Connecting Neutral and Cationic Pathways in Nickel-Catalyzed Insertion of Benzaldehyde into a C-H Bond of Acetonitrile

Article abstract of DOI:10.1021/acs.organomet.5b00405

Nickel catalysts supported by diethylamine- or aza-crown ether-containing aminophosphinite (NCOP) pincer ligands catalyze the insertion of benzaldehyde into a C-H bond of acetonitrile. The catalytic activity of neutral (NCOP)Ni(O^tBu) and cation

Full text of DOI:10.1021/acs.organomet.5b00405

Article
pubs.acs.org/Organometallics
Connecting Neutral and Cationic Pathways in Nickel-Catalyzed
Insertion of Benzaldehyde into a C−H Bond of Acetonitrile
Jacob B. Smith and Alexander J. M. Miller*
Department of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599-3290, United States
*
S Supporting Information
ABSTRACT: Nickel catalysts supported by diethylamine- or
aza-crown ether-containing aminophosphinite (NCOP) pincer
ligands catalyze the insertion of benzaldehyde into a C−H
bond of acetonitrile. The catalytic activity of neutral
t
+
(
NCOP)Ni(O Bu) and cationic [(NCOP)Ni(NCCH )] are
3
starkly different. The neutral tert-butoxide precatalysts are
active without any added base and give good yields of product
after 24 h, while the cationic precatalysts require a base
cocatalyst and still operate much more slowly (120 h in typical
runs). A series of in situ spectroscopic studies identified several
intermediates, including a nickel cyanoalkoxide complex that
was observed in all of the reactions regardless of the choice of
precatalyst. Reaction monitoring also revealed that the neutral tert-butoxide precatalysts decompose to form the cationic
acetonitrile complex during catalysis; this deactivation involves alkoxide abstraction and can be hastened by the addition of
lithium salts. While the deactivated cationic species is inactive under standard base-free conditions, catalysis can be reinitiated by
the addition of catalytic amounts of base.
2
5−27
INTRODUCTION
synthetically useful β-hydroxy nitriles.
In 2005, Ozerov
■
and co-workers reported a cationic Ni catalyst that required
stoichiometric base to couple nitriles and aldehydes. In 2013,
Guan and co-workers isolated a C-bound cyanomethyl species
Ni−CH CN) that proved to be a highly active catalyst
turnover number (TON) of up to 82 000) for C−H insertion
Organonickel complexes orchestrate an array of molecular
23
1
−8
elaborations,
but nickel catalysts have lagged behind
6
−8
precious metals in the functionalization of C−H bonds.
(
2
The choice of substrate can be a deciding factor in nickel-
mediated C−H functionalization: substrates containing a
directing group can hold a C−H bond in close proximity to
(
24
without added base.
9
−13
The two Ni catalysts were proposed to operate through quite
different mechanisms, despite the fact that both are supported
by pincer ligands. The cationic, base-promoted catalyst was
proposed to react through the Lewis acid mechanism shown in
Scheme 2A: nitrile binding to the Ni catalyst would render the
acetonitrile more acidic, facilitating deprotonation by the added
base; the resulting N-bound cyanomethyl species would then
undergo nucleophilic attack on the aldehyde to form the C−C
the metal center,
or substrates containing relatively acidic
C−H bonds can facilitate pathways that rely on deprotona-
1
4,15
tion.
Acetonitrile is an attractive substrate for nickel-catalyzed C−
3
C bond-forming reactions because the sp C−H bond is
16
relatively acidic (pK = 25 in H O) and the products would
a
2
contain a valuable nitrile functionality. Direct deprotonation of
acetonitrile is often impractical or not tolerated by nearby
functional groups, motivating the development of catalytic
methods that avoid stoichiometric amounts of strong base.
This article describes the insertion of benzaldehyde into a
C−H bond of acetonitrile (Scheme 1) catalyzed by asymmetric
Ni aminophosphinite (NCOP) pincer complexes with little or
no base added. Only two nickel catalysts are found among the
23
bond. The neutral, base-free catalyst was proposed to react
through the organometallic mechanism shown in Scheme 2B: a
highly nucleophilic Ni alkoxide would facilitate C−H bond
activation by a “concerted metalation−deprotonation”-type
28
pathway, followed by 1,2-insertion of the aldehyde into the
2
4
resulting C-bound cyanomethyl species.
1
7−22
Considering the proposed mechanisms in Scheme 2, we
hypothesized that a Ni catalyst supported by our pincer-crown
ether ligand framework could utilize the aza-crown ether donor
to hold a Lewis acidic cation near the Ni center and accelerate
handful
of late-transition-metal catalysts known to carry
23,24
out this cyanomethylation reaction,
which produces
Scheme 1
Special Issue: Gregory Hillhouse Issue
Received: May 12, 2015
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 2
2
9
1
the catalytic reaction. In addition to numerous examples of
The H NMR spectrum of 1-15c5 is consistent with a C -
s
30−32
Lewis acid-promoted insertion reactions,
aware that a related Ru catalyst showed improved aldehyde
we were also
symmetric solution structure, as a single benzylic methylene
linker resonance and two isopropyl doublets of doublets were
observed. A singlet (δ 200) was observed in the P{ H} NMR
spectrum. Slow evaporation of an ether solution of 1-15c5
under N fostered the growth of yellow crystals suitable for X-
ray diffraction analysis. The solid-state structure of 1-15c5
(Figure 1) confirmed the expected tridentate meridional
21
31
1
cyanomethylation yields in the presence of 10% NaPF6. With
this hypothesis in mind, we set out to synthesize new pincer-
crown ether complexes of nickel and assess their catalytic
activity for the insertion of benzaldehyde into a C−H bond of
acetonitrile. While the asymmetric NCOP pincer framework
can indeed support catalytic insertion of benzaldehyde into
acetonitrile, we were surprised to find that lithium salts inhibit
the reaction (Na and K+ had no effect on the rate).
Mechanistic studies to probe the origin of the cation-triggered
deactivation revealed that catalysis could be reinitiated by the
addition of base, suggesting a connection between the neutral
and cationic precatalysts.
2
+
RESULTS AND DISCUSSION
Synthesis and Characterization. Metalation of the
■
previously reported pincer-crown ether ligand containing an
29
15c5
iPr
aza-15-crown-5 macrocycle,
(
NCOP )H, was accom-
plished by heating a toluene solution of the ligand and
NiBr (DME) (DME = dimethoxyethane) at 65 °C for 3 h in
2
the presence of 3 equiv of NEt (Scheme 3). The bromide
3
Scheme 3
Figure 1. Structural representation of 1-15c5 with ellipsoids drawn at
the 50% probability level. Hydrogen atoms have been omitted for
clarity. Selected distances (Å) and angles (deg): Ni−P 2.1039(7), Ni−
N 2.030(2), Ni−Br 2.3446(5), Ni−C1 1.854(3), P−Ni−Br 94.37(2),
P−Ni−C1 82.11(8), N−Ni−Br 99.31(6), N−Ni−C1 83.94(10).
coordination mode of the ligand in a distorted square-planar
geometry, with bond distances and angles similar to those in
complex (κ3-15c5NCOP )Ni(Br) (1-15c5) was isolated in high
yield using this procedure adapted from Zargarian, who
reported the first nickel NCOP complexes using related ligands
iPr
previously reported NCOP analogues.
Addition of 1-15c5 to 1.1 equiv of KO Bu suspended in
33
t
toluene resulted in an immediate color change from yellow to
red-orange accompanied by precipitation of KBr and the clean
3
3,34
featuring simple secondary or tertiary amine donors.
Full
3
15c5
iPr
t
synthetic details are provided in the Experimental Section;
accompanying spectra can be found in the Supporting
Information (SI).
formation of (κ - NCOP )Ni(O Bu) (2-15c5) (Scheme 4).
1
The H NMR spectrum of 2-15c5 was similar to that of 1-
15c5, albeit with a characteristic tert-butoxide methyl singlet (δ
B
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 4
Scheme 6
Table 1. Comparing Catalysts and Conditions for
a
Benzaldehyde Insertion into Acetonitrile
1
.42). The 31P{ H} spectrum of 2-15c5 showed a significant
1
b
entry
catalyst
base
time (h) TON
yield (%)
upfield shift relative to the corresponding bromide complex 1-
5c5 (δ 190). An analogous diethylamino-substituted nickel
1
2
3
4
5
6
7
8
9
2-15c5
2-Et
none
none
none
24
24
104
105
0
52
52
0
1
3
Et
iPr
t
precatalyst, (κ - NCOP )Ni(O Bu) (2-Et), was prepared by
3-15c5
3-15c5
3-Et
120
120
120
120
120
24
treating the previously reported bromide complex
κ - NCOP )Ni(Br) (1-Et) with KO Bu (Scheme 4). In
contrast to the reaction with the macrocycle-containing 1-15c5,
the color did not change immediately, and complete conversion
required excess KO Bu and extended reaction times (5 h for 2-
Et vs 1 h for 2-15c5). The H and P{H} NMR spectra of 2-Et
were similar to those of 2-15c5. Complexes 2-15c5 and 2-Et
are hygroscopic and decompose on contact with moisture
apparently to form a Ni hydroxide and free BuOH); samples
3
Et
iPr
33
t
1 mol % DBU
none
100
0
50
0
(
3-Et
1 mol % DBU
1 mol % DBU
91
0
46
0
t
none
none
none
t
1
31
0.5 mol % KO Bu
24
1
12
1
t
0.5 mol % LiO Bu
24
a
Conditions: 0.5 mol % catalyst, 3 mL of 1 M benzaldehyde in
b
t
CH CN, 25 °C. Estimated uncertainty: TON ± 5.
3
(
were stored under N at −30 °C.
2
Cationic nickel complexes were also accessible from the
ambient temperature. Reactions with the diethylamino- and
aza-15-crown-5-containing complexes each proceeded with a
turnover number (TON) of ∼100 (Table 1, entries 1 and 2).
The uncertainty in the TON is estimated at ±5 turnovers on
the basis of triplicate catalytic runs under identical conditions.
corresponding bromide. Addition of AgPF to bromide 1-15c5
6
in CH Cl containing acetonitrile resulted in precipitation of
2
2
3
15c5
iPr
AgBr and the formation of [(κ - NCOP )Ni(NCCH )]-
3
[
PF ] (3-15c5) (Scheme 5). Two new signals were noted in
6
35
Given that some strong bases can catalyze similar reactions,
t
Scheme 5
1
,8-diazabicyclo[5.4.0]undec-7-ene (DBU), LiO Bu, and
t
KO Bu were tested as catalysts in the absence of Ni (Table 1,
entries 7−9). DBU and LiO Bu did not catalyze the reaction,
while KO Bu showed sluggish catalytic activity compared with
t
t
the Ni catalyst.
The cationic complexes 3-15c5 and 3-Et were completely
inactive under the typical conditions described above (Table 1,
entries 3 and 5), consistent with Ozerov’s observation that
the 31P{ H} NMR spectrum, a singlet corresponding to the
1
added base is required for activity with [(PNP)Ni(NCCH )]-
3
−
23
bound ligand (δ 203) and a septet corresponding to the PF
counterion (δ −143). The diethyl analogue [(κ - NCOP )-
Ni(NCCH )][PF ] (3-Et) was obtained using a similar
procedure (Scheme 5).
The acetonitrile complexes 3-15c5 and 3-Et exhibit distinct
NMR spectra when dissolved in CD Cl , in contrast to the
similar spectra observed in CD CN. In the case of 3-Et, the
NMR resonances in CD Cl were sharp and clearly showed a
[OTf]. Accordingly, catalysis commenced when 1 mol %
6
3
Et
iPr
DBU (2 equiv relative to the catalyst) was added to the reaction
mixture under typical conditions of 1 M benzaldehyde in
acetonitrile and 0.5 mol % cationic catalyst (3-15c5 or 3-Et).
3
6
1
Reaction monitoring by H NMR spectroscopy revealed >50%
conversion after 120 h at room temperature. The cationic
species are significantly less active than the tert-butoxide
precatalysts even in the presence of a base. As with the neutral
precatalysts, the diethylamino- and aza-15-crown-5-containing
complexes showed essentially identical results (Table 1, entries
4 and 6).
2
2
3
2
2
singlet corresponding to the methyl group of a bound
acetonitrile ligand. The crown ether-containing complex 3-
5c5 exhibited broad, poorly resolved resonances at room
1
temperature, however. The fluxional behavior is attributed to
The cationic catalyst system is noteworthy for operating with
only catalytic amounts of base. The other cationic Ni catalyst
required elevated temperatures and stoichiometric base for best
results. In fact, additional base had almost no impact on the
efficiency of 3-15c5. Variation of the concentration of base
revealed saturation behavior at low levels of DBU (Figure 2),
with base beyond 1 mol % giving no significant rate
enhancement.
competitive Ni binding by the ether groups in the macrocycle,
29
in analogy to related Ir systems. When the NMR probe was
cooled to −50 °C, the resonances for free acetonitrile and Ni-
bound acetonitrile sharpened, and the major product was the
bound acetonitrile complex (Figure S1 in the SI). Consistent
with a labile acetonitrile ligand, elemental analysis of solid 3-
23
1
5c5 exposed to high vacuum was consistent with loss of
acetonitrile.
Having assessed the baseline catalytic performance for
neutral and cationic catalysts, we turned to an examination of
the effect of cationic Lewis acids on the catalytic performance.
We expected the crown ether to hold the cations near the active
site, providing ample opportunity for interactions between the
substrates and the Lewis acid. However, no rate enhancement
was observed when the catalysis was carried out in the presence
Catalytic Hydrocyanomethylation of Benzaldehyde.
The nickel complexes were assessed as C−H insertion catalysts,
as shown in Scheme 6 and summarized in Table 1. The tert-
butoxide complexes 2-15c5 and 2-Et are active catalysts under
typical reaction conditions of 0.5 mol % catalyst in 3 mL of a 1
M solution of benzaldehyde in acetonitrile (∼10% v/v) at
C
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Figure 2. Turnover number (TON) for the insertion of benzaldehyde
into acetonitrile catalyzed by 0.5 mol % 3-15c5 as a function of added
DBU. Conditions: 3 mL of 1 M benzaldehyde in CH CN, 25 °C, 72 h.
3
A variation of TON ± 5 was generally observed in the catalytic
reactions.
of added NaPF or KPF (Table S1 in the SI), regardless of the
6
6
overall charge of the precatalyst or whether a crown ether
group was present. This result stands in contrast to Ru-
catalyzed cyanomethylation of aldehydes, which exhibit
21
significant rate enhancements in the presence of NaPF6.
Even more surprisingly, addition of LiPF had a dramatic
6
inhibitory effect on the catalysis. As illustrated in Figure 3, under
conditions in which 2-15c5 proceeded to 105 turnovers in the
absence of Li , the reaction halted after less than 15 turnovers
Figure 3. (A) Turnover number (TON) for the insertion of
benzaldehyde into acetonitrile catalyzed by 2-15c5 alone (dark-blue
+
+
in the presence of small amounts of LiPF . Some deactivation
solid circles), 2-15c5 with 0.5 mol % Li (light-blue open circles), 2-Et
6
+
was also observed in reactions catalyzed by 2-Et, but the effect
was much less dramatic than with 2-15c5, suggesting that the
macrocycle aids the deactivation process. Catalyst deactivation
was unique to the tert-butoxide complexes 2-15c5 and 2-Et, as
the lithium salts did not inhibit catalysis by the cationic
complexes 3-15c5 and 3-Et.
alone (dark-red solid squares), or 2-Et with 0.5 mol % Li (light-red
open squares). (B) TON after 24 h as a function of the equivalents of
LiPF added relative to catalyst 2-15c5 (blue circles) or catalyst 2-Et
6
(red squares). Conditions: 0.5 mol % catalyst, 3 mL of 1 M
benzaldehyde, 25 °C. A variation of TON ± 5 was generally observed
in the catalytic reactions.
Mechanistic Studies. In order to better understand the
+
origin of the Li inhibition, NMR studies were carried out to
After formation of the Ni−CD CN fragment, addition of 1
2
probe the individual steps of the reaction. Dissolution of tert-
butoxide complex 2-15c5 in CD CN at room temperature led
to complete conversion to a new species as observed by H and
P{ H} NMR spectroscopy after 40 min, along with a
equiv of benzaldehyde led to complete conversion to another
1
3
new species (Scheme 7 and Figures S2 and S3 in the SI). H
1
NMR spectroscopy showed four sets of isopropyl methyl
doublets of doublets, consistent with a lowering of the
symmetry due to the presence of a chiral center. A broad
new resonance was observed in a region consistent with a
benzylic position (δ 4.59) formed upon aldehyde insertion. The
resonance at δ 4.59 correlated with a nitrile carbon (δ 121.1)
3
1
1
t
resonance characteristic of free BuOH (Figures S2 and S3 in
the SI). The product is assigned as the C-bound cyanomethyl
3
15c5
iPr
complex (κ - NCOP )Ni(CD CN) (4-15c5-d ), as shown
2
2
1
5,24
in Scheme 7.
This formulation was confirmed by the
1
13
preparation of an authentic sample via the reaction of
and several aromatic carbons in a H− C HMBC experiment
(Figure S8 in the SI), further suggesting C−C bond formation.
To aid characterization, the protio analogue was generated in
LiCH CN with bromide 1-15c5. The spectroscopic signatures
2
3
15c5
iPr
of authentic (κ - NCOP )Ni(CH CN) (4-15c5) confirmed
2
that the tert-butoxide complex 2-15c5 reacts with acetonitrile to
give 4-15c5 (or 4-15c5-d in CD CN). The Ni alkyl hydrogen
CH CN and characterized by NMR spectroscopy with the aid
3
of a C D capillary to lock and shim. In CH CN solvent, a
2
3
6
6
3
1
atoms appear as a doublet (δ 0.11, JPH = 5 Hz) in the H NMR
spectrum of 4-15c5 and the methylene carbon atom as a
diastereotopic pair of methylene protonsabsent in samples
prepared using CD CNappeared around 3.5 ppm, and the
3
1
3
1
doublet (δ −15.16, J = 12 Hz) in the C{ H} NMR
benzylic resonance at δ 4.59 became a sharp triplet. These two
PC
2
4,36−38
1
1
spectrum, confirming the presence of a Ni−C bond.
broadening or other signs of fluxionality that might suggest
No
resonances exhibited strong correlation in a H− H COSY
experiment (Figure S12 in the SI). On the basis of this
spectroscopic evidence, the product is assigned as the
interconversion between C-bound and N-bound forms of the
15
3 15c5
iPr
cyanomethyl ligand were observed.
Cyanomethyl species have been proposed as intermediates in
cyanoalkoxide complex (κ - NCOP )Ni(OC(H)(Ph)-
(CH CN)) (5-15c5).
2
2
4
related systems, so a catalytic run was carried out with 0.5 mol
isolated 4-15c5 in a 1 M benzaldehyde solution in CH CN.
Cyanoalkoxide species have been proposed as intermediates
24
%
in Ni-catalyzed cyanomethylation but have not been isolated
or fully characterized because of their high reactivity in the
presence of acetonitrile. For example, treatment of (POCOP)-
3
The observed TON of 100 is similar to the TON for 2-15c5
104 ± 5); 4-15c5 is therefore a viable catalytic intermediate.
(
D
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Scheme 7
Ni(CH CN) with benzaldehyde in CD CN formed the
Scheme 8
2
3
deuterated product (POCOP)Ni(CD CN) and the free
2
24
cyanoalcohol, and the cyanoalkoxide was not observed. The
surprising stability of cyanoalkoxide 5-15c5 has allowed for a
much more detailed characterization of this species, presumably
because the (NCOP)Ni system is less reactive with acetonitrile.
Catalytic reactions under typical conditions (0.5 mol %
catalyst, 1 M benzaldehyde, 25 °C) were also monitored in situ
by NMR spectroscopy in CH CN. The precatalyst 2-15c5
3
3
15c5
iPr
formed both (κ - NCOP )Ni(CH CN) (4-15c5) and
κ - NCOP )Ni(OC(H)(Ph)(CH CN)) (5-15c5) at early
2
3
15c5
iPr
(
2
times (4-15c5:5-15c5 ratio ≈ 50:50), but the ratio of the
intermediates changed over the course of the reaction, with
cyanoalkoxide complex 5-15c5 slowly diminishing over 48 h
(Figures S13 and S14 in the SI). The two intermediates
constituted >70% of the Ni species during the reaction. When
3
1
precatalyst 2-Et was monitored analogously, P NMR
resonances were observed at chemical shifts very similar to
those for 4-15c5 and 5-15c5, suggesting that once again both
the cyanomethyl complex 4-Et and the cyanoalkoxide complex
inactive in the absence of a base additive (Table 1). The
identity of the deactivated Ni catalyst strongly suggests that the
5
-Et were present at the onset of catalysis (4-Et:5-Et ratio ≈
60:40) before the speciation shifted almost entirely to the
+
inhibitory role of Li involves alkoxide abstraction to form
cyanoalkoxide complex 5-Et (Figures S15 and S16 in the SI).
The differences in speciation over time for 2-Et and 2-15c5
were surprising and suggest that relatively minor changes in the
catalyst structure might lead to changes in the turnover-limiting
step. Interestingly, the cationic species 3-15c5 and 3-Et were
observed among several unidentified Ni species that grew in
slowly over time during the reaction. Under these base-free
conditions, the cationic species are likely deactivated forms of
the catalyst (Table 1, entries 3 and 5).
+
t
[
(NCOP)Ni(NCCH )] and either LiO Bu or LiOC(H)(Ph)-
3
(
CH CN). Accordingly, a white precipitate was observed upon
2
+
addition of LiPF to 2-15c5. The lack of inhibition by Na and
K salts is attributed to the greater solubility of the alkoxide
salts that would be formed.
6
+
39,40
The catalytic activity could be stopped and started by
addition of LiPF followed by addition of DBU. A catalytic
6
reaction was initiated under standard conditions of 0.5 mol %
2
-15c5 and 1 M benzaldehyde in CH CN (C D capillary for
3 6 6
The presence of both cyanomethyl and cyanoalkoxide
species during catalysis led us to investigate the reversibility
of the insertion step. To probe for reversibility, 5-15c5 was
1
lock signal) and followed by H NMR spectroscopy (Figure 4).
Addition of 0.5 mol % LiPF (red line in Figure 4) effectively
6
shut down the reaction. After ∼3 h, catalysis was reinitiated by
injection of 1 mol % DBU (green line in Figure 4), albeit at a
lower rate consistent with the longer reaction times required by
cationic catalysts paired with DBU (Table 1). Alkoxide
abstraction by lithium additives connects the neutral catalysts
synthesized in CH CN as described above, isolated as a crude
3
solid by removal of volatiles under vacuum, and redissolved in
CD CN. Over the course of 2 h, the diastereotopic methylene
3
protons of 5-15c5 disappeared and the benzylic triplet
resonance was replaced by a broad singlet (Figures S17 and
S18 in the SI). The data are consistent with reversible C−C
bond formation (Scheme 8): deinsertion of benzaldehyde
2-15c5 and 2-Et with the cationic acetonitrile complexes 3-
15c5 and 3-Et.
Independently synthesized samples of the cationic precatalyst
would be followed by activation of CD CN to form a Ni−
3-15c5 exhibited no reaction upon dissolution in CD CN, with
3
3
CD CN fragment and reinsertion of benzaldehyde to generate
no evidence for the formation of a Ni cyanomethyl species.
2
41
the isotopologue 5-15c5-d . Rapid H/D exchange of
Addition of 1 equiv of DBU (pK = 24.3 in CH CN),
2
a 3
cyanomethylnickel species has been previously observed, and
aldehyde insertion has been proposed to be reversible in related
systems.
When tert-butoxide catalyst 2-15c5 was treated with LiPF6
and monitored during turnover, alkoxide abstraction occurred
immediately to form the cationic acetonitrile complex 3-15c5 in
however, led to a mixture of species, including a broad
resonance indicating a fluxional process attributed to
2
4
competitive binding of DBU and CD CN (Scheme 9 and
3
23
Figure S24 in the SI). Addition of 10 equiv of benzaldehyde
to this solution initiated catalysis (Figures S24 and S25 in the
SI). Benzaldehyde was consumed, and the cyanoalcohol
product grew in. The major Ni species during catalysis were
the DBU adduct and 3-15c5, but about 10% of the Ni was
>90% yield (Figures S20 and S21 in the SI). Essentially no
turnover was observed under these conditions; 3-15c5 is
E
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
precatalysts both generate the same cyanoalkoxide intermedi-
ate, albeit in different steady-state concentrations.
In an attempt to avoid the formation of a base adduct of
t
nickel, the strong, bulky phosphazene base BuNP-
tBu
41
(
pyrrolidinyl) ( P , pK = 28.4 in CH CN) was also
3 1 a 3
tBu
examined. The addition of a slight excess of P1 to the
acetonitrile complex 3-15c5 in CD CN resulted in rapid
3
deprotonation to form the cyanomethyl complex 4-15c5-d2
(
Scheme 9). In contrast to the analogous reaction with DBU,
tBu
no evidence of P1 coordination to Ni was observed. A sharp
31
1
singlet in the P{ H} NMR spectrum (δ 197) indicated that 4-
5c5-d was formed in high yield, and broad resonances for
1
2
tBu
P1 and its conjugate acid were also observed (Figure S27 in
Figure 4. Turnover number (TON) over time for two parallel
reactions containing 0.5 mol % 2-15c5 and 1 M benzaldehyde in
the SI). As expected of a solution containing the
cyanomethylnickel complex, the addition of 25 equiv of
benzaldehyde initiated catalysis. During turnover, a sharp
singlet characteristic of the conjugate acid of P1 suggested
that the base was fully protonated. The cyanoalkoxide complex
1
CH CN (C D capillary) assessed by H NMR spectroscopy. The first
3
6
6
experiment (blue solid circles) was allowed to continue unabated for
tBu
1
7 h, while the second experiment (light-blue open circles) was treated
with 0.5 mol % LiPF after 2.4 h (red line), resulting in catalyst
6
5
-15c5-d2 was the major species during catalysis, again
deactivation; the activity was reinitiated by the addition of 1 mol %
DBU (5.2 h, green line).
contrasting with the situation with DBU, which generated
only a small amount of 5-15c5-d2.
Scheme 9
In a separate experiment, 0.5 mol % cationic precatalyst 3-
tBu
1
5c5 and 1.0 mol %
P1 were combined in a 1 M
benzaldehyde solution in CH CN. Benzaldehyde insertion
3
proceeded smoothly, with the TON reaching 109 after 24 h.
Whereas the 3-15c5/DBU combination was much slower than
tBu
the neutral tert-butoxide precatalyst 2-15c5, the 3-15c5/ P
1
combination provided similar levels of activity as 2-15c5. (In
tBu
the absence of any Ni species, P1 exhibited a TON of only
1
2.)
Scheme 10 depicts a proposed mechanism that is consistent
with the experimental data. The neutral pathway, shown in
green, involves interconversion of nickel cyanomethyl (4-15c5
and 4-Et) and cyanoalkoxide (5-15c5 and 5-Et) species. Both
of these species were independently synthesized and observed
during turnover, and the isolated cyanomethyl complex 4-15c5
exhibited catalytic activity essentially identical to that of the tert-
butoxide complex 2-15c5. The deuterium-labeling study
(Scheme 8) provides strong evidence that both the aldehyde
insertion and C−H bond activation steps are reversible. In
further support of the key role of equilibria, catalytic reactions
present in the form of the insertion product, cyanoalkoxide 5-
1
1
5c5-d . Significantly less of this cyanoalkoxide complex 5-
2
5c5-d was present in the case of the cationic catalyst paired
2
with base, compared with catalytic reactions involving tert-
butoxide complex 2-15c5. These studies establish the viability
of nickel cyanomethyl and cyanoalkoxide species as cyanome-
thylation intermediates. Furthermore, the neutral and cationic
t
containing 0.5 mol % 2-15c5 and 50 mol % BuOH showed
Scheme 10
F
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
1
markedly decreased activity (TON = 62 vs TON = 104 in the
absence of added alcohol). Product inhibition might therefore
explain why reactions tended to slow down before reaching
completion (50−60% yield).
column of activated alumina, and stored over 3 Å molecular sieves. H
and 13C chemical shifts are reported in parts per million relative to
residual protio solvent resonances. All of the P resonances are
reported relative to 85% H PO external standard (0 ppm).
4
3
31
3
4
1
5c5
iPr
29
Et
iPr
33
3 Et
iPr
33
+
(
NCOP )H, ( NCOP )H, (κ - NCOP )Ni(Br), and
Deactivation of the neutral cycle occurs when Li abstracts
24,44
LiCH CN,
were synthesized according to literature procedures.
2
the tert-butoxide or cyanoalkoxide, leading to insoluble
precipitates and generating cationic species 3-15c5 or 3-Et, as
shown in dashed red. The cationic pathway, shown in blue,
requires a base for activity. DBU can deprotonate Ni-bound
acetonitrile, but DBU also displaces the acetonitrile ligand,
engaging an unproductive binding equilibrium. This is
t
KO Bu was sublimed under reduced pressure (0.01 Torr) at 175 °C
prior to use. All of the other reagents were commercially available and
used without further purification. Elemental analyses were performed
by Robertson Microlit Laboratories (Ledgewood, NJ).
Single-crystal X-ray diffraction data were collected on a Bruker
Smart Apex-II diffractometer at 100 ± 2 K with Cu Kα radiation (λ =
.54175 Å). Diffraction profiles were integrated using the SAINT
1
consistent with the pseudo-zeroth-order dependence of the+
software program. Absorption corrections were applied using
SADABS. The structure was solved using direct methods and refined
using the XL refinement package via the least-squares method.
Hydrogen atoms were generated theoretically and refined isotropically
with fixed thermal factors.
catalysis rate on DBU (Figure 2). The conjugate acid H−DBU
4
1
t
(
∼
pK = 24.3 in CH CN) is also more acidic than BuOH (pK
a
3
a
42
40 in CH CN), which may facilitate removal of the
3
cyanoalkoxide species from the cycle by protonolysis. It is
difficult to rule out the alternative pathway shown in Scheme 2,
in which deprotonation of acetonitrile by DBU forms a reactive
N-bound cyanomethyl complex that undergoes outer-sphere
Synthesis of (1 NCOP )Ni(Br) (1-15c5). In a glovebox, a glass
5c5
iPr
Teflon-sealed pressure tube was charged with 168.7 mg (0.547 mmol)
of NiBr (DME) and 10 mL of toluene. In a separate flask, 241.4 mg
0.547 mmol) of ( NCOP )H was stirred with 0.229 mL (1.643
mmol) of triethylamine in 15 mL of toluene for 10 min, and then the
resulting mixture added dropwise to the stirring suspension of
NiBr (DME). The flask was sealed and heated at 65 °C for 3 h. After
heating, the solution was filtered under N , and the filtrate was
evaporated to dryness. The resulting yellow oil was extracted with
Et O and filtered. Slow evaporation yielded yellow crystals that were
washed with 3 × 5 mL of pentane and dried under vacuum (288 mg,
91% yield). Single crystals suitable for an X-ray diffraction study were
O solution of 1-15c5. H NMR
600 MHz, C D ): δ 1.18 (dd, J = 14 Hz, J = 7 Hz, 6H,
2
23
15c5
iPr
(
nucleophilic attack on benzaldehyde. We observed the same
cyanoalkoxide intermediate when monitoring catalytic reactions
carried out with neutral and cationic precatalysts, however,
suggesting a relationship between the neutral and cationic
2
tBu
2
cycles. When the stronger, bulkier phosphazene base P1 is
employed, the unproductive binding equilibrium is avoided
2
(
Scheme 10), and the base can be considered an initiator that
quickly generates the more active cyanomethyl catalyst 4-15c5.
1
The fully protonated state of the phosphazene during catalysis
grown by slow evaporation of an Et
2
tBu
3
3
(
suggests a mechanism in which P1 deprotonates the Ni-
6 6 HP HH
3
3
CH(CH ) ), 1.50 (dd, J = 17 Hz, J = 7 Hz, 6H, CH(CH ) ),
bound acetonitrile to enter the neutral cycle in Scheme 10.
3
2
HP
HH
3 2
2
.23 (m, 2H, CH(CH ) ), 3.16 (m, 2H, crown-CH ), 3.24 (m, 4H,
3
2
2
crown-CH ), 3.31 (m, 4H, crown-CH ), 3.40 (m, 2H, crown-CH ),
2
2
2
CONCLUSIONS
■
3.63 (m, 2H, crown-CH ), 3.80 (m, 2H, crown-CH ), 4.07 (m, 2H,
2
2
Neutral and cationic NCOP Ni complexes are viable catalysts
for the insertion of benzaldehyde into a C−H bond of
acetonitrile. Subtle differences between the catalyst resting
states were observed for macrocycle-containing pincer-crown
ether complexes and pincer complexes with a diethylamine
donor. A surprising deactivation occurred when neutral Ni tert-
butoxide catalysts were treated with LiPF , which revealed a
pathway connecting the neutral catalysts to cationic Ni species
that are catalytically inactive in the absence of base. Catalytic
crown-CH
2
), 4.14 (m, 2H, crown-CH3
2
), 4.47 (s, 2H, ArCH N), 6.52
2
3
(
d, J = 7 Hz, 1H, ArH), 6.63 (d, J = 8 Hz, 1H, ArH), 6.90 (t,
HH HH
3
13
1
JHH = 8 Hz, 1H, ArH). C{ H} NMR (150 MHz, C D ): δ 16.58 (d,
6
6
2JCP = 2 Hz, CH ), 17.99 (d, J = 4 Hz, CH ), 28.16 (d, J = 24 Hz,
2
1
3
CP
3
CP
CH), 56.89 (s, crown-CH ), 65.28 (s, ArCH N), 69.49 (s, crown-
2
2
CH ), 70.27 (s, crown-CH ), 70.56 (s, crown-CH ), 70.58 (s, crown-
2
2
2
3
4
CH ), 107.58 (d, J = 13 Hz, C ), 115.38 (d, J = 2 Hz, CAr),
2
CP
Ar
CP
6
2
1
(
26.68 (s, C ), 142.08 (d, J = 33 Hz, C ), 153.35 (s, C ), 165.58
Ar CP Ar Ar
2
31
1
d, J = 11, CAr). P{ H} NMR (161 MHz, C D ): δ 200.4. Anal.
C
P
6
6
Calcd for C H O NPNiBr: C, 47.70; H, 6.79; N, 2.42. Found: C,
23
39
5
+
activity switched off by the addition of Li could be switched
47.60; H, 6.50; N, 2.38.
5c5
iPr
t
back on by addition of DBU. These in situ mechanistic studies
revealed a link between the two precatalysts and provided more
detailed characterization of a proposed cyanoalkoxide inter-
mediate.
Synthesis of (1 NCOP )Ni(O Bu) (2-15c5). In a glovebox, a 20
t
mL vial was charged with 10.9 mg (0.097 mmol) of KO Bu and 2 mL
of toluene. A solution of 51.2 mg (0.088 mmol) of (1 NCOP )-
Ni(Br) in toluene was added dropwise, and the resulting solution was
allowed to stir for 1 h. The solvent was removed in vacuo, and the
reaction mixture was extracted with cold pentane (−30 °C) and
filtered. The solution was evaporated to yield a hygroscopic red-orange
5c5
iPr
EXPERIMENTAL SECTION
General Considerations. Standard vacuum line and glovebox
techniques were utilized to maintain a N2 atmosphere during
manipulation of all compounds, unless otherwise noted. NMR-scale
■
1
3
oil (47 mg, 93% yield). H NMR (400 MHz, C D ): δ 1.19 (dd, J
=
6
6
HP
3
13 Hz, J = 7 Hz, 6H, CH(CH ) ), 1.45 (s, 9H, Ni−OC(CH ) ),
HH
3
2
3 3
3
3
1.56 (dd, J = 18 Hz, J = 7 Hz, 6H, CH(CH ) ), 2.20 (m, 2H,
HP
HH
3 2
reactions were prepared under N in a glovebox and monitored in
CH(CH ) ), 3.36 (m, 14H, crown-CH ), 3.80 (m, 2H, crown-CH ),
2
3 2 2 2
Teflon-sealed tubes. Organic solvents were dried and degassed with
argon using a Pure Process Technology solvent system and stored over
4.06 (m, 2H, crown-CH ), 4.35 (s, 2H, ArCH N), 4.44 (m, 2H,
2 2
3
3
crown-CH ), 6.51 (d, J = 7 Hz, 1H, ArH), 6.58 (d, J = 8 Hz,
2 HH HH
3
13
1
3
Å molecular sieves. Under standard glovebox operating conditions,
1H, ArH), 6.90 (t, J = 8 Hz, 1H, ArH). C{ H} NMR (150 MHz,
HH
2
2
pentane, diethyl ether, benzene, toluene, and tetrahydrofuran were
C D ): δ 17.34 (d, J = 4 Hz, CH(CH ) ), 19.32 (d, J = 6,
6 6 CP 3 2 CP
1
used without purging, so traces of those solvents were present in the
CH(CH ) ), 29.92 (d, J = 20 Hz, CH(CH ) ), 36.49 (s, Ni−
3
2
CP
3 2
1
31
13
atmosphere and in the solvent bottles. H, P, and C NMR spectra
were recorded on 400, 500, or 600 MHz spectrometers. NMR
characterization data are reported at 25 °C, unless specified otherwise.
All of the NMR solvents were purchased from Cambridge Isotopes
Laboratories. Acetonitrile-d (CD CN), benzene-d (C D ), chloro-
OC(CH ) ), 56.38 (s, crown-CH ), 66.46 (s, 1C, ArCH N), 68.64 (s,
3 3 2 2
Ni−OC(CH ) ), 69.64 (s, crown-CH ), 70.84 (s, crown-CH ), 71.00
3
3
2
2
3
(s, crown-CH ), 71.47 (s, crown-CH ), 107.11 (d, J = 12 Hz, C ),
2
2
CP
Ar
4
2
115.32 (d, J = 2 Hz, C ), 126.05 (s, C ), 138.24 (d, J = 35 Hz,
CP
Ar
Ar
CP
2
31
1
C ), 154.25 (s, C ), 165.98 (d, J = 10 Hz, CAr). P{ H} NMR
3
3
6
6
6
Ar
Ar
CP
form-d (CDCl ), and methylene chloride-d (CD Cl ) were freeze−
(161 MHz, C D ): δ 189.57. Anal. Calcd for C H NNiO P: C,
56.66; H, 8.45; N, 2.45. Found: C, 55.83; H, 7.98; N, 2.29. Complex 2-
3
2
2
2
6 6 27 48 6
pump−thaw-degassed three times, dried by passage through a small
G
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
3
1
5c5 is an extremely hygroscopic oil; the high reactivity led to some
CH CH ), 16.85 (d, J = 3 Hz, CH(CH ) ), 17.64 (d, J = 5 Hz,
2 3 CP 3 2 CP
1
3
deviation from the expected elemental analysis values.
CH(CH ) ), 28.80 (d, J = 24 Hz, CH(CH ) ), 56.47 (d, J = 2
3
2
CP
3
2
CP
Et
iPr
t
3
3
Synthesis of ( NCOP )Ni(O Bu) (2-Et). In a glovebox, a 20 mL
Hz, CH CH ), 65.12 (d, J = 2 Hz, ArCH N), 109.06 (d, J = 13
2 3 CP 2 CP
t
4
vial was charged with 17.0 mg (0.152 mmol) of KO Bu and 2 mL of
toluene. A solution of 26.2 mg (0.061 mmol) of ( NCOP )Ni(Br) in
2
allowed to stir for 6 h. The solvent was removed in vacuo, and the oily
residue was extracted with cold pentane (−30 °C) and filtered. The
filtrate was evaporated to yield a red-orange powder (24 mg, 93%
yield). H NMR (400 MHz, C D ): δ 1.17 (dd, J = 12 Hz, J = 7
Hz, 6H, CH(CH ) ), 1.36 (s, 9H, Ni−OC(CH ) ), 1.58 (dd, J = 18
Hz, J = 7 Hz, 6H, CH(CH ) ), 1.62 (t, J = 7 Hz, 6H, CH CH ),
Hz, CAr), 115.80 (d, JCP = 2 Hz, CAr), 127.12 (s, 1C, Ni−NCCH
),
3
Et
iPr
2
129.16 (s, CAr), 136.18 (d, JCP = 32 Hz, CAr), 154.54 (s, CAr), 166.01
2
31
1
mL of toluene was added dropwise, and the resulting solution was
(d, JCP = 9 Hz, CAr). P{ H} NMR (242 MHz, CD
(sept, PF ), 199.0 (s). Anal. Calcd for C19 NiOP
H, 5.98; N, 5.20. Found: C, 42.06; H, 5.88; N, 4.97.
2
Cl
2
): δ −144.49
2
H
32
F
6
N
: C, 42.33;
6
2
Synthesis of (1 NCOP )Ni(CH
20 mL vial was charged with 1.5 mL of THF, 3.3 μL (0.063 mmol) of
CH
5c5
iPr
CN) (4-15c5). In a glovebox, a
2
1
3
3
6
6
HP
HH
3
CN, and a stirbar. The mixture was frozen at −196 °C, and 36.8
3
2
3
3
HP
3
3
3
μL (0.059 mmol) of n-BuLi (1.6 M in hexanes) was slowly added. The
resulting solution was allowed to warm to room temperature and was
stirred for 20 min. It was then cooled to −30 °C in the glovebox
HH
3
2
HH
2
3
1
.94 (m, 2H, CH CH ), 2.22 (m, 2H, CH(CH ) ), 3.37 (m, 2H,
2 3 3 2
3
CH CH ), 3.54 (s, 2H, ArCH N), 6.42 (d, J = 7 Hz, 1H, ArH),
2
3
2
HH
freezer. A solution of 28.4 mg (0.049 mmol) of (1 NCOP )Ni(Br)
in THF at −30 °C was then added, and the mixture was allowed to
warm to room temperature again and stirred for 2 h. The volatiles
were removed, and the residue was extracted with toluene and filtered.
The toluene was removed to yield an amber oil (19 mg, 72% yield).
5c5
iPr
3
3
6
.55 (d, J = 8 Hz, 1H, ArH), 6.89 (t, J = 8 Hz, 1H, ArH).
HH HH
1
3
1
C{ H} NMR (150 MHz, C D ): δ 13.57 (s, CH CH ), 17.31 (d,
6
6
2
3
3
3
JCP = 4 Hz, CH(CH ) ), 19.50 (d, J = 6 Hz, CH(CH ) ), 30.33 (d,
3
2
CP
3 2
1J = 20 Hz, CH(CH ) ), 36.30 (s, Ni−OC(CH ) ), 54.67 (d, J
2
3
=
CP
3
2
3
3
CP
3
Hz, CH CH ), 63.84 (d, J = 2 Hz, ArCH N), 68.38 (s, Ni−
2
3
CP
2
1
3
3
4
H NMR (400 MHz, CD
2
Cl
2
): δ 0.11 (d, JHP = 5 Hz, 2H, Ni−
OC(CH ) ), 107.00 (d, J = 12 Hz, C ), 114.15 (d, J = 2 Hz),
3
3
CP
Ar
CP
3
3
2
CH
2
CN), 1.30 (dd, JHP = 14 Hz, JHH = 7 Hz, 6H, CH(CH
3
)
2
), 1.38
), 2.37 (m, 2H,
CH(CH ) ), 3.38 (m, 4H, crown-CH ), 3.61 (m, 12H, crown-CH ),
1
25.80 (s, C ), 138.29 (d, J = 34 Hz, C ), 154.54 (s, C ), 165.90
Ar CP Ar Ar
3
3
2
31
1
(dd, JHP = 17 Hz, JHH = 7 Hz, 6H, CH(CH )
3 2
(
d, J = 10 Hz, CAr). P{ H} NMR (161 MHz, C D ): δ 188.6.
C
P
6
6
3
2
2
2
Complex 2-Et is extremely hygroscopic; the high reactivity
confounded characterization by elemental analysis or high-resolution
mass spectrometry.
3
4
.04 (m, 4H, crown-CH ), 4.35 (s, 2H, ArCH N), 6.46 (d, J = 8
2 2 HH
Hz, 1H, ArH), 6.60 (d, J = 7 Hz, 1H, ArH), 6.90 (t, JHH = 8 Hz, 1H,
HH
ArH). 13C{ H} NMR (150 MHz, CH
1
CN): δ −15.16 (d, JCP = 12
3
15c5
iPr
3
Synthesis of [( NCOP )Ni(NCCH )][PF ] (3-15c5). In a
3
6
3
3
Hz, Ni−CH CN), 17.37 (d, J = 2 Hz, CH(CH ) ), 18.33 (d, J
=
2
CP
3
2
CP
glovebox, a 20 mL vial covered with foil was charged with 150 μL
2.872 mmol) of CH CN, 49.0 mg (0.193 mmol) of AgPF , and 3 mL
1
5
Hz, CH(CH ) ), 28.77 (d, J = 24 Hz, CH(CH ) ), 58.09 (s,
3 2 CP 3 2
(
3
6
1
5c5
iPr
crown-CH ), 68.62 (s, ArCH N), 69.86 (s, crown-CH ), 71.15 (s,
2
2
2
of CH Cl A solution of 102.0 mg (0.176 mmol) of ( NCOP )-
Ni(Br) in 1 mL of CH Cl was slowly added to the stirring silver
solution. After 1 h, 4 mL of pentane was added to the mixture, and the
resulting solution was filtered. The solvent was removed in vacuo to
2
2.
crown-CH ), 71.28 (s, crown-CH ), 71.57 (s, crown-CH ), 107.88 (d,
JCP = 13 Hz, C ), 116.08 (s, C ), 126.41 (s, Ni−CH CN), 130.87 (s,
C ), 148.94 (d, J = 31 Hz, C ), 153.27 (s, C ), 165.28 (d, J =
0 Hz, CAr). P{ H} NMR (161 MHz, CD Cl ): δ 196.7. The highly
reactive cyanomethyl complex was not suitably stable to pass elemental
2
2
2
2
2
3
Ar
Ar
2
2
2
Ar
CP
Ar
Ar
CP
31
1
1
2 2
yield a yellow oil. The oil was washed with 3 × 5 mL of Et O to yield a
2
1
yellow powder (97 mg, 85% yield). H NMR (500 MHz, CD Cl , −50
2
2
3
3
analysis.
°
C): δ 1.29 (dd, J = 17 Hz, J = 7 Hz, 6H, CH(CH ) ), 1.33 (dd,
HP HH 3 2
General Procedure for Nickel-Catalyzed Insertion of
Benzaldehyde into a C−H Bond of Acetonitrile. In a glovebox,
a 20 mL vial was charged with 306 μL (3.0 mmol) of benzaldehyde,
2.5 mL of acetonitrile, and optional base (e.g., 4.5 μL, 0.03 mmol of
DBU). After addition of a 500 μL aliquot of a solution of 2-15c5 or 3-
15c5 in acetonitrile (0.03 M, 0.015 mmol), the solution was capped
and allowed to stir at room temperature for 24−120 h. After the
desired reaction time, a 300 μL aliquot was evaporated to dryness
under reduced pressure and dissolved in CDCl , and 6.4 μL (30 μmol)
of hexamethyldisiloxane internal standard was added before NMR
spectroscopic analysis.
General Procedure for NMR-Scale Reactions. In a glovebox, a
standard catalytic solution was prepared as described in the general
procedure. A 600 μL aliquot of the reaction mixture was added to a
Teflon-capped NMR tube. Most of the reactions were run in protio
solvent (CH CN), so a sealed capillary of C D was added to the tube
to provide a signal on which to lock and shim. Chemical shifts are
reported relative to the residual C D H signal of the capillary tube.
3
3
JHP = 17 Hz, J = 7 Hz, 6H, CH(CH ) ), 2.31 (m, 2H, CH(CH ) ),
HH
3
2
3 2
2
.45 (s, 3H, Ni−NCCH ), 3.16 (m, 2H, crown-CH ), 3.35 (m, 2H,
3
2
crown-CH ), 3.53 (m, 12H, crown-CH ), 3.96 (s, 2H, ArCH N), 3.98
2
2
2
3
(
m, 2H, crown-CH ), 4.61 (m, 2H, crown-CH ), 6.49 (d, J = 8 Hz,
2 2 HH
3
3
1
H, ArH), 6.64 (d, J = 7 Hz, 1H, ArH), 6.99 (t, J = 8 Hz, 1H,
HH HH
1
3
1
ArH). C{ H} NMR (150 MHz, CD Cl ): δ 3.11 (s, Ni−NCCH ),
2
2
3
3
3
1
2
6.90 (d, J = 3 Hz, CH(CH ) ), 17.77 (d, J = 5 Hz, CH(CH ) ),
8.87 (d, J = 24 Hz, CH(CH ) ), 57.93 (s, crown-CH ), 66.36 (s,
CP 3 2 CP 3 2
1
CP
3
2
2
3
ArCH N), 69.99 (s, crown-CH ), 70.11 (s, crown-CH ), 70.20 (s,
crown-CH ), 72.45 (s, crown-CH ), 109.42 (d, J = 13 Hz, CAr),
2
2
2
3
2
2
CP
4
1
1
17.13 (d, J = 2 Hz, C ), 123.74 (s, Ni−NCCH ), 129.17 (s, C ),
52.80 (s, C ), 166.19 (d, J = 9 Hz, C ). Note: the ipso carbon
CP
Ar
3
Ar
2
Ar
CP
Ar
bound to Ni was not observed, perhaps because of broadening of this
signal due to fluxional acetonitrile binding. P{ H} NMR (161 MHz,
3
1
1
3
1
1
CD CN): δ −143.4 (sept, PF ), 202.7 (s). P{ H} NMR (161 MHz,
3
6
3
6
6
CD Cl ): δ −144.5 (sept, PF ), δ 199.12 (s). Note: The CH CN
2
2
6
3
ligand was lost under high vacuum; elemental analysis was calculated
6
5
4
15c5
iPr
1
31
1
for [(κ - NCOP )Ni][PF ]. Anal. Calcd for C H F NNiO P : C,
The tube was capped and monitored for 24−120 h by H and P{ H}
NMR spectroscopy. Stoichiometric reactions were prepared and
monitored analogously.
6
23 39
6
5 2
4
2
2.88; H, 6.10; N, 2.17. Found: C, 42.78; H, 5.92; N, 2.17.
Et
iPr
Synthesis of [( NCOP )Ni(NCCH )][PF ] (3-Et). In a glovebox, a
3
6
0 mL vial covered with foil was charged with 34.6 mg (0.137 mmol)
of AgPF , 3 mL of CH Cl , and 150 μL (2.872 mmol) of CH CN. A
6
2
2
3
Et
iPr
ASSOCIATED CONTENT
solution of 53.7 mg (0.125 mmol) of ( NCOP )Ni(Br) in 2 mL of
CH Cl was slowly added to the stirring silver solution. After 1 h, 5 mL
■
2
2
* Supporting Information
S
of pentane was added to the mixture, and the resulting solution was
filtered. The solvent was removed in vacuo to yield a yellow oil, which
NMR spectra and crystallographic details (CIF). The
Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.organo-
met.5b00405.
was washed with 3 × 5 mL of Et O to yield a yellow powder (65 mg,
2
1
3
9
7% yield). H NMR (600 MHz, CD Cl ): δ 1.36 (dd, J = 15 Hz,
JHH = 7 Hz, 6H, CH(CH ) ), 1.42 (dd, J = 19 Hz, J = 7 Hz,
2 2 HP
3
3
3
3
3
2
HP
HH
6
H, CH(CH ) ), 1.66 (t, J = 7 Hz, 6H, CH CH ), 2.36 (m, 2H,
3 2 HH 2 3
CH CH ), 2.45 (s, 3H, NCCH ), 2.81 (m, 2H, CH(CH ) ), 3.06 (m,
2
3
3
3
2
AUTHOR INFORMATION
3
■
2
H, CH CH ), 4.06 (s, 2H, ArCH N), 6.52 (d, J = 8 Hz, 1H, ArH),
2 3 2 HH
3
3
Corresponding Author
*E-mail: ajmm@email.unc.edu.
6
.63 (d, J = 8 Hz, 1H, ArH), 7.03 (t, J = 8 Hz, 1H, ArH).
HH
HH
13
1
C{ H} NMR (150 MHz, C D ): δ 3.74 (s, Ni−NCCH ), 13.42 (s,
6
6
3
H
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX

Organometallics
Article
Notes
(32) West, N. M.; Miller, A. J. M.; Labinger, J. A.; Bercaw, J. E. Coord.
Chem. Rev. 2011, 255, 881−898.
The authors declare no competing financial interest.
(
33) Spasyuk, D. M.; Zargarian, D.; van der Est, A. Organometallics
2
(
009, 28, 6531−6540.
34) Spasyuk, D. M.; Zargarian, D. Inorg. Chem. 2010, 49, 6203−
213.
35) Rodriguez, A.; Bunlaksananusorn, T.; Knochel, P. Org. Lett.
000, 2, 3285−3287.
36) Lin, B. L.; Clough, C. R.; Hillhouse, G. L. J. Am. Chem. Soc.
002, 124, 2890−2891.
37) Kitiachvili, K. D.; Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem.
Soc. 2004, 126, 10554−10555.
38) Laskowski, C. A.; Bungum, D. J.; Baldwin, S. M.; Del Ciello, S.
A.; Iluc, V. M.; Hillhouse, G. L. J. Am. Chem. Soc. 2013, 135, 18272−
8275.
39) Jackman, L. M.; Smith, B. D. J. Am. Chem. Soc. 1988, 110,
829−3835.
40) Kissling, R. M.; Gagne,
010.
41) Kaljurand, I.; Ku
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019−1028.
42) Waidmann, C. R.; Miller, A. J. M.; Ng, C.-W. A.; Scheuermann,
M. L.; Porter, T. R.; Tronic, T. A.; Mayer, J. M. Energy Environ. Sci.
012, 5, 7771−7780.
43) Fulmer, G. R.; Miller, A. J. M.; Sherden, N. H.; Gottlieb, H. E.;
ACKNOWLEDGMENTS
■
6
(
2
(
2
(
Acknowledgment is made to the donors of the American
Chemical Society Petroleum Research Fund (52325-DNI3)
and the University of North Carolina at Chapel Hill for support
of this research. Matthew Kita and Peter White assisted with X-
ray crystallography.
(
REFERENCES
■
(
1) Matsunaga, P. T.; Hess, C. R.; Hillhouse, G. L. J. Am. Chem. Soc.
1
(
3
(
9
(
1
994, 116, 3665−3666.
(
2) Mindiola, D. J.; Hillhouse, G. L. J. Am. Chem. Soc. 2002, 124,
976−9977.
3) Waterman, R.; Hillhouse, G. L. J. Am. Chem. Soc. 2003, 125,
3350−13351.
4) Harrold, N. D.; Waterman, R.; Hillhouse, G. L.; Cundari, T. R. J.
Am. Chem. Soc. 2009, 131, 12872−12873.
5) Laskowski, C. A.; Miller, A. J. M.; Hillhouse, G. L.; Cundari, T. R.
J. Am. Chem. Soc. 2011, 133, 771−773.
6) Tasker, S. Z.; Standley, E. A.; Jamison, T. F. Nature 2014, 509,
99−309.
7) Henrion, M.; Ritleng, V.; Chetcuti, M. J. ACS Catal. 2015, 5,
9
(
1
(
́
M. R. J. Org. Chem. 2001, 66, 9005−
̈
tt, A.; Soovali, L.; Rodima, T.; Maemets, V.;
̈
̈
(
(
2
(
(
2
(
Nudelman, A.; Stoltz, B. M.; Bercaw, J. E.; Goldberg, K. I.
Organometallics 2010, 29, 2176−2179.
1
(
(
283−1302.
(
44) Davidson, J. G.; Barefield, E. K.; Van Derveer, D. G.
8) Ananikov, V. P. ACS Catal. 2015, 5, 1964−1971.
Organometallics 1985, 4, 1178−1184.
9) Rouquet, G.; Chatani, N. Angew. Chem., Int. Ed. 2013, 52,
1
(
1726−11743.
10) Shiota, H.; Ano, Y.; Aihara, Y.; Fukumoto, Y.; Chatani, N. J. Am.
Chem. Soc. 2011, 133, 14952−14955.
(11) Aihara, Y.; Chatani, N. J. Am. Chem. Soc. 2013, 135, 5308−5311.
(12) Aihara, Y.; Chatani, N. J. Am. Chem. Soc. 2014, 136, 898−901.
(13) Song, W.; Lackner, S.; Ackermann, L. Angew. Chem., Int. Ed.
2
014, 53, 2477−2480.
14) Yamaguchi, J.; Muto, K.; Itami, K. Eur. J. Org. Chem. 2013, 2013,
9−30.
15) Oertel, A. M.; Ritleng, V.; Chetcuti, M. J.; Veiros, L. F. J. Am.
Chem. Soc. 2010, 132, 13588−13589.
16) Anslyn, E. V; Dougherty, D. A. Modern Physical Organic
Chemistry; University Science Books: Sausalito, CA, 2006.
17) Suto, Y.; Kumagai, N.; Matsunaga, S.; Kanai, M.; Shibasaki, M.
Org. Lett. 2003, 5, 3147−3150.
18) Suto, Y.; Tsuji, R.; Kanai, M.; Shibasaki, M. Org. Lett. 2005, 7,
757−3760.
19) Goto, A.; Endo, K.; Ukai, Y.; Irle, S.; Saito, S. Chem. Commun.
(
1
(
(
(
(
3
(
2
(
008, 2212−2214.
20) Sureshkumar, D.; Ganesh, V.; Kumagai, N.; Shibasaki, M. Chem.
-
Eur. J. 2014, 20, 15723−15726.
(
21) Kumagai, N.; Matsunaga, S.; Shibasaki, M. J. Am. Chem. Soc.
2
(
004, 126, 13632−13633.
22) Kumagai, N.; Matsunaga, S.; Shibasaki, M. Chem. Commun.
2
(
(
005, 3600−3602.
23) Fan, L.; Ozerov, O. V. Chem. Commun. 2005, 4450−4452.
24) Chakraborty, S.; Patel, Y. J.; Krause, J. A.; Guan, H. Angew.
Chem., Int. Ed. 2013, 52, 7523−7526.
(
25) Kamila, S.; Zhu, D.; Biehl, E. R.; Hua, L. Org. Lett. 2006, 8,
4
(
(
429−4431.
26) Fukuda, Y.; Okamoto, Y. Tetrahedron 2002, 58, 2513−2521.
27) Ankati, H.; Zhu, D.; Yang, Y.; Biehl, E. R.; Hua, L. J. Org. Chem.
2
(
(
009, 74, 1658−1662.
28) Lapointe, D.; Fagnou, K. Chem. Lett. 2010, 39, 1118−1126.
29) Kita, M. R.; Miller, A. J. M. J. Am. Chem. Soc. 2014, 136, 14519−
1
4529.
(
30) Butts, S. B.; Strauss, S. H.; Holt, E. M.; Stimson, R. E.; Alcock,
N. W.; Shriver, D. F. J. Am. Chem. Soc. 1980, 102, 5093−5100.
31) Hazari, A.; Labinger, J. A.; Bercaw, J. E. Angew. Chem., Int. Ed.
012, 51, 8268−8271.
(
2
I
DOI: 10.1021/acs.organomet.5b00405
Organometallics XXXX, XXX, XXX−XXX