Lewis Acid Assisted C-CN Cleavage of Benzonitrile Using [(dippe)NiH] 2

Article abstract of DOI:10.1055/s-0036-1590801

Reactions of [(dippe)NiH] ₂ with benzonitrile and varying concentrations of Lewis acids (primarily BPh ₃ and BF ₃) have shown a dramatic variation of reaction rate compared to the same reaction without Lewis acids. When less than one equivalent of Lewis acid is used, the reaction rate is as much as 100 times greater than without Lewis acid. Boron exchange was observed with less than one equivalent of Lewis acid allowing the formation of a Ni(0)-η ² -aryl complex (observed by low-temperature NMR and calculations) to which the Lewis acid is postulated to re-coordinate as the rate-limiting step allowing the formation of a stable Ni(II) product. When one equivalent or greater of Lewis acid is used, the reaction shows dramatic inhibition even compared to the reaction without Lewis acid. Lewis acid dissociation can be considered as the rate-limiting step under these conditions. The overall work detailed herein has shown the importance of Lewis acids in the activation of benzonitriles due to the strengthening of the C-N bond allowing for increased stabilization of the products.

Full text of DOI:10.1055/s-0036-1590801

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
©
Georg Thieme Verlag Stuttgart · New York
2017, 28, A–G
cluster
A
en
Synlett
B. D. Swartz et al.
Cluster
Lewis Acid Assisted C–CN Cleavage of Benzonitrile Using
(dippe)NiH]2
[
Brett D. Swartz
i
i
BPh3
Pr2
P
Ni
Pr
P
William W. Brennessel
William D. Jones*
2
N
slow
C
C
Ni
N
P
P
i
i
Ph
Pr2
Pr
+ BPh3
BPh3
2
Department of Chemistry, University of Rochester, Rochester,
NY 14627, USA
jones@chem.rochester.edu
fast
–
BPh3
???
Published as part of the Cluster C–C Activation
Received: 05.05.2017
Accepted: 23.05.2017
Published online: 20.07.2017
DOI: 10.1055/s-0036-1590801; Art ID: st-2017-u0304-c
vided by Garcia and Jones that demonstrated a reversible
1c,2a–c,5,6
activation of benzonitrile (Scheme 1, R = X).
action of [(dippe)NiH]
The re-
2
and benzonitrile initially forms a η -
2
Ni(0) complex1a that is converted into a Ni(II) complex 1b,
reaching an equilibrium favoring 1b (Keq = 1.62). Further
study incorporated the use of para-substituted benzoni-
triles to examine substituent effects on the equilibrium
(Scheme 1).2b The strongly withdrawing functional groups
Abstract Reactions of [(dippe)NiH]2 with benzonitrile and varying
concentrations of Lewis acids (primarily BPh3 and BF ) have shown a
3
dramatic variation of reaction rate compared to the same reaction
without Lewis acids. When less than one equivalent of Lewis acid is
used, the reaction rate is as much as 100 times greater than without
Lewis acid. Boron exchange was observed with less than one equivalent
including CN and CF were shown to drive the equilibrium
3
2
of Lewis acid allowing the formation of a Ni(0)–η -aryl complex (ob-
very strongly toward the C–CN activated Ni(II) product
while the donating groups including NH and OCH drive
served by low-temperature NMR and calculations) to which the Lewis
acid is postulated to re-coordinate as the rate-limiting step allowing the
formation of a stable Ni(II) product. When one equivalent or greater of
Lewis acid is used, the reaction shows dramatic inhibition even com-
pared to the reaction without Lewis acid. Lewis acid dissociation can be
considered as the rate-limiting step under these conditions. The overall
work detailed herein has shown the importance of Lewis acids in the ac-
tivation of benzonitriles due to the strengthening of the C–N bond al-
lowing for increased stabilization of the products.
2
3
2
the equilibrium towards the η -Ni(0) complex. This result
suggested the stabilization of the negative charge on the
ipso-carbon of the substituent and is supported by a large,
positive ρ-value determined in a Hammett study (+6.1). Po-
lar solvents were also found to shift the equilibrium to the
right.
Key words nickel, C–C cleavage, nitriles, Lewis acids, kinetics
The activation of carbon–carbon σ-bonds has many im-
portant applications in both organic synthesis and industri-
al processes. Although frequently accomplished using or-
ganic substrates in systems where strain relief or aromati-
zation act as a driving force, recent developments have
shown an increasing number of organometallic reagents
have successfully performed this activation via the inser-
tion of a transition-metal center into the carbon–carbon σ-
Scheme 1
Lewis acids have been used industrially in conjunction
with transition-metal catalysts to improve product selectiv-
ity, reaction rates, and extend catalyst lifetimes. For exam-
ple, the catalytic hydrocyanation of olefins with Ni(0)–
phosphite catalysts using Lewis acids was studied and
showed that the Lewis acid coordinated through the nitro-
1
bond. One area of recently increased study is C–C activa-
2,3
3c
tion using alkyl and aryl nitrile substrates. A number of
studies have utilized Ni(0)–phosphine and phosphite cata-
lysts as an efficient method for the activation of C–CN
4
bonds. One particular Ni–phosphine catalyst that has been
7
studied in detail is [(dippe)NiH] , which is shown to pro-
gen lone electron pair prior to the reaction. An additional
2
duce a reactive Ni(0) fragment and has been used to acti-
vate alkyl, allyl, and aryl nitriles, including an example pro-
study of nickel-catalyzed pentenenitrile hydrocyanation in-
dicated that the presence of Lewis acids in the reaction pro-
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

B
Synlett
B. D. Swartz et al.
Cluster
motes selectivity of linear products over branched products
as a result of increased steric bulk produced upon coordina-
tion of the Lewis acid to the nitrile. Lewis acids were also
of benzonitrile and BPh coordinated to the lone electron
3
pair of the nitrogen (Figure 1, 2a). The other set of doublets
8
2
was at δ = 79.9 and 67.7 ppm with J = 67 Hz associated
PP
shown to increase the rate of reductive elimination in palla-
dium–phosphine complexes. This was thought to occur as
with 1a. The sample was heated to 65 °C and both sets of
doublets immediately broadened indicating exchange of
9
a result of a change in charge distribution on the nitrogen
and boron or aluminum following Lewis acid coordina-
BPh between the Ni(0)–benzonitrile complexes. Addition-
3
2
ally, a set of doublets at δ = 85.3 and 73.9 ppm with J = 20
PP
10
tion. The resultant complex is more susceptible to nucleo-
philic attack at the nitrile carbon by the alkyl group thereby
accelerating the rate of reaction, in this case reductive elim-
Hz associated with the C–CN activated Ni(II)–benzonitrile
complex with BPh coordinated to the cyano ligand (Figure
3
1, 2b) started to appear and increase as well as a set of dou-
8
2
ination.
blets at δ = 81.7 and 71.2 ppm with J = 20 Hz (1b). Within
PP
Other related work by Garcia et al. investigated the acti-
13 minutes only 1a, 1b and 2b remained, indicating that all
vation of acetonitrile by [(dippe)NiH] in the presence of
of the available BPh was coordinated to 2b since only 0.5
2
3
31
1
BPh . The results indicated that BPh acted to inhibit the C–
equivalent of BPh was present. At this point, the P{ H}
3
3
3
2
CN cleavage leaving only the η -nitrile nickel(0) complex,
possibly related to the steric demand of the BPh group. In
NMR spectrum showed the peaks for 1a to sharpen, indi-
2c
cating that rapid BPh exchange was no longer occurring.
3
3
contrast, Brunkan et al. examined the effect of BPh on the
Equilibrium between 1a and 1b was established after ~3
3
competitive C–C and C–H cleavage of allyl cyanide by
hours.
[
(dippe)NiH] and found that the C–CN activation of allyl
2
cyanide was both faster and irreversible in the presence of
11
BPh3. Nakao observed that boron and aluminum Lewis ac-
ids dramatically increased the rate of alkyne arylcyanation
12
using aryl nitriles.
The work detailed herein describes the reaction of
[
(dippe)NiH] and benzonitrile in the presence of Lewis ac-
2
ids. Since the substitution of functional groups in the para-
position of the arene had significant effects on the equilib-
rium of the system,2b it is reasonable that the coordination
of a Lewis acid would be anticipated to have some effect.
Additionally, a Lewis acid should act to stabilize some of the
complexes involved in the reaction possibly enabling inter-
mediates to be better observed, providing insight into the
mechanistic aspects of this system. In this paper we show
that the amount and type of Lewis acid does indeed dra-
matically affect the rate of the reaction. Specifically, excess
Lewis acid inhibits the rate of C–CN cleavage by making a
stable adduct that is unreactive. In contrast, use of less than
one equivalent of Lewis acid results in acceleration of the
rate of C–CN cleavage by allowing the Lewis acid to coordi-
nate to an intermediate formed further along the reaction
13
Figure 1 NMR spectra of the reaction of [(dippe)NiH] , benzonitrile,
coordinate.
Reaction of [(dippe)NiH] , benzonitrile, and Lewis ac-
2
and 0.5 equivalent of BPh at 65 °C
3
2
2b
ids. As described earlier, 1a is converted into 1b upon
heating to 54 °C over 46 hours (t1/2 = 607 min) and forms an
equilibrium mixture in THF of 1a/1b = 1:1.6 at 54 °C.
Reaction of [(dippe)NiH] , benzonitrile, and >1 equivalent
2
of BPh . Heating to 65 °C with one or two equivalents of
3
Reaction of [(dippe)NiH] , benzonitrile, and 0.5 equivalent
added BPh required 6 days to reach 96% completion with
2
3
of BPh . A solution of [(dippe)NiH] was treated with two
one equivalent BPh and was only 90% complete with two
3
2
3
equivalents of benzonitrile and 0.5 equivalent of BPh (per
equivalents of BPh after two weeks (Figure 2). Additionally,
3
3
Ni) in THF-d at 25 °C turning the deep red solution yellow-
the peaks remained sharp throughout these reactions sup-
8
brown with the evolution of H gas. The reaction was moni-
tored primarily through P{ H} NMR spectroscopy (Figure
porting the notion that BPh coordinates very favorably to
2
3
31
1
1a and 1b. In fact, the resonances for 2b remain sharp un-
der all conditions, indicating that once BPh3 coordinates
forming 2b, no significant exchange occurs. An orange,
block-like crystal (for 2a) and an orange, needle-like crystal
1
). The initial spectrum showed two sets of doublets at δ =
2
2
81.8 and 65.3 ppm with JPP = 41 Hz associated with the η -
coordination of the Ni(0) fragment with the nitrile portion
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

C
Synlett
B. D. Swartz et al.
Cluster
Figure 4 ORTEP drawing of 2b showing 30% probability ellipsoids. Se-
lected bond distances (Å): C1–N1 1.153(2), Ni1–C1 1.8567(19), Ni1–
C2 1.9302(17), N1–B1 1.575(2). Selected angles in degrees: Ni1–N1–
C1 176.03(16), C1–Ni1–C2 89.13(7), C1–N1–B1 175.94(17).
Figure 2 NMR spectra of the reaction of [(dippe)NiH] , benzonitrile,
2
and one equivalent of BPh at 65 °C
3
compounds) can also interact differently to rapidly produce
2b. A series of distribution of species plots for reactions
containing several concentrations of BPh is presented be-
3
low in Figure 5.
Compound 2b appears to be substantially more stable
than 1b since it is by far the major product. X-ray bond dis-
tances and IR data show significant differences supporting
this notion (Table 1). A comparison of the C–N bond dis-
tance in 1a [1.225(6) Å] vs. 2a [1.241(2) Å] shows a slight
increase with the coordination of Lewis acid, similar to
what is seen for 1b [1.148(3) Å] vs. 2b [1.153(2) Å]. This re-
sult is in stark contrast to IR data where complex 2b (ν
=
=
CN
–1
2
170 cm ) showed a much stronger CN bond than 1b (ν
CN
–1
–1
2108 cm ) by 62 cm , which follows trends observed in
Figure 3 ORTEP drawing of 2a showing 30% probability ellipsoids. Se-
lected bond distances (Å): C1–N1 1.241(2), Ni1–N1 1.8866(14), Ni1–
C1 1.8542(17), N1–B1 1.604(2). Selected angles in degrees: B1–N1–
Ni1 146.1(12), N1–C1–Ni1 72.05(10), C1–N1–B1 143.37(15).
15
the formation of M–CN–LA complexes. The increase in
stretching frequency may be attributed to the removal of
electron density from C–N σ-antibonding orbitals,
strengthening the C–N bond.10
However, when comparing the Ni–CN bond distances in
1b [1.935(2) Å] and 2b [1.8567(19) Å] a noticeable differ-
ence (~0.1 Å) is observed. The decrease in Ni–CN bond dis-
(
for 2b) suitable for X-ray diffraction were obtained and se-
lected bond lengths and angles are provided in Figure 3 and
Figure 4.
14
Curiously, these data indicate that the C–CN cleavage re-
action is accelerated when less than one equivalent of BPh3
Table 1 Selected Bond Distances and IR Stretches for Ni–Benzonitrile
Complexes
is used versus the reaction without BPh ; however, once the
3
concentration of BPh3 is increased to one equivalent or
greater, the reaction is strongly inhibited. These observa-
Complex number C–N distance (Å)
Ni–CN distance (Å)
ν
CN (cm–1
)
1
a
1.225(6)
1.148(3)
1.241(2)
1.153(2)
1.2524(19)
–a
1.862(7)
1.935(2)
1.8542(17)
1.8567(19)
1.8499(15)
–a
1745
2108
1755
2170
1722
2130
tions indicate that BPh is playing multiple roles in this re-
3
action, one that slows C–CN cleavage but another that can
increase the rate of C–CN cleavage. One clear and immedi-
1b
2
2
3
a
b
a
b
ate conclusion is that the coordination of BPh to the nitrile
3
lone pair in 1a (to give 2a) does not accelerate C–CN cleav-
age. A second conclusion is that 1a, in the presence of small
quantities of BPh (formed via dissociation of BPh from 2a,
3
3
3
31
a
No crystal structure was obtained for complex 3b.
as evidenced by broadening of the P resonances for both
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

D
Synlett
B. D. Swartz et al.
Cluster
Figure 5 Distribution of species plots for all concentrations of BPh . (a) 0 equiv BPh , (b) 0.5 equiv BPh , (c) 1.0 equiv BPh , (d) 2.0 equiv BPh .
3
3
3
3
3
tance in the presence of a Lewis acid indicates increased π-
back bonding between the metal center and the ligand ulti-
mately resulting in a stronger bond and a more stable com-
plex as demonstrated by the strong equilibrium shift to the
right.
observed in separate experiments employing one and two
equivalents of BF but unlike BPh , all reached completion
within two hours (see Supporting Information).
3
3
Surprisingly, the BF reactions occur much more rapidly
3
than the BPh3 reactions even when BF3 is at one or two
equivalents. Due to the presence of back bonding between
the boron and fluorine, BF3 is a weaker Lewis acid and
Reaction of [(dippe)NiH] , benzonitrile, and 0.5 equivalent
2
of BF . A solution of [(dippe)NiH] was treated with two
3
2
16
equivalents of benzonitrile and 0.5 equivalent of BF in THF-
forms a less stable complex than BPh . IR data of 3b show
3
3
–
1
–1
d8 at 25 °C turning the deep red solution yellow-brown with
a bond stretch of 2130 cm , 22 cm greater than 1b but 40
–1
the evolution of H gas. The reaction was monitored pri-
cm smaller than 2b (Table 1). On the basis of these results,
2
1
31
marily through P{ H} NMR spectroscopy. The initial spec-
BF is not as strong a Lewis acid as BPh , but is still able to
3
3
trum showed two sets of doublets associated with the
coordinate strongly enough in 3b to shift the equilibrium
completely to the right.
2
Ni(0)–(η -benzonitrile–BF ) complex 3a at δ = 66.3 and 59.1
3
2
ppm with J = 56 Hz and another set of doublets at δ =
The greater Lewis acidity of BPh over BF was further
PP
3
3
7
9.9 and 67.7 ppm with 2JPP = 67 Hz associated with 1a. A
demonstrated by a reaction where both BF and BPh were
3 3
third set of doublets, associated with the Ni(II)–benzoni-
in competition, by adding one equivalent of BF and one
3
trile–BF complex 3b was observed even without heating at
equivalent BPh3 simultaneously to
a
solution of
3
2
δ = 84.2 and 74.6 ppm with J = 30.5 Hz (Scheme 2).
[(dippe)NiH] and benzonitrile in THF-d . The reaction was
PP
2
8
31
1
monitored via P{ H} NMR spectroscopy and initially
showed four sets of doublets (see Supplementary Informa-
tion). Two sets of doublets were located at δ = 81.8 and 65.3
2
ppm with J = 41 Hz (2a, 52%) and δ = 85.3 and 71.9 ppm
PP
2
with J = 20 Hz (2b, 18%). The other sets of doublets were
PP
2
located at δ = 66.3 and 59.1 ppm with J = 56 Hz (3a, 15%)
PP
2
and δ = 84.2 and 74.6 ppm with J = 30.5 Hz (3b, 9%), with
PP
Scheme 2
the remaining material appearing as (dippe)Ni(CN) . After
2
15 minutes at 65 °C, 3a disappeared completely and 3b
The sample was heated to 65 °C and the Ni(0) complex
(52%) was the dominant species while 2a (19%) and 2b
resonances for 3a and 1a showed significant broadening as
(22%) were present in nearly equal quantities with the re-
observed with BPh . After one hour at 65 °C, 3a disappeared
mainder of the material consisting of (dippe)Ni(CN) . After
3
2
and only 3b, 1a, and 1b were observed. Similar results were
24 hours at 65 °C, a noticeable change was observed: 2a
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

E
Synlett
B. D. Swartz et al.
Cluster
(3%) was almost completely gone while 2b (48%) was now
was gradually warmed (–60 to 25 °C), 1c disappeared and
the dominant product over 3b (28%) with the remainder
only 2a remained. No evidence of BPh coordinated to 1c
3
being (dippe)Ni(CN) (8%) and decomposition products. At
was observed.
2
this point in the reaction, one additional equivalent of BF3
was added to see if any of the coordinated BPh could be
3
displaced, however, no changes were observed indicating
that BPh bonds more strongly to the nickel complexes, sup-
3
porting the notion of the BF3 reaction occurring more
quickly due to the lability of BF in 3a. These results indi-
3
cate that BPh3 forms a more thermodynamically stable
complex as 2b comes to be the major species over time. 3b
appears as the kinetically preferred product as it is the
dominant species early in the reaction, but is observed to
decrease in favor of 2b over time and cannot be regenerated
even upon the addition of more BF3.
Scheme 3
A second experiment was run where the BPh and ben-
zonitrile were mixed prior to the addition to the
[(dippe)NiH] solution at low temperature with the inten-
3
2
tion of forming 2c with BPh coordinated to the nitrogen
3
lone pair (Scheme 4). The 31P{ H} NMR spectrum showed
1
The reaction with BF was run at –20 °C, allowing isola-
3
tion of yellow, block-like crystals of 3a suitable for X-ray
structure determination (Figure 6). Table 1 shows a com-
parison of relevant bond distances and IR data for 1a–b, 2a–
b, and 3a.
four sets of doublets. One set of doublets was observed at
2
δ = 64.5 and 56.7 ppm with J = 46 Hz suspected to be the
PP
BPh -coordinated form of 1c (2c, 4%). A second set of dou-
3
blets observed was associated with 2a (40%). The third set
of doublets was associated with 2b (40%), and the final set
2
of doublets was observed at δ = 72.2 and 70 ppm with J
=
PP
42 Hz (10%) and is unidentified. The remaining material
represented decomposition products. As the mixture was
slowly warmed to 25 °C, 2c disappeared and 2b appeared in
the same amount (~4% by NMR). All of the other resonances
remained unchanged. The appearance of 2b is very import-
ant as it was never observed previously to be formed at low
temperature and also did not start to appear until about
four minutes at 65 °C in previous experiments. The fact that
it now appears at low temperature supports the proposal
that 2c must be an important intermediate in the mecha-
nism of Lewis acid catalyzed Ph–CN cleavage (vide infra).
Figure 6 ORTEP drawing of 3a showing 30% probability ellipsoids. Se-
lected bond distances (Å): C1–N1 1.2524(19), Ni1–N1 1.8684(13),
Ni1–C1 1.8499(15), N1–B1 1.58(2). Selected angles in degrees: B1–
N1–Ni1 146.19(11), N1–C1–Ni1 71.12(9), C1–N1–B1 144.19(14).
Low-temperature NMR studies.
A
solution of
[
(dippe)NiH] in THF-d was cooled to –78 °C and treated
2
8
with two equivalents of benzonitrile turning the deep red
solution yellow-brown with the evolution of H gas upon
2
mixing. The sample was placed in a pre-cooled NMR probe
31
1
at –60 °C and a P{ H} NMR spectrum was collected. Two
sets of doublets were observed at δ = 79.9 and 67.7 ppm
Scheme 4 Reaction with benzonitrile and BPh at –65 °C
3
2
with J = 67 Hz associated with 1a and a new set of dou-
PP
2
blets at δ = 64.5 and 59.6 ppm with J = 80 Hz associated
with a fluxional Ni(0)–η -aryl complex 1c (Scheme 3), as
observed in previous work. A solution of one equivalent of
PP
2
Use of Lewis acids with substituted benzonitriles. A
17
solution of [(dippe)NiH] was treated with two equivalents
2
BPh in THF was pre-cooled to –78 °C and added to the mix-
of 4-methoxybenzonitrile or 4-methylbenzonitrile and 0.5
equivalent BPh3 in THF-d8 at 25 °C turning the deep red
3
31
1
ture. The P{ H] NMR spectrum showed two sets of dou-
2
blets at δ = 81.8 and 65.3 ppm with J = 41 Hz (2a) and δ =
solution yellow-brown with the evolution of H gas. The re-
PP
2
2
5
8.2 and 53.4 ppm with JPP = 80 Hz (1c). As the solution
actions were heated to 65 °C to produce BPh -coordinated
3
C–CN cleavage adducts 5 and 6, respectively. Monitoring the
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

F
Synlett
B. D. Swartz et al.
Cluster
reactions by 31P{ H} NMR spectroscopy allowed a compari-
1
son with the equilibrium constants generated in previous
2b
work (Table 2).
Table 2 Comparison of Keq Data of para-Substituted Benzonitriles
p-Substituted
group
K
eq
t
1/2
K
eq
t
1/2
(min)
(no BPh
3
)
(min)
(w/ BPh
3
)
H
1.62
0.34
0.74
607.5a
333.9
464.6
large
7.6b
17b
5.99
924.2
–c
OCH
3
CH
3
a
Equilibrium achieved after 46 h at 54 °C.
Equilibrium was achieved within 48 h at 65 °C.
Not determined.
b
c
Previous results indicated that the presence of the para-
Scheme 5 Proposed mechanism for the C–CN activation of benzoni-
substituted group had an effect on the stability of the Ni(II)
complex by either strengthening (withdrawing groups) or
trile by [(dippe)NiH] in the presence of BPh3
2
2b
weakening (donating groups) the Ni–phenyl bond thereby
allowing a reaction to reach an equilibrium state based on
the relative stabilities of the Ni(0) and Ni(II) complexes. The
results of these Lewis acid experiments show that the pres-
ence of Lewis acid not only accelerates the rate of the reac-
tion but changes the equilibrium in favor of the activated
Ni(II) product. The strengthening of the Ni–CN bond by the
Lewis acid appears to be more important than the effect by
the para-substituents to the product stability. Therefore,
Lewis acids do shift the Ph–CN cleavage equilibria to the
right.
1
8
Ni(PMe ) derivative. This step was shown to be vital to
3
2
the formation of 1b; however, previous low-temperature
NMR experiments have shown 1c to be unstable with re-
spect to 1a, and 1a therefore acts to slow the reaction.
2b,17
However, we suggest that when small quantities of BPh are
3
in solution (from BPh dissociation from 2a), the Lewis acid
3
coordinates to 1c via the lone pair of electrons on the nitro-
gen, forming 2c. The current low-temperature NMR studies
showed the existence of 2c and its ability to disappear rap-
idly to form 2b upon warming. The transient appearance of
2c, coupled with the rapid and permanent appearance of
2b, indicates that 2c must be significantly lower in energy
than 1c yet unstable compared to 2b. As a result, this step,
A yellow, block-like crystal of 5 and a colorless, plate-
like crystal of 6 suitable for X-ray diffraction were obtained
and the structures were determined. Bond lengths and an-
gles are provided in the Supporting Information.
the formation of 2c from BPh and 1c, can be considered a
3
Mechanism of Lewis acid assisted C–CN cleavage.
Based on the data presented above, the following mecha-
nism is proposed for the C–CN activation of benzonitrile by
critical difference when less than one equivalent of BPh is
3
used. Additionally, by incorporation of BPh in this step, the
3
rate of C–CN cleavage can be accelerated by ~100-fold to ac-
[
(
(dippe)NiH]2 in the presence of 0.5 equivalent of BPh3
Scheme 5). The first step of the reaction shows the dissoci-
count for the observed rate increase. Consequently, BPh is
the necessary element in this step to allow the accelerated
3
ation of BPh from 2a to form 1a. Evidence for this step was
observed via P{ H} NMR spectroscopy when a significant
broadening of the peaks for 2a and 1a was observed when
formation of 2b which is facilitated by conditions allowing
3
31
1
BPh dissociation from 2a.
3
Another mechanism was proposed by Taw et al.2e show-
ing that the Lewis acid coordinates directly to the metal
center before migrating to the nitrogen of the aryl nitrile.
The order of addition experiment described earlier shows
using less than one equivalent of BPh . When using one
3
equivalent or greater of BPh , no broadening was observed
3
as the equilibrium lies strongly in favor of 2a. Since 2a
forms completely when using one equivalent or greater of
BPh , the energy associated with BPh dissociation must be
no evidence of the Ni–BPh complex under specific reaction
3
conditions indicating that this mechanism is inappropriate
for this system.
3
3
added to the barrier for reactions starting from 1a. Since 2a
is significantly lower in energy than 1a, the reaction rate
actually slows compared to the reaction without BPh3.
The next step, migration of the nickel–benzonitrile
In summary, reactions of [(dippe)NiH] with benzoni-
2
trile and varying concentrations of Lewis acids (primarily
BPh and BF ) have shown a dramatic variation of reaction
3
3
2
2
complex from η -nitrile 1a to η -arene 1c was demonstrat-
rate compared to the same reaction without Lewis acids.
When less than one equivalent of Lewis acid is used, the re-
action rate is as much as 100 times greater. Boron exchange
was observed with less than one equivalent of Lewis acid
ed both experimentally and via DFT calculations reported
17
earlier by our group and in related calculations for the
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G

G
Synlett
B. D. Swartz et al.
Cluster
2
allowing the formation of a Ni(0)–η -aryl complex (ob-
served by low-temperature NMR and calculations) to which
the Lewis acid is postulated to re-coordinate as the rate-
limiting step allowing the formation of a stable Ni(II) prod-
uct. When one equivalent or greater of Lewis acid is used,
the reaction shows dramatic inhibition even compared to
the reaction without Lewis acid. Lewis acid dissociation can
be considered a rate-limiting step under these conditions.
Other Lewis acids (BEt , AlF , SnF , or SnPh ) proved ineffec-
(5) Edelbach, B. L.; Lachicotte, R. J.; Jones, W. D. Organometallics
1
999, 18, 4040.
(
(
(
6) Ateşin, T. A.; Li, T.; Lachaize, S.; García, J. J.; Jones, W. D. Organo-
metallics 2008, 27, 3811.
7) Tolman, C. A.; Seidel, W. C.; Druliner, J. D.; Domaille, P. J.
Organometallics 1984, 3, 33.
8) McKinney, R. J.; Nugent, W. A. Organometallics 1989, 8, 2871.
(9) Huang, J.; Haar, C. M.; Nolan, S. P.; Marcone, J. E.; Moloy, K. G.
Organometallics 1999, 18, 297.
(
(
10) Purcell, K. F.; Drago, R. S. J. Am. Chem. Soc. 1966, 88, 919.
11) Brunkan, N. M.; Brestensky, D. M.; Jones, W. D. J. Am. Chem. Soc.
3
3
4
4
tive (see Supporting Information).
2
004, 126, 3627.
12) (a) Nakao, Y.; Yada, A.; Ebata, S.; Hiyama, T. J. Am. Chem. Soc.
007, 129, 2428. (b) Yada, A.; Ebata, S.; Idei, H.; Zhang, D.;
(
2
Acknowledgment
Nakao, Y.; Hiyama, T. Bull. Chem. Soc. Jpn. 2010, 83, 1170.
(
13) Swartz, B. D. Ph.D. thesis; University of Rochester: USA, 2009.
We acknowledge the U.S. Department of Energy, grant #FG02-
6ER13569, for their support of this work.
2
(14) Preparation of (dippe)Ni(h -PhCNBPh ), 2a. A solution of
8
3
[
(dippe)NiH]2 (0.01 g, 0.016 mmol) in THF-d8 was combined
with benzonitrile (3.3 μL, 0.032 mmol) producing a yellow-
brown solution with the evolution of H gas. BPh (0.0072 g,
.032 mmol) was added and the solution was transferred to an
Supporting Information
2
3
0
Supporting information for this article is available online at
NMR tube. The reaction was observed via NMR spectroscopy
and the solution was collected in a vial and allowed to slowly
https://doi.org/10.1055/s-0036-1590801.
S
u
p
p
ortioIgnfrm oaitn
S
u
p
p
ortioIgnfrm oaitn
crystallize at room temperature giving orange, block-like crys-
1
tals (80% yield by NMR). H NMR (THF-d ): δ = 0.75–1.2 (m, 24
8
References and Notes
i
i
H, Pr-CH ), 1.5–1.7 (m, 4 H, PCH CH P), 1.9–2.1 (m, 4H, Pr-CH),
3
2
2
6
.85–7.7 shows overlapping multiplets in the aryl region.
(1) For reviews of C–C activation by organometallic complexes see:
31
1
2
2
P{ H} NMR: δ = 81.8 (d, J = 41 Hz), 65.3 (d, J = 41 Hz). IR
PP
PP
(a) Murakami, M.; Ito, Y. In Topics in Organometallic Chemistry;
3
V
o.
l
-1
ν(CN) = 1755 cm .
Preparation of (dippe)Ni(CNBPh )(Ph), 2b.
Murai, S., Ed.; Springer: New York, 1999, 97–129. (b) C-C Bond
Activation, In Topics in Current Chemistry;
Springer: Heidelberg, 2014. (c) Rybtchinski, B.; Milstein, M.
Angew. Chem. Int. Ed. 1999, 38, 870. (d) Perthuisot, C.; Edelbach,
B. L.; Zubris, D. L.; Simhai, N.; Iverson, C. N.; Müller, C.; Satoh, T.;
Jones, W. D. J. Mol. Catal. A 2002, 189, 157.
A
solution of
3
3
4
V6o.
l
Dong, G., Ed.;
[
(dippe)NiH]2 (0.01 g, 0.016 mmol) in THF-d8 was combined
with benzonitrile (3.3 μL, 0.032 mmol) producing a yellow-
brown solution with the evolution of H2 gas. BPh3 (0.0072 g,
0.032 mmol) was added and the solution was transferred to an
NMR tube. The reaction was heated to 65 °C and observed via
NMR spectroscopy for 6 days. The solution was collected in a
vial and allowed to slowly crystallize at room temperature
(
(
(
2) (a) Garcia, J. J.; Jones, W. D. Organometallics 2000, 19, 5544.
(b) Garcia, J. J.; Brunkan, N. M.; Jones, W. D. J. Am. Chem. Soc.
2002, 124, 9547. (c) Garcia, J. J.; Arévalo, A.; Brunkan, N. M.;
1
giving orange, needle-like crystals (88% yield by NMR). H NMR
Jones, W. D. Organometallics 2004, 23, 3997. (d) Taw, F. L.;
White, P. S.; Bergman, R. G.; Brookhart, M. J. Am. Chem. Soc.
i
(THF-d ): δ = 0.75–1.2 (m, 24 H, Pr-CH ), 1.5–1.7 (m, 4 H,
8
3
i
PCH CH P), 2.2–2.3 (m, 4H, Pr–CH), 7.5–7.7 shows overlapping
2
2
2002, 124, 4192. (e) Taw, F. L.; Mueller, A. H.; Bergman, R. G.;
31
1
2
multiplets in the aryl region. P{ H} NMR: δ = 85.3 (d, J = 20
PP
Brookhart, M. J. Am. Chem. Soc. 2003, 125, 9808.
Hz), 73.9 (d, 2J = 20 Hz). IR ν(CN) = 2170 cm-1.
PP
3) (a) Nakao, Y.; Hirata, Y.; Tanaka, M.; Hiyama, T. Angew. Chem.
Int. Ed. 2008, 47, 385. (b) Nakao, Y.; Yukawa, T.; Hirata, Y.; Oda,
S.; Satoh, J.; Hiyama, T. J. Am. Chem. Soc. 2006, 128, 7116. (c) van
der Vlugt, J. I.; van Duren, R.; Batema, G.; den Heeten, R.;
Meetsma, A.; Fraanje, J.; Goubitz, K.; Kamer, P. C. J.; van Leeu-
wen, P. W. N. M.; Vogt, D. Organometallics 2005, 24, 5377.
(
(
15) Kristoff, J. S.; Shriver, D. F. Inorg. Chem. 1973, 12, 1788.
16) Housecroft, C. E.; Sharpe, A. G. Inorganic Chemistry, 2nd ed.;
Pearson Education Ltd: England, 2005, 307–308.
17) Ateşin, T. A.; Li, T.; Lachaize, S.; García, J. J.; Jones, W. D. Organo-
metallics 2008, 27, 3811.
(
(
18) Ohnishi, Y.-y.; Nakao, Y.; Sato, H.; Nakao, Y.; Hiyama, T.; Sakaki,
S. Organometallics 2009, 28, 2583.
(d) Nakao, Y. Bull. Chem. Soc. Jpn. 2012, 85, 731.
4) (a) Hartwig, J. F. Organotransition Metal Chemistry: From
Bonding to Catalysis; University Science Books: Sausalito, 2010,
103 674. (b) Tolman, C. A.; McKinney, R. J.; Seidel, W. C.;
Druliner, J. D.; Stevens, W. R. Adv. Catal. 1985, 33, 1.
©
Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–G