H} NMR of
the products in pentane showed a doublet for 2 at d 28.1 (d,
2JPF = 40 Hz), but no observable resonance for 3, as anticipated
given the proximity of the phosphine to the metal-centered radical.
The 1H NMR spectrum has sharp peaks for 2 at the typical expected
diamagnetic values of d 1.18 and 1.70 for the isopropyl group,
whereas broad peaks were observed for 3 at d 4.8 and 11.1 in a
6:1 ratio, and comparable to the 1H NMR spectra for the isopropyl
substituents of other paramagnetic (iPr3P)2NiX species (where
X = Cl, Br and I) [16]. The 19F{1H} NMR spectra revealed three res-
onances for 2 at d ꢀ109.1, ꢀ160.7, and ꢀ163.6 corresponding to
two ortho-fluorines, para-fluorine, and two meta-fluorines, respec-
tively. The resonances associated with the ortho and meta-fluorines
are second order multiplets of an AA0MM0N spin system, while the
para-fluorine is a triplet of triplets, with coupling to two meta-flu-
partially fluorinated aromatics in high yields with occasional
impurities attributed to C–F activation compounds. In general
these nickel-hydrides are stable at room temperature for days.
The hydride produced via the reaction of 1 and 1,2,4,5-C6F4H2
was stable even under reflux in toluene. The bulky phosphine
ancillary ligands better promote trans oxidative addition, thereby
reducing the possibility of isolating or observing any mononuclear
or dinuclear arene adducts. Herein, we will show that 1 is also
capable of C–F bond oxidative addition with perfluorinated and
partially fluorinated substrates, and that the products in some
cases are suggestive of radical pathways.
2. Results and discussion
2.1. Ni(I) and Ni(II) C–F bond activation products with C6F6
After 2 h, the reaction of the anthracene adduct, (iPr3P)2
3
orines with a JFF value of 20.4 Hz and two ortho-fluorines with a
4JFF value of 2.3 Hz. The resonance for the nickel-fluoride is at d
2
Ni(g
2-C14H10) (1) with hexafluorobenzene in pentane produced a
ꢀ409.9 and is a triplet of triplets, with a JPF value of 40 Hz and a
4JFF value of 9.8 Hz. The observed coupling constants are consistent
with the previous spectroscopic characterization of trans-(Et3P)2
NiF(C6F5) [2b]. The 19F NMR spectrum also revealed a single broad
resonance that typically appeared in the range of d ꢀ117 to ꢀ123
for 3. This peak has a peak width at half-height of 300 Hz and is
likely due to the para-F of 3. Compared to a typical para-F of a
Ni-C6F5 group it is shifted by about 45 ppm. The failure to observe
the ortho- and meta-F groups by NMR is not surprising; DFT calcu-
lations show that the ortho- and meta-F should feature Fermi con-
tact coupling to the unpaired spin density that is 100 and 35 times
greater than that of the para-F, respectively. The relative NMR
yields of 2 and 3 are 49% and 22%, which is suggestive of a third
Ni species being made in the reaction in equal amounts to 3, to
account for all the Ni, but all attempts to identify this product have
failed. Although (iPr3P)2NiF might be expected as a concomitant
product with 3, no species with sufficient intensity peaks was
observed in the solution spectra, and no precipitate was observed;
it is notable that although (iPr3P)2NiX species (where X = Cl, Br, and
I) all give similar paramagnetically shifted 1H NMR spectra, previ-
ous attempts to prepare (iPr3P)2NiF as well as trans-(iPr3P)2NiF2
failed [16].
Crystallization of the pentane soluble products at ꢀ40 °C gave a
precipitate that contained both 2 and 3, which were easy to distin-
guish under a microscope due to their distinct crystal habits and
colors. Compound 2 provided large yellow cubes, whereas com-
pound 3 was pale yellow. Both were suitable for structural analysis
by X-ray crystallography, and ORTEPs of the solid-state molecular
structures of 2 and 3 are shown in Fig. 1. X-ray crystallographic col-
lection and refinement parameters are included in the Supporting
Information. Complex 2 has trans-disposed triisopropylphosphine
ligands and a pseudo-square planar geometry around the nickel
metal. This compound crystallized with half an equivalent of
anthracene in the lattice. The pentafluorobenzene moiety is nearly
orthogonal to the P–Ni–P plane. The P(1)–Ni(1)–P(2) angle of
165.57(5)° is attributed to the relief of steric crowding by the bulky
triisopropylphosphine ligands; the phosphine donors bend away
from the aromatic ring and towards the smaller fluoride ligand.
The Ni(1)–F(1) bond distance of 1.842(3) Å is within the range
expected [2b,3c,3f,4d]. Compound 3 has planar geometry around
the nickel metal, and similar to 2 the pentafluorobenzene moiety
is twisted out of the P–Ni–P plane. The P(1)–Ni(1)–P(2) angle of
145.22(2)° has decreased significantly in the absence of the
nickel-fluoride to relieve steric crowding by the bulky triisopropyl-
phosphine ligands. Complex 2 is easily distinguished from the
related Ni(II) hydride. For example, in comparison trans-(iPr3P)2
NiH(2,3,5,6-C6F4H) has a P–Ni–P angle of 155.98(2)° which is about
mixture of the anticipated CꢀF bond oxidative addition product,
trans-(iPr3P)2NiF(C6F5) (2) and the unexpected Ni(I) complex
(iPr3P)2Ni(C6F5) (3), as shown in Scheme 1. This reaction is consid-
erably faster than the reaction of (Et3P)4Ni with C6F6, which reacts
over weeks at room temperature. Unlike sources of the smaller
diphosphine nickel fragment (Et3P)2Ni, which reacts rapidly to pro-
vide mononuclear and dinuclear adducts of C6F6, [2a,7,13] no
adducts were observed by multinuclear NMR spectroscopy in this
reaction, though an NMR scale reaction using the dinitrogen com-
plex [(iPr3P)2Ni]2N2 as an alternate (iPr3P)2Ni source revealed a
short-lived intermediate assigned as (iPr3P)2Ni( 2-C6F6) (4), as
g
shown on the bottom of Scheme 1. As with many (R3P)2Ni-arene
complexes, 4 features rapid ring whizzing in solution, as suggested
a septet in the 31P{1H} NMR at d 42.5 and a triplet in the 19F NMR
3
with an identical JPF of 8.3 Hz, which is comparable to the struc-
turally characterized [bis(ditertbutylphosphino)ethane]Ni(
[3a]. The rate of CꢀF bond activation to provide 2 and 3 was compa-
g
2-C6F6)
rable to the rate of formation of (iPr3P)2Ni( 2-C6F6), and so 4 was not
g
isolated. The addition of Ph3SiF as a 1H and 19F NMR internal standard
allowed for the determination of NMR yields of 49% for 2 and 22% for
3. The ratio of 2 to 3 does not change over the course of the reaction,
suggesting they are formed concurrently. Complex 3 is unreactive
towards excess C6F6, as well as to C6F5H.
Analysis of the crude reaction mixture by 1H, 31P{1H} and
19F{1H} NMR spectroscopy did not reveal any significant impuri-
ties, such as difluorophosphoranes, which have been reported to
F
F
+ C6F6
25 °C
2 h
F
F
F
F
F
F
F
pentane
+
F
Ni
–C14H10
iPr3P
iPr3P Ni PiPr3
Ni
PiPr3
iPr3P
PiPr3
F
2
3
1
(iPr3P)2Ni(η2-C14H10
)
(iPr3P)2NiF(C6F5) (iPr3P)2Ni(C6F5)
49% NMR yield
paramagnetic
22% NMR yield
1 h
+ C6F6
25 °C
pentane
0.5 [(iPr3P)2Ni]2(N2)
(iPr3P)2Ni(η2-C6F6)
4
Scheme 1. Intermediates and products in the reaction of (iPr3)2Ni sources with
C6F6.
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
3
Fig. 2. Experimental and modeled 77 K 9.422 GHz EPR spectrum of a powdered
sample of crystalline paramagnetic 3 (also containing EPR inactive 2).
+C6F5MgI
F
(iPr3P)2NiCl
(iPr3P)2NiCl
F
F
F
+Bu3SnC6F5
F
X
Ni
iPr3P
PiPr3
1% Na/Hg
2
3
0.5 [(iPr3P)2Ni]2N2
comproportionation
(iPr3P)2Ni(C6F5)2
Scheme 2. Attempted routes to 3.
NMR and EPR spectroscopy. Attempts to react the Ni(II) complex
2 with reducing agents such as 1% Na/Hg, excess 1, or [(iPr3P)2Ni]2-
N2 also failed; the failure of these latter reactions support that 3 is
not generated from 2 in the reaction of 1 with C6F6. In contrast,
(iPr3P)2Ni(C6F5)2, prepared from the reaction of (iPr3P)2NiCl2 with
C6F5MgBr, does undergo comproportionation with [(iPr3P)2Ni]2N2
over the course of hours to provide 3.
Fig. 1. (a) ORTEP of the solid-state molecular structure of 2 as determined by X-ray
crystallography. Hydrogen atoms and 0.5 equiv of co-crystallized anthracene are
omitted for clarity. The 30% probability ellipsoids are shown. Selected bond lengths
(Å): Ni(1)–F(1), 1.842(3); Ni(1)–P(1), 2.2426(14); Ni(1)–P(2), 2.2521(14);
Ni(1)–C(1), 1.888(5); F(2)–C(2), 1.360(6); F(3)–C(3), 1.354(5); F(4)–C(4), 1.352(5);
F(5)–C(5), 1.345(6); F(6)–C(6), 1.371(5). Selected bond angles (°): P(1)–Ni(1)–P(2),
165.57(5); C(1)–Ni(1)–F(1), 177.73(16). (b) ORTEP of the solid-state molecular
structure of 3 as determined by X-ray crystallography. Hydrogen atoms are omitted
for clarity. The 30% probability ellipsoids are shown. Selected bond lengths (Å):
Ni(1)–P(1), 2.2431(5); Ni(1)–P(2), 2.2329(5); Ni(1)–C(1), 1.973(2). Selected bond
angles (°): P(1)–Ni(1)–P(2), 145.22(2); C(1)–Ni(1)–P(1), 105.84(6); C(1)–Ni(1)–P(2),
108.85(6).
Organometallic Ni(I) species featuring
r-bound carbyl frag-
ments are rare, and are typically only isolated from a combination
of sterically bulky and/or chelating ligands. Examples include the
2-coordinate Ni(I) complex (IPr)Ni[CH(SiMe3)2] [18], three coordi-
nate [bis(ditertbutyl)phosphinoethane]NiR, where R = CHCMe3,
CH2SiMe3 and CH 2CMe2Ph [17]. An early example of a three-
coordinate
r-bound Ni(I) carbyl compound was prepared from
the reaction (PPh3)2NiCl2 with Li{C(SiMe3)2(SiMe2C5H4N-2)},
which presumably acts as both a source of the chelate and a reduc-
ing agent [19]. It is known that nickel(I) halides are intermediates
in oxidative addition reactions of aryl halides with nickel(0) com-
plexes [3d,11], and that in some cases these Ni(I) halides can be
significant reaction impurities. The typical fate of the carbyl radical
in the reactions that produce these impurities is to abstract a
hydrogen atom from the solvent. In the formation of 3 it is likely
that the C6F5 radical generated from electron transfer from (iPr3P)2
Ni to C6F6 followed by loss of Fꢀ survives and reacts with (iPr3P)2Ni.
Complex 3 is observed since it is stable with respect to dispropor-
tionation to Ni(II) and Ni(0) products.
The reaction of 1 with C6F6 to generate an isolable Ni(I) byprod-
uct appears unique to this substrate, though some unusual prod-
ucts are observed in the reaction of other substrates. Scheme 3
shows the reaction products obtained from reaction of 1 with pen-
tafluorobenzene and the three tetrafluorobenzene isomers. With
the exception of 1,2,3,4-tetrafluorobenzene these reactions were
not clean enough to isolate pure products, though the use of 1H,
31P and 19F NMR spectroscopy allows for unambiguous product
assignments.
10° degrees larger than in 2 [14]. The Ni(1)–C(1) bond length in 3 is
1.973(2) Å, which is 0.085(5) Å longer in comparison to compound
2. In contrast, the NiꢀP bond lengths are shorter on average for the
Ni(I) complex 3 versus 2, but by less than 0.01 Å.
The 77 K EPR spectrum of the isolated crystalline solid contain-
ing 2 and 3 is shown in Fig. 2. The spectrum shows a non-axially
symmetric environment at Ni, with three transitions correspond-
ing to gx, gy, and gz values of 2.388, 2.250, and 2.145, respectively.
The gy signal displays resolvable hyperfine coupling. A reasonable
fit was obtained when the gy signal was simulated as an overlap-
ping triplet of triplets, with hyperfine coupling constants of 80 G
and 75 G from coupling to two 31P nuclei and the two ortho 19F
nuclei. The magnitude of coupling to 31P is comparable to related
complexes [17]. The in situ reaction of 1 with C6F6 was monitored
by low-temperature EPR spectroscopy and yielded similar spectra.
A number of attempts to synthesize 3 from other routes were
attempted, as shown in Scheme 2. The product mixtures from
the reaction of (iPr3P)2NiCl with either C6F5MgI or Bu3SnC6F5 failed
to reveal any detectable amount of 3, as analyzed by 1H and 19F
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

4
J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
2 weeks
25 °C
iPr3P Ni PiPr3
iPr3P Ni PiPr3
+
iPr3P Ni PiPr3
1 week
pentane
–C14H10
F
H
F
5
6
7
major product (85 %)
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
F
+
iPr3P Ni PiPr3
iPr3P Ni PiPr3
iPr3P Ni PiPr3
(iPr3P)2Ni(η2-C14H10
)
+
F
1 week
pentane
–C14H10
1
H
F
9
F
56
6
8
not observed!!!
F
F
F
F
F
F
F
F
F
F
iPr3P
Ni
PiPr3
iPr3P
Ni
PiPr3
+
36 h
toluene
F
11
F
10
–C14H10
> 95 % yield by NMR
81 % yield isolated
F
F
F
F
F
F
F
F
F
F
F
+
iPr3P Ni PiPr3
iPr3P Ni PiPr3
2 weeks
toluene
50 °C
H
F
12
13
–C14H10
Scheme 3. Reactions of 1 with pentafluorobenzene and the tetrafluorobenzenes.
2.2. Pentafluorobenzene activation
true only for reactions performed in aromatic solvents. If the reac-
tion is performed in non-aromatic solvents, such as pentane, the
outcome is more complicated. After 6 h, the expected reaction time
for complete hydride formation, analysis of the 19F{1H} NMR
spectrum reveals the presence of several fluorine containing
compounds. The major product is the C–H activation product, 8,
with the additional resonances assigned as the hydrodefluorina-
tion products, trans-(iPr3P)2NiF(2,3,5,6-C6F4H) (6) and 1,2,4-
C6F3H3, as shown in Scheme 3. Note that the C–F bond oxidative
addition product, trans-(iPr3P)2NiF(2,4,5-C6F3H2) (9), is not
observed. While it is not possible at this point to determine the
exact mechanism by which this occurs, both a hydride/fluoride
ligand exchange or a radical CꢀF bond activation mechanism are
plausible. Similar unexpected products have been reported using
PEt3 as the ancillary ligand [2c].
It has been shown that the reaction of the anthracene adduct 1
with C6F5H provides the C–H activated complex trans-(iPr3P)2
NiH(C6F5) (5) in high spectroscopic yields, although it is not isola-
ble [14]. Analysis of the crude reaction mixture showed there were
less than 5% of other impurities that could be identified as the C–F
activation products: the para-F activation compound trans-(iPr3P)2
NiF(2,3,5,6-C6F4H) (6), and the ortho-F activation product trans-
(iPr3P)2NiF(2,3,4,5-C6F4H) (7) in a ratio of 6:1. Compound 5 isomer-
izes to 6 and 7 over weeks at room temperature, as shown in
Scheme 3. Unlike the activation of pentafluorobenzene with
(Et3P)2Ni sources, which strongly favored activation at the 1-posi-
tion [7], the major product is the para-fluorine activation compound,
which is also the expected site for both radical mechanisms and
nucleophilic attack [3c,f,20]. These products can be synthesized
directly via the in situ reaction of 2 with C6F5H heated for 72 h at
50 °C, with similar selectivity. No Ni(I) complexes were observed
by either 1H or 19F NMR in these reactions.
2.4. 1,2,3,4-Tetrafluorobenzene activation
To date, there are a limited number of examples reported in the
literature involving products of C–F bond oxidative addition of the
tetrafluorobenzenes by nickel complexes [3c,13]. The reaction of 1
with 1,2,3,4-C6F4H2 in toluene results in an immediate color
change from red to orange. Analysis of the crude solution by
NMR spectroscopy reveals that the C–F activated product, trans-
(iPr3P)2NiF(2,3,6-C6F3H2) (10) formed over 36 h at room tempera-
ture in greater than 95% NMR selectivity, as analyzed by 19F
NMR. The only significant impurity present was less than 5% of
the C–F oxidative addition product at the 1-site, trans-(iPr3P)2
NiF(2,3,4-C6F3H2) (11). No adducts were observed. Activation at
2.3. 1,2,4,5-Tetrafluorobenzene activation
As previously noted in the literature, the reaction of 1 with
1,2,4,5-C6F4H2 proved to be the most successful in synthesizing
an isolable C–H bond activation product, trans-(iPr3P)2
NiH(2,3,5,6-C6F4H) (8), with no observable impurities and was
remarkably stable up to 100 °C in toluene [14]. Solutions of 8 could
stand for weeks without any spectroscopic evidence of isomeriza-
tion to the thermodynamic C–F bond activation product. This holds
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
5
2.5. 1,2,3,5-Tetrafluorobenzene activation
We have previously shown that the reaction of 1 with 1,2,3,5-
C6F4H2 yielded the isolable C–H oxidative addition product com-
plex trans-(iPr3P)2NiH(2,3,4,6-C6F4H) (12). Analysis of the crude
reaction mixture showed there was less than 2% of an impurity
identified as the C–F activation product, trans-(iPr3P)2NiF(2,3,5-
C6F3H2) (13). This compound can be directly synthesized via the
in situ reaction of 1 with 1,2,3,5-C6F4H2 heated for 2 weeks at
50 °C in toluene. Spectroscopic analysis reveals that this reaction
is selective for activation at the 1-site, and 13 is present as the
major product; however, compound 12 remained in 30% yield by
NMR spectroscopy, as well as a dark precipitate attributed to Ni
metal in the NMR tube. Compound 13 was identified by 19F{1H}
NMR spectroscopy by three multiplets at d ꢀ113.8, ꢀ120.0,
ꢀ138.0, and the fluoride located at ꢀ387.9 which is a triplet of
2
doublets with coupling to two 31P nuclei with a JFP value of
4
39.8 Hz, and a single ortho-fluorine with a JFF value of 11.5 Hz. A
complimentary doublet was observed in the 31P{1H} NMR spectra.
Fig. 3. ORTEP depiction of the solid-state molecular structure of 10 as determined
by X-ray crystallography. Hydrogen atoms are omitted for clarity. The 30%
probability ellipsoids are shown. Selected bond lengths (Å): Ni(1)–F(1),
1.8427(12); Ni(1)–P(1), 2.322(6); Ni(1)–P(2), 2.2319(6); Ni(1)–C(1), 1.883(2);
F(2)–C(2), 1.349(3); F(3)–C(3), 1.357(3); F(4)–C(6), 1.359(3). Selected bond angles
(°): P(1)–Ni(1)–P(2), 171.80(2); C(1)–Ni(1)–F(1), 175.60(9).
2.6. Characterization of C–F bond activation products with C6F3H3
The successful selective C–F activation of trifluorobenzene and
its derivatives by nickel complexes with the N-heterocyclic car-
bene ancillary ligand, [iPr2Im], (where [iPr2Im] = 1,3-di(isopro-
pyl)imidazole-2-ylidene), has been reported [3f]. Carbene ligands
have donor properties that are reminiscent of the most electron-
donating trialkylphosphines; therefore it is reasonable to expect
that analogous reactions utilizing the anthracene adduct, (iPr3P)2-
the 2-position is also the expected product for both radical
mechanisms and nucleophilic attack [21]. Complex 10 was easily
isolated by removal of solvent under vacuum, extraction into
pentane, removal of the insoluble anthracene byproduct by
filtration, and crystallization at ꢀ40 °C. Single crystals suitable
for structural analysis of 10 by X-ray crystallography were directly
obtained from this synthesis. An ORTEP depiction of the solid-state
molecular structure is shown in Fig. 3, and X-ray crystallographic
collection and refinement parameters are included in the Support-
ing Information. The nickel-fluoride bond distance of 1.8427(12) Å
is comparable to other C–F activated compounds found in litera-
ture that contain two ortho-fluorines on the aromatic backbone
[2b,3c,3f,4d].
Ni(g
2-C14H10) (1) with a phosphine ancillary ligand could produce
similar results. Although it proved possible to determine the selec-
tivity of the (iPr3P)2Ni moiety in CꢀF activation, as shown in
Scheme 4, in general the NMR yields and selectivity were poor,
and so none of these compounds were isolated.
A toluene solution of 1 and 1,2,3-C6F3H3 on an NMR scale reacts
slowly to produce the C–F activation products, trans-(iPr3P)2-
NiF(2,3-C6F2H3) (14) and trans-(iPr3P)2NiF(2,6-C6F2H3) (15) in low
yields. At the end of one week, the reaction had still not gone to
completion; however a dark precipitate of nickel metal was pres-
ent in the reaction tube. Compound 14 is present as the major
The structure of complex 10 was further confirmed through
characterization via multinuclear NMR spectroscopy and elemen-
tal analysis. The room temperature 1H NMR spectra in C6D6
revealed two aromatic proton environments at d 6.18 and 6.40,
with the former resonance being broad and slightly overlapped
with the protons of the second isomer, trans-(iPr3P)2NiF(2,3,4-
C6F3H2) (11). The 31P{1H} NMR revealed a doublet at d 26.8 for
the major product activated at the 2-position (10), and a second
F
F
F
F
F
F
F
PiPr3
+
iPr3P
Ni
iPr3P
Ni
PiPr3
1 week
toluene
–C14H10
2
smaller doublet belonging to compound 11, both with JPF values
F
14
F
15
of 40.3 Hz, which is typical for bis(phosphine)nickel-fluoride com-
plexes [2a].
18 % NMR yield
F
10 % NMR yield
The 19F{1H} NMR spectrum shows three aromatic resonances
for each isomer. Compound 10 displays three resonances at d
ꢀ88.3, ꢀ106.2, and ꢀ146.4, where the two resonances downfield
are the ortho-fluorine substituents. The nickel-fluoride is shifted
upfield, as is anticipated for a nickel-fluoride complex [22], to d
ꢀ401.8, and is a triplet of doublets of doublets. This can be mod-
F
F
F
F
iPr3P
Ni
PiPr3
(iPr3P)2Ni(η2-C14H10
)
1 week
toluene
–C14H10
1
F
16
2
eled as coupling to two 31P nuclei with a JFP value of 40.3 Hz
4
and two ortho-fluorines with JFF values of 9.2 and 9.1 Hz. The
F
F
observed coupling constants are consistent with the previous spec-
troscopic characterization of nickel-fluorides [2b,3c,7]. Compound
11 (activation at the 1-position) displays three resonances at d
ꢀ106.5, ꢀ147.1, and ꢀ165.7 for the three fluorines on the aromatic
backbone. The upfield shift of the nickel-fluoride at d ꢀ387.6 can
be modeled as a triplet of doublets with coupling to two 31P nuclei
F
F
F
iPr3P
Ni
PiPr3
1 week
toluene
–C14H10
F
17
2
4
with a JFP value of 41.3 Hz and a single ortho-fluorine with a JFF
value of 11.2 Hz.
Scheme 4. Reactions of the trifluorobenzenes with 1.
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

6
J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
product, with activation at the 1-position being the favored site for
nucleophilic attack. Similar results were obtained by Radius and
co-workers with the N-heterocyclic carbene ancillary ligand,
[iPr2Im] [3f]. Compound 14 was identified by two peaks in the
19F{1H} NMR spectra at d ꢀ108.8 and ꢀ141.6. The resonances at
d ꢀ108.8 and ꢀ141.6 are doublets, coupled to a single 19F with a
3JFF value of 31.1 Hz. The nickel-fluoride was located in the fluoride
region at d ꢀ383.3 and is a triplet of doublets, coupled to two
to yield the expected compound, trans-(iPr3P)2NiF(2,4,5-C6F3H2),
and instead produced the hydrodefluorination products, trans-
(iPr3P)2NiF(2,3,5,6-C6F4H) and 1,2,4-C6F3H3. This reactivity once
again may be indicative of a radical pathway for CꢀF activation,
though this is difficult to discern from a hydride/fluoride ligand
exchange mechanism. The trifluorobenzenes underwent CꢀF bond
activation in poor spectroscopic yields; only 1,2,3-C6F3H2 under-
went unselective C–F bond activation. Unlike C6F6, Ni(I) complexes
were not identified for these substrates.
2
equivalent 31P nuclei with a JPF value of 40.2 Hz and one ortho-
4
fluorine by a JFF value of 11 Hz. Compound 15 displayed a single
resonance in the aromatic region of the 19F{1H} NMR spectra at d
ꢀ81.7. The fluoride was a triplet of triplets at d ꢀ397.8, coupled
4. Experimental
2
to two 31P nuclei with a JPF value of 39.4 Hz and two equivalent
4.1. General procedures
19F nuclei with a 4JFF value of 8 Hz. A complimentary set of doublets
in the 31P{1H} NMR spectra were located at d 25.9 and 26.5 for 14
and 15, respectively.
Unless otherwise stated, all manipulations were performed
under an inert atmosphere of nitrogen using either standard
Schlenk techniques or an MBraun glovebox. Dry, oxygen-free sol-
vents were employed throughout. Anhydrous pentane and toluene
were purchased from Aldrich, sparged with dinitrogen, and passed
through activated alumina under a positive pressure of nitrogen
gas; toluene and hexanes were further deoxygenated using a
Grubbs’ type column system [23]. Benzene-d6 was dried by heating
at reflux with Na/K alloy in a sealed vessel under partial pressure
then trap-to-trap distilled and freeze–pump–thaw degassed three
times. Toluene-d8 was purified in an analogous manner by heating
at reflux over Na. 1H, 31P{1H}, 13C{1H}, and 19F{1H} NMR spectra
were recorded on a Bruker Spectrometer operating at either
300 MHz 500 MHz with respect to proton nuclei. All chemical
shifts are recorded in parts per million, and all coupling constants
are reported in hertz. 1H NMR spectra were referenced to residual
protons (C6D5H, d 7.16; C7D7H, d 2.09) with respect to tetramethyl-
silane at d 0.00. 31P{1H} NMR spectra were referenced to external
85% H3PO4 at d 0.00. 13C{1H} NMR spectra were referenced relative
to solvent resonances (C6D6, d 128.0; C7D8, d 20.4). 19F{1H} NMR
spectra were referenced to an external sample of 80% CCl3F in
CDCl3 at d 0.0. EPR spectra were collected using an X-band Bruker
ESR 300E spectrometer. The compounds benzene-d6 and toluene-
d8 were purchased from Cambridge Isotope Laboratory. Hexaflu-
orobenzene, pentafluorobenzene, and all isomers of tetrafluoro-
benzene and trifluorobenzene were purchased from Aldrich and
used as received. Compound 1 and [(iPr3P)2Ni]2N2 were prepared
as described in the literature [14,16].
The reaction of 1 with either 1,2,4- or 1,3,5-trifluorobenzene
produces observable amounts of their respective C–H activation
products, trans-(iPr3P)2NiH(2,3,6-C6F3H2) and trans-(iPr3P)2
NiH(2,4,6-C6F3H2) within 48 h. Continued monitoring of these
reactions on an NMR scale for one week by 19F NMR spectroscopy
revealed small resonances that could be identified as the C–F
activation products, trans-(iPr3P)2NiF(2,5-C6F2H3) (16) and trans-
(iPr3P)2NiF(3,5-C6F2H3) (17) for 1,2,4- and 1,3,5-C6F3H3, respec-
tively. At the end of one week, the reaction mixtures contained
very little C–F activation product, and began to decompose. Heat-
ing the reaction mixtures also resulted in decomposition to nickel
metal. The 19F{1H} NMR spectrum showed two doublets in the
aromatic region for compound 16 at d ꢀ90.5 and ꢀ123.1, with a
five-bond coupling value of 21.2 Hz, which is within the range
expected for para-fluorine coupling [22]. The fluoride region
revealed a triplet of doublets at d ꢀ383.8, coupled to two equiva-
2
lent 31P nuclei with a JPF value of 43 Hz and one ortho-fluorine
4
by a JFF value of 6.5 Hz. Conversely, the analogous C–F activation
product, trans-(iPr3P)2NiF(2,4-C6F2H3) would have produced two
resonances in the aromatic region. However, its JFF value for
meta-fluorine coupling would have been much smaller than the
21 Hz observed for para-fluorine coupling, and therefore, can be
ruled out as a product. Compound 17 was identified in the
19F{1H} NMR spectrum by a singlet at d ꢀ115.4. The nickel-fluoride
was located at d ꢀ378.6 and is a triplet. The 31P{1H} NMR spectrum
displayed a doublet at d 25.1 coupled to the nickel-fluoride with a
2JPF value of 41.4 Hz.
3. Conclusions
4.2. Synthesis, characterization, and reactivity of complexes
The reaction of the anthracene adduct, (iPr3P)2Ni(
g
2-C14H10),
4.2.1. Synthesis of trans-(iPr3P)2NiF(C6F5)ꢁ0.5 C13H9, 2 and
-(iPr3P)2Ni(C6F5), 3
with a variety of perfluorinated and partially fluorinated aromatics
yielded a variety of C–F bond oxidative addition products. The use
of the bulky iPr3P ancillary ligand in the Ni(0) bis(phosphine) moi-
ety with C6F6 drastically improved the reaction time from weeks to
hours with better yields for the C–F activation product trans-(iPr3-
P)2NiF(C6F5). This reaction also yielded the unexpected paramag-
netic nickel(I) species, trans-(iPr3P)2Ni(C6F5), which suggests that
this CꢀF activation has at least some product arising from a radical
pathway, despite the observation of the mononuclear adduct 4 as
an intermediate.
The para-fluorine activation product obtained from the reaction
with C6F5H differed greatly from the ortho-fluorine selectivity
observed with PEt3 as the ancillary ligand [7]. The products
obtained from partially fluorinated substrates, such as tetra- and
trifluorobenzene and its derivatives yielded selectivities in accor-
dance with what is predicted for the kinetic products of nucleo-
philic aromatic substitution, with the exception of 1,2,4,5-C6F4H2.
Carbon–fluorine activation was observed with 1,2,4,5-C6F4H2 in
pentane although not in the predicted manner. The reaction failed
A
pentane solution of the anthracene adduct, (iPr3P)2
Ni(g ) (1) (500 mg, 0.9 mmol) and hexafluorobenzene
2-C14H10
(167 mg, 0.9 mmol) were stirred for 2 h. The precipitated anthra-
cene was filtered, and the filtrate cooled to ꢀ40 °C. Yellow X-ray
quality crystals of the products coprecipitated from solution, were
filtered off and dried in vacuo (405 mg, isolated). The crystalline
sample obtained in this manner contained 0.5 equiv of cocrystal-
lized anthracene associated with 2 and pale yellow-colored blocks
of the paramagnetic compound, (iPr3P)2Ni(C6F5) (3) which could not
be separated. Compound 2: 1H NMR (C6D6, 298 K, 300.13 MHz): d
i
3
3
1.18 (dd, 36H, Pr–CH3, JHH = 6.6 Hz, JPH = 10.4 Hz); 1.70 (m, 6H,
iPr–CH). 31P{1H} NMR (C6D6, 298 K, 121.5 MHz):
d 28.1 (d,
2JPF = 40 Hz). 13C{1H} NMR (C6D6, 298 K, 75.5 MHz): d 19.3 (s, Pr-
CH3); 23.1 (AMM0 virtual t, Pr-CH, JCP = 9.2 Hz); 112.8 (m, ipso–C);
1
128.1 (m, Ar–C); 133.7 (dm, Ar–C, JCF = 257 Hz); 148.1 (dm, Ar–
C, JCF = 248 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz): d ꢀ109.1
1
(AA0MM0N second order m, 2F, 2,6–Ar–F); ꢀ160.7 (tt, 1F, 4–Ar–F,
4
3JFF = 20.4 Hz, JFF = 2.3 Hz); ꢀ163.6 (AA0MM0N second order m,
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
7
2F, 3,5–Ar–F); ꢀ409.9 (tt, 1F, Ni–F, 2JFP = 40 Hz, 4JFF = 9.8 Hz). Com-
pound 3: 1H NMR (C5H12, 298 K, 500.13 MHz): d 5.1 (broad s, 36H,
NMR (C7D8, 298 K, 282.4 MHz): d ꢀ111.6 (AA0MM0 second order
m, 2F, 2,6–Ar–F); ꢀ142.1 (AA0MM0 second order m, 2F, 3,5–Ar–F);
i
2
4
iPr–CH3); d 11.7 (broad s, 6H, Pr–CH) 19F{1H} NMR (C5H12, 298 K,
ꢀ406.1 (tt, 1F, Ni–F, JFP = 40 Hz, JFF = 6 Hz).Compound 7: 31P{1H}
2
470.5 MHz): d ꢀ121.2 (broad s, 1F, para-F).
NMR (C7D8, 298 K, 121.5 MHz): d 27.0 (dm, JPF = 40 Hz). 19F{1H}
NMR (C7D8, 298 K, 282.4 MHz): d ꢀ109.7 (m, 1F, 2–Ar–F); ꢀ143.2
(m, 2F, 5–Ar–F); ꢀ160.8 (m, 1F, 3–Ar–F); ꢀ166.1 (m, 1F, 4–Ar–F);
4.3. NMR characterization of 4
2
4
ꢀ392.5 (dtm, 1F, Ni–F, JFP = 40 Hz, JFF = 10 Hz).
The dinitrogen complex [(iPr3P)2Ni]2N2 (41 mg, 0.053 mmol)
was dissolved in a solution composed of hexafluorobenzene
(44 mg, 0.24 mmol), the internal standard Ph3SiF (10 mg,
0.0384 mmol) and pentane (500 mg). The bright yellow solution
was identified by NMR spectroscopy to be a mixture of CꢀF activa-
tion products, trans-(iPr3P)2NiF(C6F5) and (iPr3P)2Ni(C6F5) as well as
4.7. Characterization of the reaction of 1,2,4,5-tetrafluorobenzene
with 1
A solution of the anthracene adduct, (iPr3P)2Ni( 2-C14H10) (1)
g
(100 mg, 0.179 mmol) and 1,2,4,5-tetrafluorobenzene (28 mg,
0.179 mmol) were dissolved in pentane, and the reaction was
monitored for 1 week by NMR spectroscopy. The initial reaction
mixture consisted primarily of the starting materials. After 6 h,
the solution was analyzed again by 19F and 31P{1H} NMR spectros-
copy, which revealed the presence of a C–H activation product,
trans-(iPr3P)2NiH(2,3,5,6-C6F4H) (8), and smaller intensity reso-
nances attributed to two fluorine containing compounds: the para
C–F activation product of pentafluorobenzene, trans-(iPr3P)2-
NiF(2,3,5,6-C6F4H) (6), as well as 1,2,4-trifluorobenzene. At the
end of one week, peaks associated with 6 and 1,2,4-trifluoroben-
zene continued to increase in intensity. The expected C–F activa-
tion product, trans-(iPr3P)2NiF(2,4,5-C6F3H2) (9) was not observed.
the short lived intermediate (iPr3P)2Ni( 2-C6F6) (4). After 0.5 h, 4
g
was no longer observed by 19F and 31P{1H} NMR. Integration of
1H and 19F NMR spectra relative to the internal standard are consis-
tent with a 49% conversion to 2 and 22% conversion to 3. Final
crude 1H and 19F NMR are provided in the Supporting Information.
Compound 4: 19F{1H} NMR (C5H12, 298 K, 470.5 MHz): d ꢀ162.4 (t,
3JPF = 8.3 Hz). 31P{1H} (C5H12, 298 K, 202.1 MHz): d 42.5 (septet,
3JPF = 8.3 Hz).
4.4. Synthesis of trans-(iPr3P)2Ni(C6F5)2
A 20 mL solution of C6F5MgCl (4 mmol) in THF was added to a
purple solution of (iPr3P)2NiCl2 (700 mg, 1.56 mmol) in 10 mL of
THF at ꢀ40 °C and then warmed to room temperature. Over the
course of 2 h the solution turned from purple to yellow. The THF
was then removed under vacuum and the leftover precipitate
was extracted with 150 mL of n-pentane, filtered through Celite,
4.8. Synthesis of trans-(iPr3P)2NiF(2,3,6-C6F3H2), 10 and 11
A solution of (1) (400 mg, 0.72 mmol) and 1,2,3,4-tetrafluoro-
benzene (110 mg, 0.72 mmol) were stirred for 36 h in toluene.
The solvent was removed in vacuo, and the resultant solid
extracted with pentane. The precipitated anthracene was filtered,
and the filtrate cooled to ꢀ40 °C. Orange-colored X-ray quality
crystals of the 10, were filtered off and dried in vacuo (313 mg,
81% isolated). Concentration of the mother liquor failed to yield a
suitable second crop. NMR spectroscopic analysis of the crude
reaction revealed a mixture of 10 and <5% of the C–F activation iso-
mer, trans-(iPr3P)2NiF(2,3,4-C6F3H2), 11. Compound 10: 1H NMR
(C6D6, 298 K, 300.13 MHz): d 1.26 and 1.28 (d, 36H, iPr–CH3,
3JHH = 6.3 Hz); 1.83 (m, 6H, iPr–CH); 6.18 (br m, 1H, 5–Ar–H);
and recrystallized at ꢀ40 °C as a yellow powder. 1H NMR (298 K,
3
C6D6, 500 MHz): d 0.85(d of virtual triplets (apparent q), JHP
+
5JHP ꢂ 13.3 Hz JHH ꢂ 6.8 Hz, 36H, P(CH(CH3)2)3) d 1.51 (septet of
3
virtual
triplets,
3JHH = 6.8 Hz,
2JHP + 4JHP ꢂ 7.0 Hz,
6H,
P(CH(CH3)2)3,). 31P{1H} NMR (298 K, THF, 121.5 MHz): d 25.4 (s,
2P) 19F{1H} NMR (298 K, THF, 470.5 MHz): d ꢀ108.9 (2nd order
3
multiplet, o-F, 4F) d ꢀ162.0 (t, p-F, JFF = 19.8 Hz, 2F) d ꢀ164.5
(2nd order multiplet, m-F, 4F).
4.5. Synthesis of 3 from comproportionation of (iPr3P)2Ni(C6F5)2
with [(iPr3P)2Ni]2N2
3
3
4
6.40 (dddd, 1H, 4–Ar–H, JHF = 17.7 Hz, JHH = 8.6 Hz, JHF = 5.4 Hz,
4JHF = 4.7 Hz). 31P{1H} NMR (C6D6, 298 K, 121.5 MHz): d 26.8 (d,
2JPF = 40.3 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz): d ꢀ88.3
A slight excess of the dinitrogen complex [(iPr3P)2Ni]2(
l-N2)
(8 mg, 0.010 mmol) was added to (iPr3P)2Ni(C6F5)2 (12 mg,
0.017 mmol) and dissolved in 550 mg of THF. Over the period of
4 h, 1H, 19F and 31P{1H} NMR spectroscopy showed the production
of compound 3 and the disappearance of both trans-(iPr3P)2
3
4
(dd, 1F, 2–Ar–F, JFF = 16.8 Hz, JFF = 8.9 Hz); ꢀ106.2 (dd, 1F,
5
4
6–Ar–F, JFF = 29.5 Hz, JFF = 8.9 Hz); ꢀ146.4 (dd, 1F, 3–Ar–F,
5JFF = 29.5 Hz, JFF = 8.9 Hz); ꢀ401.8 (tdd, 1F, Ni–F, JFP = 40.3 Hz,
4
2
4JFF = 9.2 Hz, JFF = 9.1 Hz). 13C{1H} NMR (C6D6, 298 K, 75.5 MHz):
4
Ni(C6F5)2 and the majority of [(iPr3P)2Ni]2(
4.6. Synthesis of trans-(iPr3P)2NiF(C6F4H), 6 and 7
The anthracene adduct, (iPr3P)2Ni( 2-C14H10
l-N2).
d 19.5 (s, Pr–CH3); 23.2 (AMM0 virtual t, iPr–CH, JCP = 8.3 Hz);
i
2
2
108.0 (dm, 5–Ar–C, JCF = 32.7 Hz); 111.4 (dd, 4–Ar–C, JCF = 19.6
Hz, JCF = 9.2 Hz); 116.0 (m, ipso–C, JCF = 31.4 Hz); 146.6 (ddm,
3
2
1
2
g
)
(1) (500 mg,
2–Ar–C, JCF = 245.9 Hz, JCF = 19.0 Hz); 153.8 (ddm, 6–Ar–C,
2
4
0.89 mmol) and pentafluorobenzene (150 mg, 0.89 mmol) were
stirred for 72 h in toluene at 50 °C. The crude reaction mixture
was monitored by the disappearance of the C–H activation prod-
uct, trans-(iPr3P)2NiH(2,3,5,6-C6F4H) (5), via multinuclear NMR
spectroscopy. The solvent was removed in vacuo, and the resultant
oil was extracted with pentane. The precipitated anthracene was
filtered, and the dark yellow filtrate placed in the freezer. The
brown-colored viscous oil was identified by NMR spectroscopy to
be a mixture of C–F bond activation isomers: para activation prod-
uct trans-(iPr3P)2NiF(2,3,5,6-C6F4H) (6) and ortho activation prod-
uct trans-(iPr3P)2NiF(2,3,4,5-C6F4H) (7) (85% of 6 and 15% of 7 by
NMR spectroscopy).Compound 6: 1H NMR (C7D8, 298 K,
1JCF = 226.3 Hz; JCF = 24.0 Hz, JCF = 2.0 Hz); 162.9 (dd, 3–Ar–C,
1JCF = 224.0 Hz; JCF = 20.6 Hz). Anal. Calc. for C24H44F4NiP2 (M.W.
2
529.24): C, 54.47; H, 8.38. Found: C, 54.39; H, 8.82%.
Compound 11: 31P{1H} NMR (C6D6, 298 K, 121.5 MHz): d 25.7
2
(d, JPF = 41.3 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz):
d
3
4
ꢀ106.5 (dd, 1F, 2–Ar–F, JFF = 29.6 Hz, JFF = 9.6 Hz); ꢀ147.1 (dd,
3
4
1F, 4–Ar–F, JFF = 19.2 Hz, JFF = 9.6 Hz); ꢀ165.7 (dd, 1F, 3–Ar–F,
3JFF = 29.6 Hz, JFF = 19.2 Hz); ꢀ387.6 (td, 1F, Ni–F, JFP = 41.3 Hz,
3
2
4JFF = 11.2 Hz).
4.9. Characterization of the reaction of 1,2,3,5-tetrafluorobenzene
with 1
i
3
300.13 MHz): d 1.21 (q, 36H, Pr–CH3, JHH = 6.6 Hz); 1.46 (m, 6H,
3
4
iPr–CH); 6.29 (tt, 1H, 4–Ar–H, JHF = 8.3 Hz, JHF = 7.7 Hz). 31P{1H}
A solution of 1 (100 mg, 0.179 mmol) and 1,2,3,5-tetrafluoro-
benzene (28 mg, 0.179 mmol) were dissolved in toluene, and the
NMR (C7D8, 298 K, 121.5 MHz): d 28.1 (dm, JPF = 40 Hz). 19F{1H}
2
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

8
J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
reaction was monitored for 2 weeks at 50 °C by multinuclear NMR
spectroscopy. The initial reaction mixture consisted primarily of
the starting materials. After 24 h, the solution was analyzed again
by 19F and 31P{1H} NMR spectroscopy, which revealed the presence
of the C–H activation product, trans-(iPr3P)2NiH(2,3,4,6-C6F4H),
and smaller intensity resonances attributed to the C–F activation
product, trans-(iPr3P)2NiF(2,3,5-C6F3H2) (13). After 2 weeks, com-
pound 13 was the major product; however, trans-(iPr3P)2-
NiH(2,3,4,6-C6F4H) was present in 30% yield, and the NMR tube
contained precipitated nickel metal. Compound 13: 1H NMR
(C6D6, 298 K, 300.13 MHz): d 1.26 and 1.28 (d, 36H, iPr–CH3,
3JHH = 6.3 Hz); 1.83 (m, 6H, iPr–CH); 6.18 (br m, 1H, 5–Ar–H);
resonances (<2%) attributed to the C–F activation product trans-
(iPr3P)2NiF(3,5-C6F2H3) (17). After 1 week, resonances attributed to
17 did not continue to increase significantly in intensity. Heating
the reaction mixture resulted only in decomposition into nickel
metal. Compound 17: 31P{1H} NMR (C6D6, 298 K, 121.5 MHz):
2
d 25.1 (d, JPF = 41.4 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz):
2
d ꢀ115.4 (s, 2F, 3,5–Ar–F); ꢀ378.6 (t, 1F, Ni–F, JFP = 41.4 Hz).
4.13. X-ray crystallography
The X-ray structures were obtained at low temperatures, with
the crystals covered in Paratone and placed rapidly into the cold
N2 stream of the Kryo-Flex low-temperature device. The data were
collected using the SMART [24] software on a Bruker APEX CCD
3
3
4
6.40 (dddd, 1H, 4–Ar–H, JHF = 17.7 Hz, JHH = 8.6 Hz, JHF = 5.4 Hz,
4JHF = 4.7 Hz). 31P{1H} NMR (C6D6, 298 K, 121.5 MHz): d 26.1 (d,
2JPF = 39.8 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz): d ꢀ113.8
(m, 1F, 2–Ar–F); ꢀ120.0 (dd, 1F, 5–Ar–F, 5JFF = 16.1 Hz, 4JFF = 2.2 Hz);
diffractometer using a graphite monochromator with Mo Ka radi-
ation (k = 0.71073 Å). A hemisphere of data was collected using a
counting time of 10–30 s per frame. Full details of crystallographic
information in CIF format for compounds 2, 3 and 10 are provided
on the Supporting Information. Data reductions were performed
using the SAINT [25] software, and the data were corrected for
absorption using SADABS [26]. The structures were solved by direct
methods using SIR97 [27] and refined by full-matrix least-squares
on F2 with anisotropic displacement parameters for the non-H
3
ꢀ138.0 (dm, 1F, 3–Ar–F, JFF = 31.9 Hz); ꢀ387.9 (td, 1F, Ni–F,
4
2JFP = 39.8 Hz, JFF = 11.5 Hz).
4.10. Characterization of the reaction of 1,2,3-trifluorobenzene with 1
A solution of 1 (20 mg, 0.036 mmol) and 1,2,3-trifluorobenzene
(47 mg, 0.36 mmol, 10 equiv) were dissolved in C6D6, and the reac-
tion was monitored for 1 week by NMR spectroscopy. The initial
reaction mixture consisted primarily of the starting materials.
After 24 h, the solution was analyzed again by 19F and 31P{1H}
NMR spectroscopy, which revealed the presence of two C–F activa-
tion compounds identified as, trans-(iPr3P)2NiF(2,3-C6F3H3) (14)
and trans-(iPr3P)2NiF(2,6-C6F3H3) (15). After 1 week, the reaction
had not gone to completion at room temperature (18% of 14 and
10% of 15 yield by NMR spectroscopy). Compound 14: 31P{1H}
atoms using SHELX-97 [28] and the
WinGX [29] software package,
and thermal ellipsoid plots were produced using ORTEP32 [30].
Acknowledgement
Acknowledgement is made to the National Sciences and
Engineering Research Council (NSERC) of Canada and the Shared
Hierarchical Academic Research Computing Network (SHARCNET)
for their support.
2
NMR (C6D6, 298 K, 121.5 MHz): d 25.9 (d, JPF = 40.2 Hz). 19F{1H}
NMR (C6D6, 298 K, 282.4 MHz):
d
ꢀ108.8 (dm, 1F, 2–Ar–F,
3
3JFF = 31.1 Hz); ꢀ141.6 (d, 1F, 3–Ar–F, JFF = 31.1 Hz); ꢀ383.3 (td,
2
4
Appendix A. Supplementary material
1F, Ni–F, JFP = 40.2 Hz, JFF = 11 Hz).
Compound 15: 31P{1H} NMR (C6D6, 298 K, 121.5 MHz): d 26.5
(d, JPF = 39.4 Hz). 19F{1H} NMR (C6D6, 298 K, 282.4 MHz): d ꢀ81.7
2
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2014.07.023.
2
4
(s, 2F, 2,6–Ar–F); ꢀ397.8 (tt, 1F, Ni–F, JFP = 39.4 Hz; JFF = 8.0 Hz).
References
4.11. Characterization of the reaction of 1,2,4-trifluorobenzene with 2
[1] L. Keyes, J.A. Love, RSC Catal. Ser. 11 (2013) 159.
[2] (a) S.A. Johnson, J.A. Hatnean, M.E. Doster, Prog. Inorg. Chem. 255 (2012) 255;
(b) L. Cronin, C.L. Higgitt, R. Karch, R.N. Perutz, Organometallics 16 (1997)
4920;
A solution of 1 (50 mg, 0.09 mmol) and 1,2,4-trifluorobenzene
(12 mg, 0.09 mmol) were dissolved in toluene-d8, and the reaction
was monitored for 1 week by NMR spectroscopy. The initial reac-
tion mixture consisted primarily of the starting materials. After
48 h, the solution was analyzed again by 19F and 31P{1H} NMR
spectroscopy, which revealed the presence of a C–H activation
product, trans-(iPr3P)2NiH(2,3,6-C6F3H2), and smaller intensity res-
onances attributed to the C–F activation product trans-(iPr3P)2
NiF(2,5-C6F2H3) (16). After 1 week, resonances attributed to 16
continued to develop. Heating the reaction mixture resulted only
in decomposition into nickel metal. Compound 16: 31P{1H} NMR
(c) S.A. Johnson, C.W. Huff, F. Mustafa, M. Saliba, J. Am. Chem. Soc. 130 (2008)
17278;
(d) W. Liu, K. Welch, C.O. Trinkdle, M. Sabat, W.H. Meyers, W.D. Harman,
Organometallics 26 (2007) 2589;
(e) T. Schaub, U. Radius, Chem. Eur. J. 11 (2005) 5024.
[3] (a) I. Bach, K.-R. Poerschke, R. Goddard, C. Kopiske, C. Krueger, A. Rufinska, K.
Seevogel, Organometallics 26 (1996) 4959;
(b) T. Braun, L. Cronin, C.L. Higgitt, J.E. McGrady, R.N. Perutz, M. Reinhold, New
J. Chem. 25 (2001) 19;
(c) M.E. Doster, S.A. Johnson, Angew. Chem., Int. Ed. 48 (2009) 2185;
(d) D.R. Fahey, J.E. Mahan, J. Am. Chem. Soc. 99 (1977) 2501;
(e) T. Schaub, M. Backes, U. Radius, J. Am. Chem. Soc. 128 (2006) 15964;
(f) T. Schaub, P. Fischer, A. Steffen, T. Braun, U. Radius, A. Mix, J. Am. Chem. Soc.
130 (2008) 9304;
(g) P. Fischer, K. Goetz, A. Eichhorn, U. Radius, Organometallics 31 (2012) 1374;
(h) T. Schaub, P. Fischer, T. Meins, U. Radius, Eur. J. Inorg. Chem. 2011 (2011)
3122.
2
(toluene-d8, 298 K, 121.5 MHz): d 25.6 (d, JPF = 43 Hz). 19F{1H}
NMR (toluene-d8, 298 K, 282.4 MHz): d ꢀ90.5 (d, 1F, 2–Ar–F,
5
5JFF = 21.2 Hz); ꢀ123.1 (d, 1F, 3–Ar–F, JFF = 21.2 Hz); ꢀ383.8 (td,
2
4
1F, Ni–F, JFP = 43 Hz, JFF = 6.5 Hz).
[4] (a) S.J. Archibald, T. Braun, J.A. Gaunt, J.E. Hobson, R.N. Perutz, Dalton Trans.
(2000) 2013;
4.12. Characterization of the reaction of 1,3,5-trifluorobenzene with 2
(b) T. Braun, R.N. Perutz, M.I. Sladek, Chem. Commun. (2001) 2254;
(c) S. Burling, P.I.P. Elliott, M.A. Jasim, R.J. Lindup, J. McKenna, R.N. Perutz, S.J.
Archibald, A.C. Whitwood, Dalton Trans. (2005) 3686;
(d) T. Schaub, M. Backes, U. Radius, Eur. J. Inorg. Chem. (2008) 2680.
[5] (a) T. Braun, S.P. Foxon, R.N. Perutz, P.H. Walton, Angew. Chem., Int. Ed. 38
(1999) 3326;
(b) A. Steffen, M.I. Sladek, T. Braun, B. Neumann, H.-G. Stammler,
Organometallics 24 (2005) 4057.
[6] T. Braun, R.N. Perutz, Chem. Commun. 23 (2002) 2749.
[7] S.A. Johnson, E.T. Taylor, S.J. Cruise, Organometallics 28 (2009) 3842.
A solution 1 (20 mg, 0.036 mmol) and 1,3,5-trifluorobenzene
(47 mg, 0.36 mmol, 10 equiv) were dissolved in C6D6, and the reac-
tion was monitored for 1 week by NMR spectroscopy. The initial
reaction mixture consisted primarily of the starting materials. After
48 h, the solution was analyzed again by 19F and 31P{1H} NMR
spectroscopy, which revealed the presence of a C–H activation
product, trans-(iPr3P)2NiH(2,4,6-C6F3H2) (8), and smaller intensity
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023

J.A. Hatnean et al. / Inorganica Chimica Acta xxx (2014) xxx–xxx
9
[8] (a) A.D. Sun, J.A. Love, Organic Lett. 13 (2011) 2750;
(b) A.D. Sun, K. Leung, A.D. Restivo, N.A. La Berge, H. Takasaki, J.A. Love,
Chem. – Eur. J. 20 (2014) 3162;
(c) X. Yang, H. Sun, S. Zhang, X. Li, J. Organomet. Chem. 723 (2013) 36.
[9] (a) D. Breyer, J. Berger, T. Braun, S. Mebs, J. Fluorine Chem. 143 (2012) 263;
(b) A.D. Sun, J.A. Love, Dalton Trans. 39 (2010) 10362.
[10] A. Nova, M. Reinhold, R.N. Perutz, S.A. MacGregor, J.E. McGrady,
Organometallics 29 (2010) 1824.
[11] T.T. Tsou, J.K. Kochi, J. Am. Chem. Soc. 101 (1979) 6319.
[12] J.A. Hatnean, S.A. Johnson, Organometallics 31 (2012) 1361.
[13] S.A. Johnson, N.M. Mroz, R.V. Valdizon, S. Murray, Organometallics 30 (2011)
441.
[14] J.A. Hatnean, R. Beck, J.D. Borrelli, S.A. Johnson, Organometallics 29 (2010)
6077.
[19] C. Eaborn, M.S. Hill, P.B. Hitchcock, J.D. Smith, Chem. Commun. (Cambridge)
(2000) 691.
[20] (a) B.L. Edelbach, W.D. Jones, J. Am. Chem. Soc. 119 (1997) 7734;
(b) B.L. Edelbach, A.K.F. Rahman, R.J. Lachicotte, W.D. Jones, Organometallics
18 (1999) 3170;
(c) M.K. Whittlesey, R.N. Perutz, M.H. Moore, Chem. Commun. (1996) 787.
[21] P.K. Freeman, R. Srinivasa, J. Org. Chem. 52 (1987) 252.
[22] R.E. Banks, Fluorocarbons and Their Derivatives, Macdonald and Co.
(Publishers) Ltd., London, 1970. p. 231.
[23] A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, F.J. Timmers,
Organometallics 15 (1996) 1518.
[24] SMART, Molecular Analysis Research Tool, Bruker AXS Inc., Madison, WI, 2001.
[25] SAINTPlus, Data Reduction and Correction Program, Bruker AXS Inc., Madison,
WI, 2001.
[15] A. Arevalo, A. Tlahuext-Aca, M. Flores-Alamo, J.J. Garcia, J. Am. Chem. Soc. 136
(2014) 4634.
[26] SADABS, An Empirical Absorption Correction Program, Bruker AXS Inc., Madison,
WI, 2001.
[16] R. Beck, M. Shoshani, J. Krasinkiewicz, J.A. Hatnean, S.A. Johnson, Dalton Trans.
42 (2013) 1461.
[17] K.D. Kitiachvili, D.J. Mindiola, G.L. Hillhouse, J. Am. Chem. Soc. 126 (2004)
10554.
[18] C.A. Laskowski, D.J. Bungum, S.M. Baldwin, S.A. Del Ciello, V.M. Iluc, G.L.
Hillhouse, J. Am. Chem. Soc. 135 (2013) 18272.
[27] A. Altomare, M.C. Burla, M. Camalli, G.L. Cascarano, C. Giacovazzo, A.
Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, J. Appl. Crystallogr. 32
(1999) 115.
[28] G.M. Sheldrick, SHELXL-97, Universitat Gottingen, Gottingen, 1997.
[29] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[30] L.J. Farrugia, J. Appl. Crystallogr. 30 (1997) 565.
Please cite this article in press as: J.A. Hatnean et al., Inorg. Chim. Acta (2014), http://dx.doi.org/10.1016/j.ica.2014.07.023