Carbon dioxide cleavage by a Ni2 complex supported by a binucleating bis(N-heterocyclic carbene) framework Dedicated to Prof. John E. Bercaw on the occasion of his 70th birthday.

Article abstract of DOI:10.1016/j.poly.2014.06.041

A binucleating bis(N-heterocyclic carbene) ligand was designed as a means to coordinate and proximally constrain two transition metal centers. Using an imidazopyridine-based NHC afforded a framework structurally related to previously reported para-terphenyl diphosphines. Bimetallic copper, cobalt, and nickel complexes supported by this framework were synthesized and structurally characterized. Strong interactions between the metal centers and the central arene were observed in all nickel complexes. Dinickel(0) complexes of this ligand framework were found to react with CO₂ to form a dicarbonyl-bridged dinickel(0) product, demonstrating facile CO₂ reduction.

Full text of DOI:10.1016/j.poly.2014.06.041

Polyhedron 84 (2014) 103–110
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Carbon dioxide cleavage by a Ni2 complex supported by a binucleating
bis(N-heterocyclic carbene) framework
⇑
Emily Y. Tsui, Theodor Agapie
Division of Chemistry and Chemical Engineering, California Institute of Technology, 1200 E. California Blvd. MC 127-72, Pasadena, CA 91125, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A binucleating bis(N-heterocyclic carbene) ligand was designed as a means to coordinate and proximally
constrain two transition metal centers. Using an imidazopyridine-based NHC afforded a framework
structurally related to previously reported para-terphenyl diphosphines. Bimetallic copper, cobalt, and
nickel complexes supported by this framework were synthesized and structurally characterized. Strong
interactions between the metal centers and the central arene were observed in all nickel complexes.
Dinickel(0) complexes of this ligand framework were found to react with CO2 to form a dicarbonyl-
bridged dinickel(0) product, demonstrating facile CO2 reduction.
Received 6 May 2014
Accepted 18 June 2014
Available online 2 July 2014
Dedicated to Prof. John E. Bercaw on the
occasion of his 70th birthday.
Ó 2014 Elsevier Ltd. All rights reserved.
Keywords:
Bimetallic complexes
NHC
Carbon dioxide activation
Nickel
Metal-arene interactions
1. Introduction
different donor properties of the NHC donors compared to phos-
phines. In this work, we describe the synthesis of this ligand frame-
In recent years, multimetallic complexes have become attrac-
tive synthetic targets due to their potential uses in chemical trans-
formations involving the transfer of multiple electrons through
cooperative substrate binding and functionalization, particularly
as several enzymatic active sites are known to contain multiple
metal centers [1]. As such, there have been many efforts to design
and synthesize new binucleating ligand frameworks [2]. Our group
has used a para-terphenyl diphosphine ligand (A) to support a vari-
ety of binuclear Ni (Fig. 1, left), Pd, Fe, and Co complexes [3]. Labile
metal-arene interactions between the metal centers and the
central arene moiety have been observed in these compounds,
with the proposed effect of stabilizing and electronically tuning
these binuclear moieties.
Over the past two decades, N-heterocyclic carbene (NHC)
ligands have gained prominence in organometallic chemistry and
catalysis since they were first popularized by Arduengo and
co-workers [5]. There have been many examples of polydentate
NHC ligand frameworks; although these have primarily been che-
lating ligands that support a single metal center [6], some support
multimetallic complexes [7]. A new dinucleating NHC ligand
framework (Fig. 1, right) containing an aryl linker was designed
to incorporate both the hemilabile arene moiety of A and the
work and of its binuclear first row transition metal complexes, as
well as initial reactivity studies.
2. Experimental
2.1. General considerations
Unless stated otherwise, all synthetic manipulations were car-
ried out using standard Schlenk techniques under a nitrogen atmo-
sphere, or in a M. Braun glovebox under a nitrogen atmosphere.
Reactions were carried out in oven-dried glassware cooled under
vacuum. Anhydrous THF was purchased from Aldrich in 18 L
Pure-Pac™ containers. Anhydrous benzene, hexanes, pentane,
diethyl ether, and THF were purified by sparging with nitrogen
for 15 min and then passing under nitrogen pressure through a col-
umn of activated A2 alumina (Zapp’s). All non-dried solvents used
were reagent grade or better. Deuterated solvents were purchased
from Cambridge Isotope Laboratories, Inc. C6D6 was dried over
sodium/benzophenone ketyl, degassed by three freeze–pump–
thaw cycles and vacuum-transferred prior to use. Deuterated
dichloromethane (CD2Cl2) was dried over calcium hydride,
degassed by three freeze–pump–thaw cycles, and vacuum-trans-
ferred prior to use. 1H NMR and 13C NMR spectra were recorded
on Varian 300, 400, or 500 MHz instruments, with shifts reported
relative to the residual solvent peaks (7.16 and 128.06 ppm
⇑
Corresponding author. Tel.: +1 626 395 3617.
E-mail address: agapie@caltech.edu (T. Agapie).
http://dx.doi.org/10.1016/j.poly.2014.06.041
0277-5387/Ó 2014 Elsevier Ltd. All rights reserved.

104
E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
Fig. 1. Dinickel(I) dichloride complexes supported by diphosphine A (1, left) [3a] and by N-heterocyclic carbene ligand 1,3-bis(2,6-diisopropylphenyl)imidazol-2-ylidene (IPr,
center) [4]. This work: dinucleating bis(N-heterocyclic carbene) framework DIPr (right).
respectively for C6D6, 5.32 and 53.84 ppm respectively for CD2Cl2,
and 7.27 and 77.16 pm respectively for CDCl3). 19F NMR spectra
were referenced relative to the lock signal of the solvent. Elemental
analyses were performed by Robertson Microlit, Ledgewood, NJ.
High resolution mass spectrometry (HRMS) was performed at the
California Institute of Technology Mass Spectrometry Facility. IR
spectra were collected as thin films on a Bruker Alpha FT-IR
spectrometer with a diamond ATR attachment.
Unless indicated otherwise, all reagents were used as received.
Silver triflate was purchased from Alfa Aesar. Chloromethyl piva-
late was purchased from Alfa Aesar and distilled from calcium
hydride. Copper(I) chloride, nickel bis(cyclooctadiene), sodium
tert-butoxide, and nickel(II) chloride dimethoxyethane adduct
were purchased from Strem. Co[N(SiMe3)2]2 [8], 1,4-benzenedibo-
ronic acid [9] and 2-(2,6-diisopropylphenyl)imine-6-bromopyri-
dine [10] were prepared according to literature procedures.
(200 mL), ethanol (40 mL), and water (40 mL). The mixture was
degassed via three freeze–pump–thaw cycles, then Pd(PPh3)4
(0.735 g, 0.636 mmol) was added under N2. The flask was sealed
with a Teflon stopper, then heated to 100 °C for 22 h. The mixture
was cooled to room temperature, then the organics were extracted
with ethyl acetate and dichloromethane, dried over magnesium
sulfate, filtered, then concentrated under reduced pressure. The
resulting brown solid was recrystallized from diethyl ether/
dichloromethane to yield an off-white solid. The solid was dissolved
in dichloromethane (50 mL), then tetrabutylammonium chloride
(2.75 g) was added. The mixture was stirred for 2 h, then concen-
trated under reduced pressure. The residue was taken up in diethyl
ether/THF, and the solid was collected, washed with THF, then
recrystallized from diethyl ether/chloroform to yield the product
as a white solid (2.43 g, 27%). 1H NMR (CDCl3, 300 MHz): d 12.19
(s, 2 H, NC(H)N), 7.99 (d, J = 9.3 Hz, 2 H, Ar–H), 7.87 (s, 4 H, central
arene), 7.85 (d, J = 1.5 Hz, 2 H, Ar–H), 7.56 (t, J = 7.8 Hz, 2 H, Ar–H),
7.49 (dd, J = 9.3, 6.9 Hz, 2 H, Ar–H), 7.30 (d, J = 7.8 Hz, 4 H, Ar–H),
7.25 (d, J = 6.9 Hz, 2 H, Ar–H), 2.06 (sept, J = 6.9 Hz, 4 H, CH(CH3)2),
1.15 (d, J = 6.9 Hz, 12 H, CH3), 1.11 (d, J = 6.9 Hz, 12 H, CH3). 13C
NMR (CD2Cl2, 75.463 MHz): d 141.7, 135.8, 134.4, 133.6, 131.80,
131.75, 131.0, 129.8, 129.1, 126.0, 118.8, 118.7, 115.7, 21.3,
17.8 ppm. HRMS (FAB+) for C44H48N4Cl: 667.3567. Found: 667.3560.
2.2. Synthesis of compounds
2.2.1. Preparation of 5-bromo-2-(20,60-
diisopropylphenyl)imidazo[1,5a]pyridinium triflate
In the glovebox, silver trifluoromethanesulfonate (5.21 g,
20.27 mmol) was suspended in methylene chloride (15 mL) in a
scintillation vial equipped with a stir bar. Chloromethyl pivalate
(2.92 mL, 20.27 mmol) was added via syringe, and the mixture
was stirred at room temperature in the dark for 45 min., changing
to a dark red mixture. The mixture was filtered through Celite to
remove silver chloride and washed with methylene chloride, then
added to a solution of 2-(2,6-diisopropylphenyl)imine-6-bromo-
pyridine (5.0 g, 14.48 mmol) in methylene chloride (5 mL). The
dark red-brown solution was transferred to a reaction bomb
equipped with a stir bar that was then sealed with a Teflon stopper.
The reaction was stirred at 45 °C in the dark for 16 h, then cooled to
room temperature. Ethanol (10 mL) was added to quench the reac-
tion and the mixture was concentrated in vacuo to yield a viscous
brown oil that crystallizes upon standing. The solid was washed
with diethyl ether and then recrystallized from diethyl ether/
methylene chloride to yield the product as a white solid (5.27 g,
72%). 1H NMR (CDCl3, 300 MHz): d 9.24 (d, J = 1.8 Hz, 1 H, NC(H)N),
8.48 (d, J = 1.8 Hz, 1 H), 8.33 (d, J = 9.0 Hz, 1 H, o-Br), 7.63 (t,
J = 8.0 Hz, 1 H, m-Br), 7.57 (d, J = 7.2 Hz, 1 H), 7.407.33 (m, 3 H),
2.14 (sept, J = 6.6 Hz, 2 H, CH(CH3)2), 1.22 (d, J = 6.6 Hz, 6 H, CH3),
1.21 (d, J = 6.6 Hz, 6 H, CH3) ppm. 13C NMR (126 MHz, CDCl3) d
145.2, 132.7, 132.6, 130.4, 126.6, 125.7, 125.0, 124.0, 119.7,
119.5, 112.0, 28.9, 24.6, 24.4 ppm. HRMS (FAB+, m/z) for
2.2.3. Preparation of DIPr[CuCl]2 (2)
In the glovebox, a scintillation vial equipped with a stir bar was
charged with DIPrꢀ2HCl (0.040 g, 0.057 mmol), sodium tert-butox-
ide (0.011 g, 0.11 mmol, 2 equiv), and copper(I) chloride (0.011 g,
0.11 mmol, 2 equiv). THF (5 mL) was added, and the yellow mix-
ture was stirred at room temperature for 2 h. The mixture was fil-
tered through Celite to remove an insoluble yellow precipitate, and
the yellow filtrate was concentrated in vacuo. The mixture was
washed with benzene, then extracted with THF and concentrated
again under reduced pressure to yield a yellow solid. This solid
was washed with acetonitrile to yield the product as a yellow solid
(0.011 g, 23%). X-ray quality single crystals were grown by vapor
diffusion of diethyl ether into a dichloromethane solution of 2.
1H NMR (CD2Cl2, 400 MHz): d 7.79 (s, 4 H, central arene), 7.52 (d,
J = 7.6 Hz, 2 H, Ar–H), 7.48 (d, J = 9.6 Hz, 2 H, Ar–H), 7.38 (s, 2 H,
benzylic H), 7.30 (d, J = 8.0 Hz, 4 H, Ar–H), 7.13–7.09 (m, 4 H,
Ar–H), 2.26 (sept, J = 6.8 Hz, 4 H, CH(CH3)2), 1.20 (d, J = 6.8 H, 12
H, CH3), 1.14 (d, J = 6.8 Hz, 12 H, CH3) ppm. 13C NMR (CD2Cl2,
101 MHz): d 145.8, 139.4, 136.3, 136.2, 132.1, 130.1, 130.7, 124.4,
124.3, 117.3, 116.6, 114.3, 28.7, 24.7, 24.4 ppm. Anal. Calc. For
C44H46Cl2Cu2N4: C, 63.76; H, 5.59; N, 6.76. Found: C, 63.51; H,
C18H22BrN2: 357.0966. Found: 357.0951.
5.31; N, 6.51%.
2.2.2. Preparation of DIPrꢀ2HCl
2.2.4. Preparation of DIPr[Co(Cl)N(SiMe3)2]2 (3)
A reaction bomb equipped with a stir bar was charged with
5-bromo-N-(2,6-diisopropylphenyl)imidazo[1,5-a]pyridinium tri-
flate (12.92 g, 25.45 mmol), 1,4-benzenediboronic acid (2.11 g,
12.73 mmol), potassium carbonate (10.55 g, 76.36 mmol), toluene
In the glovebox, a scintillation vial equipped with a stir bar was
charged with DIPrꢀ2HCl (0.046 g, 0.065 mmol) and Co[N(SiMe3)2]2
(0.050 g, 0.13 mmol, 2 equiv). Benzene (6 mL) was added, and the
green mixture was stirred for 12 h, then filtered through Celite

E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
105
and concentrated under reduced pressure. The yellow–brown res-
idue was washed with pentane, then extracted with diethyl ether
and concentrated in vacuo, A benzene solution was allowed to
stand at room temperature, slowly forming the product as yellow
with solvent residual peak, identified by gCOSY), 7.07 (d,
J = 7.5 Hz, 2 H, Ar–H), 6.99 (d, J = 7.5 Hz, 2 H, Ar–H), 6.89 (d,
J = 6.5 Hz, 2 H, Ar–H), 6.85 (d, J = 9.0 Hz, 2 H, Ar–H), 6.80 (s, 2 H,
Ar–H), 6.56 (t, J = 8.0 Hz, 2 H, Ar–H), 5.56 (s, 2 H, central arene
H), 3.53 (s, 2 H, central arene H), 2.63 (sept, J = 7.0 Hz, 2 H,
crystals (0.008 g, 11%). 1H NMR (C6D6, 300 MHz):
d
89.0
1/2 = 53 Hz), 10.1
1/2 = 442 Hz), ꢁ7.2
1/2 = 115 Hz), ꢁ51.1 1/2 = 42 Hz),
1/2 = 324 Hz), ꢁ70.0 ( 1/2 = 942 Hz) ppm.
(
(
(
D
D
D
m
m
m
1/2 = 65 Hz), 64.6 (
1/2 = 104 Hz), 3.3 (
1/2 = 26 Hz), ꢁ15.2
D
m
1/2 = 90 Hz), 36.1 (
D
m
m
CH(CH3)2), 2.17 (sept, J = 7.0 Hz, 2 H, CH(CH3)2), 1.15 (d,
D
m1/2 = 371 Hz), ꢁ2.3 (
D
J = 7.0 Hz, 6 H, CH3), 1.01 (d, J = 7.0 Hz, 6 H, CH3), 0.93 (d,
J = 7.0 Hz, 6 H, CH3), 0.91 (d, J = 7.0 Hz, 6 H, CH3) ppm. 13C NMR
(C6D6, 126 MHz): d 129.4, 147.6, 147.1, 145.4, 137.3, 130.7, 129.7,
125.3, 123.8, 123.5, 112.8, 112.4, 106.8, 105.7, 97.3, 42.4, 28.6,
28.5, 26.7, 24.7, 24.2, 22.6 ppm. Compound 6 is unstable at room
temperature and decomposes in solution and in the solid-state
over several hours.
Procedure B: In the glovebox, a Schlenk flask equipped with a
stir bar was charged with a benzene solution of 5. The solution
was degassed with three freeze–pump–thaw cycles, then placed
under an atmosphere of dihydrogen. The reaction mixture was stir-
red for 30 min. at room temperature, then dried in vacuo. The
resulting brown solid was washed with pentane to yield the prod-
uct as a brown solid (0.023 g, 82%).
(
D
m
(Dm
ꢁ54.8 (
D
m
Dm
2.2.5. Preparation of DIPr[Ni( -Cl)]2 (4)
l
In the glovebox, a round-bottom flask equipped with a stir bar
was charged with DIPrꢀ2HCl (1.18 g, 1.68 mmol), sodium tert-
butoxide (0.327 g, 3.4 mmol), nickel bis(cyclooctadiene) (0.459 g,
1.67 mmol), and nickel dichloride dimethoxyethane adduct
(0.367 g, 1.67 mmol). THF (100 mL) was added, and the brown
mixture was stirred at room temperature for 12 h. The dark brown
mixture was filtered through Celite, washing with THF (50 mL),
and the filtrate was concentrated in vacuo. The brown residue
was taken up in benzene (60 mL), filtered through Celite, then con-
centrated in vacuo. The resulting brown solid was washed with
pentane (30 mL), and dried under vacuum to yield the product as
a brown solid (0.300 g, 22%). X-ray quality single crystals were
grown by vapor diffusion of pentane into a diethyl ether solution
of 4 at room temperature. 1H NMR (C6D6, 500 MHz): d 14.34 (s, 4
H, central arene), 8.49 (s, 2 H, benzylic H), 7.59 (d, J = 7.0 Hz, 4 H,
Ar–H), 7.53 (t, J = 7.0 Hz, 2 H, Ar–H), 7.19 (d, J = 9.5 Hz, 2 H, Ar–
H), 6.94 (d, J = 6.5 Hz, 2 H, Ar–H), 5.86 (dd, J = 6.5, 9.5 Hz, 2 H,
Ar–H), 2.86 (sept, J = 6.5 Hz, 4 H, CH(CH3)2), 1.64 (d, J = 6.5 Hz, 12
H, CH3), 1.12 (d, J = 6.5 Hz, 12 H, CH3) ppm. 13C NMR (C6D6,
126 MHz): d 157.4, 150.7, 150.0, 149.4, 148.7, 147.9, 147.4, 147.2,
131.1, 125.0, 122.3, 117.4, 109.3, 33.18, 27.8, 25.6 ppm.
2.2.8. Preparation of DIPr[Ni(l-CO)]2 (7)
Procedure A: In the glovebox, a J. Young NMR tube was charged
with a solution of 6 (0.017 g, 0.019 mmol) in C6D6 (0.7 mL). The
solution was degassed via three freeze–pump-thaw cycles, then
placed under an atmosphere of CO2. The reaction was allowed to
proceed at room temperature, and monitored by 1H NMR spectros-
copy. Compound 7 was identified as the major species in the mix-
ture by 1H NMR spectroscopy after 20 h.
Procedure B: In the glovebox, a scintillation vial equipped with a
stir bar was charged with 5 (0.021 g, 0.022 mmol) and paraformal-
dehyde (0.015 g, 0.044 mmol, 2 equiv). THF (5 mL) was added, and
the brown mixture was stirred at room temperature for 1 h, turning
dark purple. The mixture was concentrated in vacuo, then crystal-
lized from diethyl ether/pentane to yield the product as a brown-
purple solid (0.008 g, 47%). 1H NMR (C6D6, 500 MHz): d 7.10 (t,
J = 7.0 Hz, 2 H, Ar–H), 7.03 (d, J = 7.0 Hz, 4 H, Ar–H), 6.81 (s, 2 H,
Ar–H), 6.74 (dd, J = 2.0, 7.8 Hz, 2 H, Ar–H), 6.38 (m, 4 H, Ar–H),
6.24 (bs, 4 H, central arene), 2.64 (sept, J = 7.0 Hz, 4 H, CH(CH3)2),
1.27 (d, J = 7.0 Hz, 12 H, CH3), 0.99 (d, J = 7.0 Hz, 12 H, CH3) ppm.
13C NMR (C6D6, 126 MHz): d 184.9, 145.5, 142.9, 137.4, 129.8,
127.2, 124.1, 123.8, 115.0, 114.3, 112.5, 109.9, 98.3 (bs, located at
50 °C). 28.6, 24.8, 23.8 ppm. The 13C NMR signal corresponding to
the carbonyl ligands could not be located, presumably due to flux-
ional exchange processes. FT-IR (ATR film, C6D6): 1825 cmꢁ1 (CO).
2.2.6. Preparation of DIPr[Ni(COD)]2 (5)
In the glovebox, an oven-dried round-bottom flask equipped
with a stir bar was charged with DIPrꢀ2HCl (0.200 g, 0.284 mmol),
sodium tert-butoxide (0.055 g, 0.572 mmol, 2 equiv), and Ni(COD)2
(0.150 g, 0.546 mmol, 1.9 equiv). Benzene (40 mL) was added, and
the mixture was stirred for 90 min. at room temperature, turning
from yellow to green to dark brown. The mixture was then filtered
through Celite, then concentrated in vacuo to yield a dark brown
solid. This residue was washed with hexanes (50 mL) and diethyl
ether (10 mL), then dried under vacuum to yield the product as a
dark brown solid (0.150 g, 55%). Due to the equilibrium between
5 and 6 in benzene solution, NMR spectra were collected after
the addition of 2 equivalents of excess 1,5-cyclooctadiene. 1H
NMR (C6D6, 400 MHz): d 7.26 (t, J = 7.6 Hz, 2 H, Ar–H), 7.13 (d,
J = 7.6 Hz, 4 H, Ar–H), 6.79 (s, 2 H, A–H), 6.65 (d, J = 8.8 Hz, 2 H,
Ar–H), 6.45 (d, J = 6.4, 10.4 Hz, 2 H, Ar–H), 6.34 (d, J = 6.8 Hz, 2 H,
Ar–H), 6.25 (bs, 4 H, central arene), 4.2 (bs, 8 H, Ni(COD) vinyl
H), 2.82 (sept, J = 6.8 Hz, 4 H, CH(CH3)2), 1.9 (bs, 12 H, Ni(COD)
methylene H), 1.28 (d, J = 6.8 Hz, 12 H, CH3), 0.96 (d, J = 6.8 Hz,
12 H, CH3) ppm. 13C NMR (C6D6, 100.5 MHz): d 194.33, 148.09,
146.27, 138.10, 130.78, 129.74, 128.83, 125.14, 123.75, 112.49,
111.88, 106.14, 81.04, 30.70, 28.72, 26.54, 22.36 ppm.
3. Results and discussion
3.1. Synthesis of ligand framework
The bis(carbene) precursor, abbreviated DIPrꢀ2HCl, was pre-
pared in three steps from commercially available starting materials
based upon a previously published route [11]. Condensation of 2-
bromo-6-pyridinecarboxaldehyde with 2,6-diisopropylaniline
yielded the corresponding imine (Scheme 1). Cyclization by treat-
ment with silver triflate and chloromethyl pivalate formed the
bromo-substituted imidazo[1,5-a]pyridinium salt. Suzuki coupling
of this imidazo[1,5-a]pyridinium triflate with 1,4-benzenediboron-
ic acid, followed by anion substitution using tetrabutylammonium
chloride, yielded the ligand precursor DIPrꢀ2HCl.
2.2.7. Preparation of DIPr[Ni(C6H6)]2 (6)
Procedure A: In the glovebox, a scintillation vial equipped with
a stir bar was charged with a solution of 4 (0.272 g, 0.333 mmol) in
THF (10 mL). A 1 M THF solution of LiBEt3H (0.67 mL, 0.666 mmol)
was added via syringe. The mixture was stirred at room tempera-
ture for 10 min. and the mixture was concentrated in vacuo. The
brown residue was taken up in benzene, filtered through Celite,
then concentrated in vacuo. The resulting solid was washed with
hexanes and diethyl ether to yield the product as a brown solid
(0.074 g, 25%). 1H NMR (C6D6, 500 MHz): d 7.15 (2 H, overlapped
3.2. Metallation with copper(I) and cobalt(II)
A small number of monometallic complexes supported by
mono-NHC versions of DIPr are known, with Cu, Pd, Ir, Rh, and

106
E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
Scheme 1. Synthesis of DIPrꢀ2HCl.
Au [11,12]. With the ligand precursor DIPrꢀ2HCl in hand, bimetallic
complexes of the bis(carbene) framework were targeted. Attempts
to deprotonate DIPrꢀ2HCl using bases such as sodium tert-butox-
ide, sodium hydride, or sodium bis(trimethylsilyl)amide to yield
the free bis(carbene) resulted in intractable mixtures of insoluble
products, possibly due to decomposition of the free carbene. Fortu-
nately, one-pot metallation reactions of DIPrꢀ2HCl proved to be
more successful. Treatment of DIPrꢀ2HCl with two equivalents of
sodium tert-butoxide and two equivalents of copper(I) chloride in
THF formed the corresponding dicopper(I) dichloride compound
DIPr[CuCl]2 (2, Scheme 2). X-ray quality crystals were grown from
vapor diffusion of diethyl ether into a dichloromethane solution of
2. A X-ray diffraction (XRD) study of these crystals (Fig. 2 left)
showed that the two copper(I) centers are oriented on opposite
faces of the central arene, and are coordinated in a linear geometry
(C–Cu–Cl angle of 175.3(2)°). Compound 2 is structurally related to
a monomeric and less sterically hindered version [12d], whose
solid-state structure displays a Cu–Cl distance of 2.102 Å and a
Cu–C distance of 1.875 Å (average of two molecules in the asym-
metric unit). In comparison, compound 2 has a Cu–Cl distance of
Scheme 2. Synthesis of dicopper(I) compound 2 and dicobalt(II) compound 3.

E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
107
Fig. 2. Solid-state structures of 2 (left) and 3 (right) as 50% thermal ellipsoids. Hydrogen atoms and solvent molecules not shown for clarity. Structure of 2 contains a mixture
of bromide and chloride ligands.
2.210(5) Å and a Cu–C distance of 1.886(2) Å. Although the imi-
dazopyridine-based NHC has higher donor ability [12b], the above
structures are similar to that of a copper complex supported by an
imidazole-based NHC (IPrCuCl) [13], with a Cu–Cl distance of
2.106(2) Å and a C–Cu distance of 1.881(7) Å. No interactions with
the central arene are observed in the XRD structure (shortest
Cu–C(arene) distance 3.06 Å); this is also borne out in the 1H
NMR spectrum, in which the signal corresponding to the central
arene protons is located at 7.79 ppm.
Rather than use an exogenous base, DIPrꢀ2HCl can also be met-
allated via aminolysis of metal amide compounds. The addition of
two equivalents of Co[N(SiMe3)2]2 to DIPrꢀ2HCl in benzene results,
after two days at room temperature, in yellow crystals of the
dicobalt(II) compound DIPr[Co(Cl)N(SiMe3)2]2 (3, Fig. 2 right). As
with 2, the two metal centers are oriented on opposite faces of
the central arene and no metal-arene interactions are observed
(shortest Co–C(arene) distance 2.95 Å). The coordination sphere
of each cobalt(II) center is completed by a single chloride and
bis(trimethylsilyl)amide ligand. A related monocobalt complex
with Clꢁ, N(SiMe3)ꢁ2 , and imidazole-based NHC ligands has previ-
ously been isolated and structurally characterized.[14] The 1H
NMR spectrum of a C6D6 solution of 3 displays broad peaks between
-70 and 90 ppm, consistent with a paramagnetic compound.
DIPr[Ni(l-Cl)]2 (4) as a dark brown solid (Scheme 3). As with both
the NHC-supported and diphosphine-supported complexes, the 1H
NMR spectrum of 4 shows peaks in the diamagnetic region, consis-
tent with a NiI–NiI interaction. The signal corresponding to the cen-
tral arene protons is a downshifted singlet at 14.24 ppm. In
comparison, the central arene 1H NMR signal of the corresponding
diphosphine-supported dinickel(I) dichloride complex 1 is shifted
upfield from that of free ligand [3a], indicating differences in the
nickel-arene interactions. The observation of a singlet for the cen-
tral arene and of only one methine peak for the isopropyl moieties,
suggests that the dinickel(I) unit is involved in a fluxional process
that exchanges positions over the central arene fast on the NMR
timescale at room temperature.
Single crystals of 4 were grown by vapor diffusion of pentane
into a diethyl ether solution of the complex. A XRD study shows
that the two nickel(I) centers are located on the same face of the
arene ring, with the two chlorides bridging the metal centers,
and a Ni–Ni distance of 2.3936(3) Å (Fig. 3), shorter than previously
reported NiI2
(l-Cl)2 moieties supported by imidazole-NHC (Ni–
Ni ꢂ 2.51 Å) [4]. The shorter NiꢁNi distance in 4 may be a conse-
quence of the geometric constraints of the ligand framework that
distorts at the aryl-aryl linkages due to the cofacial bimetallic moi-
ety. There are strong interactions between the nickel centers and
the central arene moiety, with short CꢁNi distances (2.05–2.17 Å).
Additionally, the C–C bond distances of the central arene show sig-
nificant bond localization (Fig. 3 inset). The solid-state structure of
compound 1 shows similar nickel-arene interactions (Ni–C distances
2.05–2.10 Å) and central arene bond localization distances. Due to
the longer Ni–P distances (2.17–2.18 Å) compared to the NiꢁC(car-
bene) distances (1.91 Å), the terphenyl backbone of the diphosphine
ligand is more distorted to accommodate the dinickel moiety, with
the outer aryl rings of the diphosphine bent 15–17° out of the plane
vs. 9–10° for 4. Compound 4 displays a torsion angle of 62.72–68.75°
between the central arene and the rings at the periphery, while the
pendant aryl groups of 1 are oriented closer to perpendicular to the
central arene, with torsion angles of 74.4 and 78.8°. As a result, the
nickel(I) centers of 4 are located further toward the C–H bonds com-
pared to 1, potentially causing the differences in the 1H NMR spectra.
The NiꢁNi distance is longer in 4 (2.3936(3) Å) compared to 1
3.3. Synthesis of dinickel complexes
Although compounds 2 and 3 are bimetallic, the metal centers
are located on opposite faces of the central arene, potentially lim-
iting access to cooperative reactivity. To test the ability of ligand
DIPr to support metals in close proximity with potential for bime-
tallic reactivity, species displaying metal–metal interactions were
targeted. Both NHC ligands and diphosphine A have been shown
to support dinickel(I) dichloride complexes containing a NiI–NiI
bond. These complexes were prepared by the comproportionation
of NiCl2(dme) (dme = 1,2-dimethoxyethane) and Ni(COD)2
(COD = 1,5-cyclooctadiene) (Fig. 1, left and center)[3a,4]. Treatment
of a THF suspension of DIPrꢀ2HCl with sodium tert-butoxide (2
equiv) and equimolar amounts of Ni(COD)2 and NiCl2(dme)
resulted in the formation of comproportionation product

108
E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
Scheme 3. Synthesis of dinickel complexes.
Fig. 3. Solid-state structure of DIPr[Ni(
l-Cl)]2 (4) as 50% thermal ellipsoids. Hydrogen atoms and solvent molecules not shown for clarity. Inset: selected bond distances (Å).
(2.3658(2) Å), possibly a consequence of the differences in the ligand
distortions, but also of the different electronic character of the
ligands [12b].
on opposite faces of the central arene, each displaying an
2-interaction with the vicinal diene of the central arene moiety
(NiꢁC distances 2.13–2.15 Å).
g
More reduced dinickel(0) compounds were also targeted, as a
previous example of a NHC-supported dinickel(0) compound was
demonstrated to be capable of CO2 deoxygenation, presumably
by the formation of an equivalent of a nickel carbonate species
[15]. A benzene suspension of DIPrꢀ2HCl was treated with sodium
tert-butoxide and Ni(COD)2 to form a dark brown compound 5
(Scheme 3). The 1H NMR spectrum of a C6D6 solution of 5 displays
two broad signals at 4.2 and 1.9 ppm corresponding to the protons
of the coordinated 1,5-cyclooctadiene ligands. Additionally, the
signal corresponding to the central arene protons is shifted upfield,
to 6.25 ppm, consistent with interactions between the nickel(0)
centers and the central arene. A XRD study of single crystals of 5
grown from THF/hexanes at ꢁ35 °C (Fig. 4) confirmed the structure
to be a dinickel(0) complex in which the metal centers are located
In benzene solution, compound 5 was observed to be in equilib-
rium with a second DIPr ligand-containing complex (6), along with
free COD. Addition of excess COD resulted in complete conversion
to 5, however. Compound 6 is proposed to be the benzene adduct
of the dinickel(0) complex. Addition of excess dihydrogen to a ben-
zene solution of 5 facilitates hydrogenation of COD and formation
of clean 6 (1H NMR spectroscopy). Compound 6 can also be inde-
pendently prepared by reduction of the dinickel(I) compound 4
with two equivalents of LiBEt3H or two equivalents of sodium
naphthalenide in THF, followed by benzene. Unfortunately, com-
pound 6 is unstable at room temperature in solution and in the
solid-state, decomposing within several hours. As such, it has been
challenging to grow X-ray quality single crystals of 6, and struc-
tural confirmation of its assignment via XRD has not been achieved

E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
109
These differences may be a consequence of the structural parame-
ters of ligands A and DIPr. The shorter M–C compared to M–P bond
distances and the steric profile of the para-terphenyl linker are
expected to disfavor mononickel(0) complexes with DIPr vs A.
3.4. CO2 reactivity studies
Nickel complexes have long been examined for carbon dioxide
reduction activity due to the proposed role of the nickel center in
carbon monoxide dehydrogenase [17]. The first known example
of a nickel carbon dioxide complex, (Cy3P)2Ni(CO2) was prepared
by Aresta and co-workers in 1975 [18]. In this example, the CO2
complex was found to yield small amounts of carbonyl and carbon-
ate products upon heating via the metal-mediated disproportion-
ation of CO2. Other examples of tri- and diphosphine supported
nickel(0) complexes have been demonstrated to cleave CO2 via
the formation of nickel carbonyl and carbonate complexes, or via
the oxidation of the phosphine ligands to form phosphine oxides
[19]. A recent report by Caulton and co-workers described the
cleavage of the C–O bond of CO2 by a PNP pincer-supported nick-
el(I) complex to transpose an oxygen center from CO2 with the
amide nitrogen of the ligand framework [20]. Multinickel com-
plexes have also been studied for the reduction of carbon dioxide.
Sadighi and co-workers reported a dinickel(0) compound that
reacts with CO2 to form a carbonyl-bridged dinickel complex with
proposed concomitant loss of an unobserved nickel carbonate spe-
cies [15]. A recent example of dinickel(I) complexes supported by
b-diketiminate ligands demonstrated the formation of carbonyl,
carbonate, and oxalate nickel complexes [21]. Dinickel and trinick-
el complexes have also been reported as effective electrocatalysts
in the reduction of CO2 to CO [22].
Fig. 4. Solid-state structure of DIPr[Ni(COD)2] (5) as 50% thermal ellipsoids.
Hydrogen atoms and outer sphere solvent molecules not shown for clarity.
Upon treatment of benzene solutions of 5 or 6 with excess CO2,
complete consumption of starting material was observed by 1H
NMR spectroscopy within minutes to form one major species that
has eluded isolation thus far. Over 7 h at room temperature, this
species converts to a mixture of products, with one major species
(7) observed by 1H NMR spectroscopy. The 1H NMR spectrum of
7 displays one signal for the isopropyl methine protons and a broad
singlet at 6.2 ppm corresponding to the central arene ring, indicat-
ing nickel-arene interactions, albeit weaker than those observed in
6. An IR spectrum of 7 shows a strong band at 1825 cmꢁ1, consis-
Scheme 4. Reaction of dinickel(0) complexes 5 and 6 with CO2.
yet. The 1H NMR spectrum of a C6D6 solution of 6 displays two
singlets each integrating to 2 protons corresponding to the central
arene at 5.56 and 3.53 ppm, both shifted far upfield relative to 5,
consistent with greater backbonding into the central arene ligand
due to weaker backbonding into the coordinated benzene ligands
compared to COD. The observation of two central arene signals, as
well as two isopropyl methine signals, indicates that the fluxional
motion of the nickel centers is slower than that of 4 and 5, again
consistent with stronger metal-arene interactions. In comparison,
reduction of phosphine-based system 1 or metallation of A with
nickel(0) sources has typically yielded only a mononickel(0) com-
plex rather than dinickel complexes, except in cases in which the
compounds are stabilized via bridging carbonyl ligands [3d,16].
tent with the stretching vibration of
a bridging CO moiety
(Scheme 4) [23]. When 13CO2 was added to 5, a signal at
171 ppm was initially observed in the 13C NMR spectrum (see Sup-
porting information), possibly indicating a carbon dioxide complex
of nickel, while the final product mixture showed two sharp signals
Fig. 5. Solid-state structure of 7 as 50% thermal ellipsoids. Hydrogen atoms and solvent molecules not shown for clarity. Inset: selected bond distances (Å).

110
E.Y. Tsui, T. Agapie / Polyhedron 84 (2014) 103–110
at 161 ppm, consistent with nickel carbonate complexes [24]. The
13C NMR signal corresponding to the bridging carbonyl ligands of 7
could not be observed by 13C NMR spectroscopy even at elevated
temperatures, possibly due to fluxional exchange processes of
the dinickel unit about the central arene ring. Compound 7
from the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
deposit@ccdc.cam.ac.uk. Supplementary data associated with this
article can be found, in the online version, at http://dx.doi.org/
10.1016/j.poly.2014.06.041.
was prepared independently from the reaction of
paraformaldehyde.
5 with
References
A XRD study of crystals grown from vapor diffusion of pentane
into a diethyl ether solution of 7 reveals the structure to be a
dinickel(0) complex in which the two nickel centers are located
on the same face of the central arene and are bridged by two
[1] R.H. Holm, P. Kennepohl, E.I. Solomon, Chem. Rev. 96 (1996) 2239.
[2] A.L. Gavrilova, B. Bosnich, Chem. Rev. 104 (2004) 349.
[3] (a) A. Velian, S. Lin, A.J.M. Miller, M.W. Day, T. Agapie, J. Am. Chem. Soc. 132
(2010) 6296;
(b) S. Lin, M.W. Day, T. Agapie, J. Am. Chem. Soc. 133 (2011) 3828;
(c) S. Lin, D.E. Herbert, A. Velian, M.W. Day, T. Agapie, J. Am. Chem. Soc. 135
(2013) 15830;
(d) K.T. Horak, A. Velian, M.W. Day, T. Agapie, Chem. Commun. 50 (2014)
4427.
l-CO ligands (Fig. 5). Notably, the cofacial arrangement of metals
supported by DIPr occurs in connection with small ligands that
show a propensity to bridge (Clꢁ or CO). The nickel centers also
interact with the vicinal diene of the central arene in a similar
manner as 4, with bond localization of the central arene C–C bonds
(Fig. 5, inset). However, the dinickel moiety is located further from
the ipso-carbons of the central arene than that of 4, with torsion
angles of 50.38–51.28° between the pendant and central aryl
groups. The structure of compound 7 also displays an unusually
short Ni–Ni distance (2.248(1) Å). In comparison, the Ni–Ni dis-
tance of compound 4 is 2.3936(3) Å and the median Ni–Ni distance
of carbonyl-bridged species from a CSD search of dinickel com-
plexes is 2.387 Å. This shorter distance compared to other car-
bonyl-bridged nickel species may be due to the constraints of the
dinucleating ligand framework. The conversion of 5 to 7 indicates
that the metal-arene interactions are hemilabile and allow for
transfacial to cofacial isomerization. The isolation of 7 demon-
[4] B.R. Dible, M.S. Sigman, A.M. Arif, Inorg. Chem. 44 (2005) 3774.
[5] (a) A.J. Arduengo, R.L. Harlow, M. Kline, J. Am. Chem. Soc. 113 (1991) 361;
(b) W.A. Herrmann, C. Köcher, Angew. Chem., Int. Ed. 36 (1997) 2162;
´
(c) S. Diez-Gonza´lez, N. Marion, S.P. Nolan, Chem. Rev. 109 (2009) 3612.;
(d) G.C. Fortman, S.P. Nolan, Chem. Soc. Rev. 40 (2011) 5151.
[6] (a) J.A. Mata, M. Poyatos, E. Peris, Coord. Chem. Rev. 251 (2007) 841;
(b) D. Pugh, A.A. Danopoulos, Coord. Chem. Rev. 251 (2007) 610;
(c) M. Poyatos, J.A. Mata, E. Peris, Chem. Rev. 109 (2009) 3677.
[7] (a) U.J. Scheele, M. John, S. Dechert, F. Meyer, Eur. J. Inorg. Chem. 2008 (2008)
373;
(b) K.-M. Lee, J.C.C. Chen, I.J.B. Lin, J. Organomet. Chem. 617–618 (2001) 364;
(c) U.J. Scheele, S. Dechert, F. Meyer, Inorg. Chim. Acta 359 (2006) 4891;
(d) U.J. Scheele, S. Dechert, F. Meyer, Tetrahedron Lett. 48 (2007) 8366;
(e) Y. Li, L. Yang, Q. Chen, C. Cao, P. Guan, G. Pang, Y. Shi, Z. Anorg. Allg. Chem.
639 (2013) 575;
(f) B. Liu, C. Chen, Y. Zhang, X. Liu, W. Chen, Organometallics 32 (2013) 5451;
(g) S.A. Reindl, A. Poethig, M. Drees, B. Bechlars, E. Herdtweck, W.A. Herrmann,
F.E. Kuehn, Organometallics 32 (2013) 4082;
(h) L. Yang, J. Zhao, Y. Li, K. Ge, Y. Zhuang, C. Cao, Y. Shi, Inorg. Chem. Commun.
22 (2012) 33;
(i) J. Zhao, L. Yang, K. Ge, Q. Chen, Y. Zhuang, C. Cao, Y. Shi, Inorg. Chem.
Commun. 20 (2012) 326;
strates that CO2 cleavage can occur at
a dinickel complex
supported by a binucleating NHC-based ligand to generate metal-
CO species. Addressing the fate of the oxygen atom in an atom
economical manner remains a challenge.
(j) Y. Zhou, Z. Xi, W. Chen, D. Wang, Organometallics 27 (2008) 5911.
[8] R.A. Andersen, K. Faegri, J.C. Green, A. Haaland, M.F. Lappert, W.P. Leung, K.
Rypdal, Inorg. Chem. 27 (1988) 1782.
4. Conclusions
[9] J.-F.o. Lemonnier, L. Gue´ne´e, G.r. Bernardinelli, J.-F.o. Vigier, B. Bocquet, C.
Piguet, Inorg. Chem. 49 (2010)1252.
[10] T. Irrgang, S. Keller, H. Maisel, W. Kretschmer, R. Kempe, Eur. J. Inorg. Chem.
2007 (2007) 4221.
[11] C. Burstein, C.W. Lehmann, F. Glorius, Tetrahedron 61 (2005) 6207.
[12] (a) M. Alcarazo, S.J. Roseblade, A.R. Cowley, R. Fernández, J.M. Brown, J.M.
Lassaletta, J. Am. Chem. Soc. 127 (2005) 3290;
A novel bis(N-heterocyclic carbene) framework containing a
central arene donor was prepared and employed to support dinu-
clear complexes of first row transition metals. Dicopper(I), dico-
balt(II), and dinickel complexes of varying oxidation state were
synthesized and structurally characterized. While the copper and
cobalt complexes display metals on opposite sides of the central
arene, the dinickel(I) complex is cofacial with a strong Ni-Ni inter-
action. This compound was utilized as a precursor to dinickel(0)
species that reacted with CO2 to form a dicarbonyl-bridged dinick-
el(0) product. Although further elaboration of this chemistry will
require a suitable acceptor for the oxygen atom, the current reac-
tivity demonstrates that dinickel complexes supported by a binu-
cleating NHC ligand can promote facile CO2 reduction.
(b) A. Fürstner, M. Alcarazo, H. Krause, C.W. Lehmann, J. Am. Chem. Soc. 129
(2007) 12676;
(c) M. Alcarazo, T. Stork, A. Anoop, W. Thiel, A. Fürstner, Angew. Chem., Int. Ed.
49 (2010) 2542;
(d) B.A. Ondrusek, S.M. Opalka, O. Hietsoi, M. Shatruk, D.T. McQuade, Synlett
24 (2013) 1211.
[13] (a) N.P. Mankad, T.G. Gray, D.S. Laitar, J.P. Sadighi, Organometallics 23 (2004)
1191;
(b) H. Kaur, F.K. Zinn, E.D. Stevens, S.P. Nolan, Organometallics 23 (2004)
1157.
[14] (a) A.A. Danopoulos, P. Braunstein, N. Stylianides, M. Wesolek,
Organometallics 30 (2011) 6514;
(b) A.A. Danopoulos, P. Braunstein, Dalton Trans. 42 (2013) 7276.
[15] C.H. Lee, D.S. Laitar, P. Mueller, J.P. Sadighi, J. Am. Chem. Soc. 129 (2007) 13802.
[16] S. Suseno, K.T. Horak, M.W. Day, T. Agapie, Organometallics 32 (2013) 6883.
[17] J.-H. Jeoung, H. Dobbek, Science 318 (2007) 1461.
[18] M. Aresta, C.F. Nobile, V.G. Albano, E. Forni, M. Manassero, J. Chem. Soc., Chem.
Commun. (1975) 636.
[19] (a) C. Bianchini, C. Mealli, A. Meli, M. Sabat, Inorg. Chem. 23 (1984) 2731;
(b) J.S. Anderson, V.M. Iluc, G.L. Hillhouse, Inorg. Chem. 49 (2010) 10203;
(c) L. Gonzalez-Sebastian, M. Flores-Alamo, J.J. Garcia, Dalton Trans. 40 (2011)
9116;
(d) R. Beck, M. Shoshani, J. Krasinkiewicz, J.A. Hatnean, S.A. Johnson, Dalton
Trans. 42 (2013) 1461.
[20] B.C. Fullmer, H. Fan, M. Pink, K.G. Caulton, Inorg. Chem. 47 (2008) 1865.
[21] B. Horn, C. Limberg, C. Herwig, B. Braun, Chem. Commun. 49 (2013) 10923.
[22] (a) D.L. DeLaet, R. Del Rosario, P.E. Fanwick, C.P. Kubiak, J. Am. Chem. Soc. 109
(1987) 754;
Acknowledgements
This work was supported by the California Institute of
Technology, NSF CAREER Grant CHE-1151918 (to T.A.), and a
Sandia Campus Executive Fellowship (to E.Y.T.). TA is a Sloan, Cott-
rell, and Dreyfus fellow. We thank L. M. Henling for assistance with
crystallography. The Bruker KAPPA APEXII X-ray diffractometer
was purchased via an NSF Chemistry Research Instrumentation
award to Caltech (CHE-0639094). The 400 MHz NMR spectrometer
was purchased via NIH Award RR027690.
(b) K.S. Ratliff, R.E. Lentz, C.P. Kubiak, Organometallics 11 (1992) 1986;
(c) E. Simón-Manso, C.P. Kubiak, Organometallics 24 (2004) 96.
[23] P. Jutzi, T. Redeker, B. Neumann, H.-G. Stammler, J. Organomet. Chem. 498
(1995) 127.
[24] B. Horn, C. Limberg, C. Herwig, M. Feist, S. Mebs, Chem. Commun. 48 (2012)
8243.
Appendix A. Supplementary data
CCDC 999595–999599 contains the supplementary crystallo-
graphic data for 2, 3, 4, 5, and 7. These data can be obtained free
of charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or