Synthesis, characterization, and electrocatalytic activity of bis(pyridylimino)isoindoline Cu(ii) and Ni(ii) complexes

Article abstract of DOI:10.1039/d0dt03030a

Two NNN pincer complexes of Cu(ii) and Ni(ii) with BPIMe- [BPIMe- = 1,3-bis((6-methylpyridin-2-yl)imino)isoindolin-2-ide] have been prepared and characterized structurally, spectroscopically, and electrochemically. The single crystal structures of the two complexes confirmed their distorted trigonal bipyramidal geometry attained by three equatorial N-Atoms from the ligand and two axially positioned water molecules to give [Cu(BPIMe)(H2O)2]ClO4 and [Ni(BPIMe)(H2O)2]ClO4. Electrochemical studies of Cu(ii) and Ni(ii) complexes have been performed in acetonitrile to identify metal-based and ligand-based redox activity. When subjected to a saturated CO2 atmosphere, both complexes displayed catalytic activity for the reduction of CO2 with the Cu(ii) complex displaying higher activity than the Ni(ii) analogue. However, both complexes were shown to decompose into catalytically active heterogeneous materials on the electrode surface over extended reductive electrolysis periods. Surface analysis of these materials using energy dispersive spectroscopy as well as their physical appearance suggests the reductive deposition of copper and nickel metal on the electrode surface. Electrocatalysis and decomposition are proposed to be triggered by ligand reduction, where complex stability is believed to be tied to fluxional ligand coordination in the reduced state.

Full text of DOI:10.1039/d0dt03030a

Dalton
Transactions
View Article Online
View Journal
PAPER
Synthesis, characterization, and electrocatalytic
activity of bis(pyridylimino)isoindoline Cu(II) and
Ni(II) complexes†
Cite this: DOI: 10.1039/d0dt03030a
a
b
a
Soumen Saha, Sha Tamanna Sahil, Md. Motiur R. Mazumder,
a
a
a
b
Alexander M. Stephens, Bryan Cronin, Evert C. Duin, Jonah W. Jurss
Byron H. Farnum
and
a
*
−
−
Two NNN pincer complexes of Cu(II) and Ni(II) with BPIMe [BPIMe = 1,3-bis((6-methylpyridin-2-yl)imino)
isoindolin-2-ide] have been prepared and characterized structurally, spectroscopically, and electrochemi-
cally. The single crystal structures of the two complexes confirmed their distorted trigonal bipyramidal
geometry attained by three equatorial N-atoms from the ligand and two axially positioned water mole-
cules to give [Cu(BPIMe)(H
complexes have been performed in acetonitrile to identify metal-based and ligand-based redox activity.
When subjected to a saturated CO atmosphere, both complexes displayed catalytic activity for the
reduction of CO with the Cu(II) complex displaying higher activity than the Ni(II) analogue. However, both
2 2 4 2 2 4
O) ]ClO and [Ni(BPIMe)(H O) ]ClO . Electrochemical studies of Cu(II) and Ni(II)
2
2
complexes were shown to decompose into catalytically active heterogeneous materials on the electrode
surface over extended reductive electrolysis periods. Surface analysis of these materials using energy dis-
persive spectroscopy as well as their physical appearance suggests the reductive deposition of copper
and nickel metal on the electrode surface. Electrocatalysis and decomposition are proposed to be trig-
gered by ligand reduction, where complex stability is believed to be tied to fluxional ligand coordination
in the reduced state.
Received 30th August 2020,
Accepted 7th December 2020
DOI: 10.1039/d0dt03030a
rsc.li/dalton
base and may coordinate to the metal center to form either
Introduction
2
2
2
: 1 (ligand : metal) or 1 : 1 complexes.
In recent years utilization of these transition metal BPI
−
Meridionally coordinating tridentate ligands, so-called ‘pincer’
ligands, have been extensively studied in inorganic, organo- complexes toward energy storage and catalysis have become
1
–5
23–30
metallic, and molecular catalysis.
Among this class of active research areas.
and PNP,
For example, several first-row metal
6
–8
9–11
12–15
16–19
−
ligands (e.g. NCN,
CNC,
NNN,
where BPI complexes have been shown to act as potential anolyte
2
3
letters represent the atomic nature of each coordinating atom), materials for nonaqueous redox flow batteries. These pseudo
bis(pyridylimino) isoindoles (BPIH, NNN) are of interest octahedral 2 : 1 complexes displayed multiple, reversible
because of their ease of synthesis, thermal stability, and tun- ligand-based redox potentials. The reversibility of the redox
ability by modification of the aromatic rings. A broad range of events and the stability of the complexes over multiple charge–
derivatized BPIH ligands have been reported as metal chela- discharge cycles was attributed to the strong chelating environ-
tors and their coordination chemistry has been explored since ment and coordination saturation of the metal center. In
2
0–22
−
the late twentieth century.
These ligands act as anionic terms of catalysis, 1 : 1 complexation between BPI and the
−
chelating pincer ligands (i.e. BPI ) in the presence of weak metal is preferred to have open coordination sites at the metal
2
5,30–32
2
3
center.
As an example, a series of κ -(N,N) and κ -(N,N,
N) coordinated iridium and osmium complexes were reported
a
to be catalytically active in the transfer hydrogenation of
Department of Chemistry and Biochemistry, Auburn University, Auburn, AL 36832,
ketones and imines. A variety of BPI− derivatives have been
shown to produce 1 : 1 complexes, however, one strategy which
encourages their formation is to install sterically bulky substi-
tuents in the ortho position relative to the pyridine nitrogen.
Derivatization at this position discourages coordination in the
equatorial plane and promotes axial coordination above and
31
USA. E-mail: farnum@auburn.edu
b
Department of Chemistry and Biochemistry, University of Mississippi, University,
MS 38677, USA
†
Electronic supplementary information (ESI) available: X-ray crystal structures,
crystallographic data, CV, SEM/EDS, ESI-MS, ATR-IR and NMR. CCDC 2006230,
012499, 1904565 and 1913288. For ESI and crystallographic data in CIF or other
2
electronic format see DOI: 10.1039/d0dt03030a
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
below the BPI− ligand. This strategy has been used in particu- (Incoatec IμS DIAMOND microfocus sealed tube,
λ =
−
13
lar toward asymmetric catalysis using chiral BPI ligands.
1.54178 Å). The integrations and global cell refinements were
In this context we report the synthesis and catalytic pro- performed by using APEX3 software which includes Bruker
perties of Cu(II) and Ni(II) complexes containing a BPI− ligand SAINT software package. Finally, the structures were solved by
3
4,35
derivatized with a methyl group at the ortho position relative to using Intrinsic Phasing/Direct Methods (ShelXT)
each pyridine nitrogen, BPIMeH. This functionalization is squares refinement was performed using ShelXL in APEX3.
shown to discourage the formation of 2 : 1 complexes and Olex 2 solvent mask was employed for [Cu (BPIMe (OH)]ClO
forms the corresponding [Cu(BPIMe)(H O) ]ClO ([Cu-BPI ] ) Mask void volume: 187 Å volume corresponding to 68 elec-
and least-
2
)
2
4
.
+
3
2
2
4
Me
+
and [Ni(BPIMe)(H
2 2 4
O) ]ClO ([Ni-BPIMe] ) species. To our knowl- trons, closely matching three methanols and one water per
edge, these complexes have not been previously reported in unit cell.
the literature. Characterization of their physical properties was
Scanning electron microscopy (SEM) was performed using a
performed using single crystal X-ray diffraction, UV-visible Zeiss EVO 50VP scanning electron microscope with energy dis-
absorption spectroscopy, electron paramagnetic spectroscopy, persive spectroscopy (EDS, INCA). SEM/EDS was performed on
and electrochemistry. The electrocatalytic reduction of CO by GC working electrodes following controlled potential electroly-
2
+
each complex was also assessed, where [Cu-BPIMe] exhibited sis of each metal complex and a gentle rinse. Elemental peaks
+
more electrocatalytic behavior compared to [Ni-BPIMe
]
with were calibrated using INCA software and a copper standard.
Caution! Metal complexes containing perchlorate ions are
each producing CO as product. Both complexes, however, were
found to decompose to heterogeneous materials over extended potentially explosive and should be used carefully.
electrolysis periods. These decomposition products were also
Synthesis of 1,3-bis((6-methylpyridin-2-yl)imino)isoindolin
found to be catalytically active for CO reduction to CO.
2
(
BPIMeH)
The ligand BPIMeH was synthesized by following a previously
reported procedure with slight modifications to the extraction
Experimental section
General considerations
1
5,36,37
step.
Phthalonitrile (Alfa Aesar, 98%, 2.56 g, 20.0 mmol)
was added to a 3-necked round bottom flask along with CaCl
2
All chemicals were used as received with the exception of (Sigma Aldrich, ≥ 93%, 1.10 g, 9.20 mmol) and 2-methyl-6-
n-tetrabutylammonium hexafluorophosphate (TBAPF ), which aminopyridine (Sigma Aldrich 98%, 4.54 g, 42.0 mmol). The
was recrystallized twice prior to electrochemical experiments. mixture was purged for 20 min with N followed by addition of
H and C NMR were recorded on a Bruker 600 MHz spectro- butanol (Sigma Aldrich, 99.8%, 30 mL) and an additional
meter using methanol-d (Cambridge Isotope Laboratories, purge of 1 h with N . The reaction mixture was then refluxed
9.8%). Chemical shifts are reported in parts per million (δ) for 48 h at 117 °C. Once the mixture cooled, a yellow-green pre-
6
2
1
13
4
2
9
and referenced against residual internal solvent signals. Mass cipitate was formed that was washed with water and filtered
spectrometry analyses were performed on a quadrupole time- three times. After filtration, the green solid mixture was dis-
of-flight mass spectrometer (Q-Tof Premier, Waters) with elec- solved in dichloromethane (DCM, VWR, 99.5%) and subjected
trospray ionization (ESI) in positive mode using Masslynx soft- to extraction with water (1 : 1, 3 times) to remove any excess
ware (V4.1). The samples were infused in acetonitrile via a salt. The organic layer was evaporated to dryness and the final
syringe pump. Elemental analysis was performed by Atlantic product was recrystallized in methanol (MeOH, VWR, 98%) to
Microlab with microcrystalline powder samples. UV-visible result in a yellow microcrystalline powder. X-ray quality crystals
absorbance spectra were recorded on an Agilent HP 8454 were grown by dissolving the solid in DCM followed by slow
diode array spectrophotometer in acetonitrile (MeCN) for all evaporation at room temperature. Yield: 4.40 g, 67%. ATR-IR
−
1
samples. Attenuated total reflectance Fourier transform infra- (cm ): 3277, 1633, 1556, 1438, 1203, 1146, 774, 677. UV-Vis
−
1
−1
−1
−1
red spectroscopy (ATR-IR) was performed using a Nicolet iS-50 (MeCN): 275 nm (ε/M cm , 13 900), 367 nm (ε/M cm
,
1
spectrometer with a built-in diamond ATR. Characterization 12 400). H-NMR: 7.99–7.97 (q, J = 6, 2H), 7.67–7.64 (m, J = 12,
data for NMR, ATR-IR, and ESI-MS are given in the ESI.† 4H), 7.09–7.07 (d, J = 12, 2H), 7.00–6.98 (d, J = 12, 2H), 2.35 (s,
1
3
Electron paramagnetic resonance spectroscopy was per- 6H) ppm. C-NMR: 161.08, 158.71, 154.79, 140.55, 136.41,
formed using an X-band Bruker EMX spectrometer fitted with 133.66, 123.90, 121.48, 119.18, 24.69 ppm. ESI mass (m/z) for
+
an ER-4119-HS (high-sensitivity) perpendicular-mode cavity. [C H N + H] : calcd: 328.1562, found: 328.1555. Elemental
2
0
17 5
17 5
Samples were prepared by dissolving crystalline solid in MeCN analysis (%) for C20H N : calcd: C, 73.37; H, 5.23; N, 21.39;
and transferring to an EPR tube where the sample was frozen found: C, 73.08; H, 5.38; N, 21.25.
using liquid nitrogen. The spectra were recorded at 77 K with a
field modulation frequency of 100 kHz, a modulation amplitude
Synthesis of [Cu(BPIMe)(H O) ]ClO
4
2
2
of 6.00 G, a microwave power of 1.995 mW, and a frequency of BPIMeH (0.327 g, 1.00 mmol) was added to 30 mL of MeOH
.366 GHz. EPR data were simulated using the program with methanolic tetra-n-butylammonium hydroxide (TBAOH,
9
33
Hyperfine Spectrum (W R Hagen Visual Software, V 1.0, 2009).
Single crystal X-ray diffraction was carried out on a Bruker tion of Cu(ClO
D8 VENTURE κ-geometry diffractometer using Cu Kα radiation 0.966 mmol) in 10 mL MeOH was warmed at 50 °C and then
Alfa Aesar, 1.0 mL, 1.0 M) and warmed at 50 °C for 1 h. A solu-
4
)
2
·6H (Sigma Aldrich, 98% 0.370 g,
2
O
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
added to the stirring mixture. After 1 h, the solution was fil-
C
20
H
20
N
5
NiClO
6
: calcd: C, 46.15; H, 3.87; N, 13.45; found: C,
tered to remove unreacted solid. The filtrate was collected and 45.96; H, 3.94; N, 13.42.
evaporated to dryness. The resulting solid was washed with
Electrochemistry
5
0 mL DCM and dried in air. X-ray quality crystals were grown
by dissolving the solid in MeOH followed by slow evaporation TBAPF (Sigma Aldrich, 98%) was recrystallized twice in absol-
6
at room temperature. Yield 0.40 g, 79%. ATR-IR (cm− ): 3402, ute ethanol (KOPTEC, 200 proof), washed with diethyl ether
1
1
583, 1518, 1446, 1192, 1041, 782, 708, 619, 511. UV-Vis (Avantor, 99%), dried under vacuum, and stored in a desicca-
−
1
−1
−1
−1
(MeCN) 277 nm (ε/M cm , 22 500), 322 nm (ε/M cm
,
tor. All cyclic voltammetry (CV) experiments were performed in
5 800), 338 nm (ε/M cm , 23 100), 431 nm (ε/M cm−1, MeCN with 0.1 M TBAPF
electrolyte purged with either N or
(Bone dry, Airgas). Experiments were performed at room
−
1
−1
−1
2
2
6
2
−
1
cm−1, 21 400) ESI mass (m/z) for CO
5 300), 459 nm (ε/M
2
+
[C H N Cu] : calcd: 389.0701, found: 389.0733. Elemental temperature using a CH Instruments 600E Series potentiostat
20 16 5
analysis (%) for C20
20 5 6
H N CuClO : calcd: C, 45.72; H, 3.84; N, with a glassy carbon (GC) disk working electrode (3 mm dia-
+
1
3.33; found: C, 45.59; H, 3.87; N, 13.23.
meter, BASi), a Ag/Ag nonaqueous reference electrode (BASi)
with 0.001 M AgNO3 (BASi) in MeCN, and a platinum wire
counter electrode (BASi). The GC working electrode was
Synthesis of [Cu (BPI ) (OH)]ClO4
2
Me 2
BPIMeH ligand (0.654 g, 2.00 mmol) along with triethylamine polished with 0.05 μm alumina powder (Allied High Tech
TEA, Sigma Aldrich, >99%, 418 µL, 3.00 mmol) was added to Products Inc., DeAgglomerated). The reduction potential of
a 30 mL 1 : 1 mixture of MeCN (VWR, 99.95%)/DCM. A solu- ferrocene (Fc; Alfa Aesar, 99%) was recorded before and after
tion of Cu(ClO ·6H O (0.740 g, 1.93 mmol) in 10 mL of all electrochemical experiments in a separate MeCN (0.1 M
MeCN was then added to the stirring ligand solution. After TBAPF ) solution to confirm the consistency of the reference
(
4
)
2
2
6
+
/0
1
h, the solution was filtered, the green solid was collected, electrode. All potentials are reported versus the Fc couple. All
washed with 50 mL of dichloromethane, and recrystallized in data were collected after compensating for the internal solu-
MeOH. X-ray quality crystals were obtained by vapor diffusion tion resistance and consisted of three continuous cycles
of diethyl ether over MeOH solution at room temperature. scanned with an initial negative direction over the potential
−
1
Yield 0.60 g, 62%. ATR-IR (cm ): 3402, 1583, 1523, 1438, range. Data shown are those of the third cycle.
−
1
1
188, 1086, 804, 706, 617, 513. UV-Vis (MeCN): 277 nm (ε/M
Controlled potential electrolysis (CPE) experiments were
performed in a 3-neck pear-shaped glass cell with a GC rod
−
1
−1
−1
−1
−1
cm , 22 800), 316 nm (ε/M cm , 26 400), 340 (ε/M cm
2
1
,
4 400), 430 nm (ε/M cm , 21 800), 460 nm (ε/M cm−1, (2 mm diameter, type 2, Alfa Aesar) working electrode, a silver
6 000) ESI mass (m/z) for [C H N OCu ] : calcd: 795.1523. wire quasi-reference electrode, and a platinum mesh (2.5 cm
−
1
−1
−1
+
2
4
0
33
5
2
Found: 389.0708 (50%), 390.0777 (100%). The observed area, 150 mesh) counter electrode that was housed in an iso-
masses indicate the dissociation of the complex into two units lation chamber, separated from the other electrodes with a
+
of [Cu(BPI )] . The mass of 389.0708 is consistent with fine glass frit. Solutions were degassed with CO for 30 min
Me
+
2
II
[
[
Cu (BPIMe)] while the mass of 390.0777 is consistent with before collecting data. Applied potentials in CPE experiments
Cu (BPIMeH)] . We believe the reduction of the copper center were determined by cyclic voltammetry. Constant stirring was
I
+
and protonation occur during the ESI-MS experiment as such maintained during electrolysis. Headspace gases in the airtight
3
8,39
events have been previously observed in the literature.
ESI† vessel were analyzed by gas chromatography using a custom
shows simulations of calculated spectra based on a mixture of PerkinElmer Clarus 680 Gas Chromatograph (Agilent
ions. Elemental Analysis for C H N Cu ClO : calcd: C, 53.60; PorapakQ column, 6 ft, 1/8 in. OD) with a dual detector system
40
33 10
2
5
H, 3.71; N, 15.63 (%); found: C, 53.17; H, 3.50; N, 15.53 (%).
(TCD and FID). Integrated gas peaks were quantified with cali-
bration curves generated from known standards purchased
from BuyCalGas.com. CO was measured at an FID detector
Synthesis of [Ni(BPIMe)(H O) ]ClO
2
2
4
BPIMeH ligand (0.327 g, 1.00 mmol) along with triethylamine equipped with a methanizer, while H
TEA, Sigma Aldrich, ≥99%, 209 µL, 1.50 mmol) was added to TCD detector. faradaic efficiencies were determined from the
a 30 mL MeCN/DCM (1 : 1) mixture. A solution of Ni experimental amount of product formed during electrolysis
ClO ·6H O (STREM, 99%, 0.365 g, 0.975 mmol) in 10 mL of divided by the theoretical amount of product possible based
2
was quantified at the
(
(
4
)
2
2
MeCN was then added to the stirring ligand mixture. After 1 h, on accumulated charge passed and the electron stoichiometry
the solution was filtered and the green solid was further of the reaction × 100.
washed by 50 mL of DCM and recrystallized in MeOH. X-ray
quality crystals were obtained by vapor diffusion of diethyl analysis were performed with
ether over a methanol solution at room temperature. Yield Bipotentiostat. A rotating disk electrode (RDE) with inter-
CPE experiments on GC disk electrodes used for SEM/EDS
a
Pine Wavedriver 20
−
1
0
1
.41 g, 81%. ATR-IR (cm ): 3384, 1579, 1509, 1448, 1189, changeable GC disks (5 mm diameter, Pine Research) was
−
1
+
034, 899, 778, 710, 620, 518. UV-Vis (MeCN): 289 nm (ε/M
used as the working electrode with a Ag/Ag nonaqueous refer-
ence electrode (BASi) with 0.001 M AgNO3 (BASi) in MeCN,
and a platinum mesh counter electrode. All experiments were
cm , 15 300), 321 nm (ε/M−1 cm , 23 800), 337 nm (ε/M
−
1
−1
−1
cm , 20 600), 431 nm (ε/M−1 cm , 30 000), 460 nm (ε/M
−
1
−1
−1
−
1
+
cm , 28 800). ESI mass (m/z) for [C20
16 5 2
H N Ni] : calcd: performed on freshly polished GC disk electrodes in CO -satu-
3
84.0759, found: 384.0841. Elemental analysis (%) for rated MeCN with 0.1 M TBAPF6 electrolyte at the indicated
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
applied potentials under constant stirring. Following CPE
experiments, GC disk electrodes were removed from the RDE [Cu(BPIMe)(H O) ]ClO ([Cu-BPI ] )and [Ni(BPI )(H O) ]ClO
and gently rinsed with acetone before analysis with SEM/EDS.
Solid state crystal structures of the mononuclear complexes
+
2
2
4
Me
Me
2
2
4
+
([Ni-BPIMe] ) are shown in Fig. 2. Their structures are highly
similar, with coordination by three N-atoms from the ligand
and two water molecules positioned axially to attain a dis-
torted trigonal bipyramidal geometry. The average Cu–Npy
Results and discussion
(
2.028 Å) and Ni–Npy (2.077 Å) distances were found within the
4
1–43
The ligand BPIMeH (Fig. 1) containing methyl groups in the range of similar reported complexes.
The shorter dis-
ortho position with respect to the pyridine nitrogen was syn- tances for Cu–Nind (1.937 Å), Ni–Nind (1.933 Å), and the back-
1
5,36,37
thesized by following an earlier literature procedure.
bone CvN (1.297 Å/1.303 Å; Cu/Ni) distances suggests the
Single crystal X-ray crystallography revealed that BPIMeH crys- coordination of anionic BPIMe− to the respective metal centers.
tallizes with a dimeric asymmetric unit with one pyridine ring In the free BPIMeH ligand, one of the pyridine rings is twisted
puckered above the isoindole-pyridine plane (Fig. S3†) This is from the plane (torsion angle; C7–N2–C6–N1 ∼31.34°) because
believed to be the result of π–π interactions between neighbor- of the presence of the ortho methyl groups in the π–π stacking
ing molecules and the steric repulsion of ortho-methyl groups. dimeric unit. Upon complexation to both Cu(II) and Ni(II), the
The crystallographic parameters are closely matched with an two pyridine rings become coplanar with torsion angles of 6.5°
4
0
+
+
earlier report of this ligand structure.
for [Cu-BPIMe] and 1.1° for [Ni-BPI ] . Additionally, the posi-
Me
Coordination of Cu(II) and Ni(II) were achieved with a tion of the metal center in the coordination pocket is almost
mixture of 1 eq. of BPIMeH and 1 eq. Cu(ClO ) ·6H O or Ni perfectly T-shaped with average angles for N –Cu–N of 91.78°
4 2
2
py
ind
(ClO
4
)
2
·6H
2
O in MeOH at room temperature in the presence of and N –Ni–N
of 91.24°. The bond angles for N –Cu–N
py py
py
ind
base. In the case of Cu(II), the base used for deprotonation of (176.43°) and N –Ni–N (176.81°) are nearly linear. However,
py
py
BPIMeH resulted in either a mononuclear (TBAOH) or dinuclear the two axially positioned water molecules differ between the
TEA) Cu(II) 1 : 1 ligand : metal complex. Similar dinuclear coordi- two complexes by 14°; O–Cu–O (141.23°) and O–Ni–O (155.23°).
(
−
+
nation between BPI ligands and Cu(II) has also been observed in Interestingly, the presence of the o-methyl groups in [Cu-BPIMe
the literature.
]
41,42
In the case of Ni(II), the 1 : 1 mononuclear decreases the Nind–Cu–O (108.06° and 110.68°) bond angle com-
complex was produced with both TBAOH and TEA.
pared to an analogous Cu(II) complex (Nind–Cu–O: 120.61° and
36.19°) having p-methyl substituents. This comparison high-
1
lights the steric repulsion of the o-methyl groups, resulting in
coordinated water molecules which are distorted from the trigo-
nal bipyramidal geometry to angles greater than 120°. Further
crystallographic parameters are listed in the ESI.†
In contrast to the mononuclear complexes, the dinuclear
2 2 4
[Cu (BPIMe) (OH)]ClO complex (Fig. S4†) produced Cu(II)
centers with distorted tetrahedral geometries with coordi-
nation from three N-atoms of the ligand and one µ-OH group
bridged between the Cu(II) centers. The average Cu–O bond
distance (1.928 Å) is similar with another reported µ-OH
4
1
Fig. 1 Chemical structure of BPIMeH ligand.
bridged Cu(II) complex.
Fig. 2 ORTEP diagrams of (a) [Cu(BPIMe)(H
2
O)
2
]ClO
4
and (b) [Ni(BPIMe)(H
2
O)
2
]ClO
4
with thermal ellipsoids shown at 50% probability. Solvent mole-
cules outside the coordination sphere are omitted for clarity.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
−
In all structures reported here, the BPIMe ligand design is
shown to produce 1 : 1 ligand : metal complexes. We believe
the o-methyl groups discourage the coordination of a fourth
ligand in the equatorial plane around Cu(II) and Ni(II).
However, we note that axial coordination is not predicated on
ortho substitution, as Cu(II) and Ni(II) complexes with both
axial and equatorial coordination have been reported with
4
4–46
unsubstituted pyridine arms.
Fig. 3 shows the UV-visible extinction coefficient spectra of
+
+
[
Cu-BPIMe] and [Ni-BPIMe] in MeCN solution. Both complexes
display similar spectra with a broad range of features between
00–475 nm and are overall similar to those observed for other
2
−
47
transition metal and main group BPI complexes. These
absorption features have been assigned to π–π* transitions
−
within the conjugated BPIMe ligand, as the vibronic pro-
gression of peaks can be clearly observed. The frequency differ-
ences of these vibrational sublevels are nearly constant and
close to 1440 cm− (Table S9†), which is consistent with a C–H
bending mode and similar to what is observed in the elec-
tronic absorbance spectrum of anthracene. Notably, these
transitions are red shifted with respect to those of the free
ligand. This observation has been previously assigned to an
increase in planarity of the ligand structure, resulting in both
1
+
Fig. 4 EPR spectrum of 1 mM [Cu-BPIMe
used for simulation: g
.000 (W = 50 G).
]
in MeCN at 77 K. Parameters
4
8
x
y z
= 2.213 (W = 40 G), g = 2.208 (W = 50 G), g =
2
the increase in molar absorptivity and red shift in absence of any axial ligands; however, this could be an artifact
4
7,49,50
energy.
mixed ligand-to-metal charge transfer character has been
In some cases, such as Pt(II)-BPI complexes, of the desolvation process in the ESI-MS experiment (Fig. S7†).
+
Fig. 4 shows the X-band EPR spectrum of [Cu-BPIMe] col-
Given the similarity lected at 77 K in MeCN. The signal was best simulated with
4
9,50
observed in the range of 450–600 nm.
of the spectra for [Cu-BPIMe] and [Ni-BPIMe] , we believe these
transitions to be mostly π–π* in nature. As an added note, dis- 50 G). The similarity of g
solution of [Cu
+
+
g
x
= 2.213 (W = 40 G), g
y
= 2.208 (W = 50 G), and g
and g (g⊥,avg = 2.211) as well as g
z
z
= 2.000 (W =
x
y
(BPIMe
)
2
(OH)]ClO
4
in MeCN resulted in an (g||) ∼ g
e
(=2.0023) gives a clear indication of a trigonal bipyra-
2
+
1
2
9
identical UV-visible absorbance spectrum as for [Cu-BPIMe] . midal geometry around the S =
d metal center, consistent
This would indicate dissociation of the dimer in solution to with the X-ray crystal structure. The difference between square
+
−
produce two equivalents of [Cu-BPIMe
]
and OH . Given the planar and trigonal bipyramidal geometries in copper com-
π–π* character of the absorbance spectrum, it is difficult to plexes has been correlated to the g
⊥
and g|| values, where g
< g|| indicates square
It is unclear whether the water molecules found in
⊥
>
−
ascertain coordination of OH , H O, or MeCN to the metal
g|| indicates trigonal bipyramidal and g
⊥
2
5
1–53
center in solution. Mass spectrometry data indicated the planar.
the X-ray crystal structure are responsible for axial coordi-
nation or if they have been replaced by MeCN. The linewidths
for each g-tensor are notably broad (>40 Gauss) and could indi-
cate unresolved hyperfine coupling to the I = 3/2 spin of the
+
copper metal center. [Ni-BPIMe] was found to be EPR silent in
the X-band region, consistent with an S = 1 spin-state and sup-
1
ported by H-NMR which exhibited broad, paramagnetically
shifted peaks.
Electrochemistry
+
+
Cyclic voltammograms of [Cu-BPIMe
MeCN with 0.1 M TBAPF electrolyte under an atmosphere of
are shown in Fig. 5. Table 1 shows a summary of cathodic and
anodic peak potentials (Epc and E ) as well as E ((Epa + E )/2)
]
and [Ni-BPIMe
]
in
6
N
2
pa
1/2
pc
potentials for BPIMeH and the series of complexes. For
+
[
−
Cu-BPIMe] , three quasi-reversible redox waves with Epc1
=
+
/0
0.49 V, Epc2 = −1.82 V, and Epc3 = −2.32 V vs. Fc
were
observed. Epc1 is assigned to the Cu(II/I) redox couple while
pc2 and Epc3 are attributed to ligand-based reductions as both
peaks are comparable to those observed for the free ligand
+
E
Fig. 3 UV-visible extinction coefficient spectra of BPIMeH, [Cu-BPIMe] ,
+
and [Ni-BPIMe
]
in MeCN.
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
The reversibility of the first ligand-based reduction is
+
+
notably higher for [Ni-BPIMe
BPIMeH, with peak splitting (ΔE
.33 V, respectively. We hypothesize that the reversibility, or
]
than for [Cu-BPIMe
]
and
p
) values of 0.07, 0.43, and
0
lack thereof, for this wave is related to rotation of one of the
pyridine arms out of the coordination pocket. Such fluxional
behavior in this ligand class has been noted in the literature
with examples of monometallic and bimetallic complexes
showing coordination to the rotated Npyridine in addition to a
2
4,31
backbone Nimine
.
Trimetallic complexes have even been
4
2
observed in the case of Zn(II) metal centers. For BPIMeH, the
pyridine arm is free to rotate out of the coordination pocket
upon reduction; however, when a metal center is present, this
rotation is likely inhibited. In the case of both metal com-
plexes, the metal is present in the +1 oxidation state by the
1
0
time the ligand reduction occurs. For Cu(I), the d electronic
configuration would yield no crystal-field stabilization energy
+
Fig. 5 CV data of 1 mM BPIMeH (blue), 0.5 mM [Cu-BPIMe
]
(red), and (CFSE) for coordination and we believe this results in rotation
+
9
0
.5 mM [Ni-BPIMe
TBAPF
OC) and initial scan direction.
]
(green) in MeCN under N
2
atmosphere with 0.1 M of the pyridine arm upon ligand reduction. For Ni(I), the d
. Scan rate: 100 mV s− . Arrows indicate the resting potential
1
6
configuration could provide some level of CFSE and therefore
the pyridine arm remains coordinated to the metal center on
the timescale of the CV experiment, allowing for a reversible
oxidation of the ligand.
(
V
BPIMeH. DFT calculations for the LUMO of the BPI class of
ligands indicate the electron for the first ligand-based
reduction is within a π* orbital delocalized over the entire
ligand structure. We speculate that the reduction at Epc
2.20 V could be related to the Cu(I/0) redox couple. For
Ni-BPIMe] , similar ligand reductions at Epc2 = −1.84 V and
Epc3 = −2.26 V were observed while the Ni(II/I) redox couple was
shifted cathodically to E = −1.14 V with respect to the M(II/I)
Electrocatalysis
5
0
=
In the last few years tridentate pincer complexes of first-row
transition metals have been used for electrocatalytic CO2
reduction because of their high molecular tunability and, in
−
[
+
5
,28,54–57
some cases, ligand-based redox activity.
To this end,
pc1
+
+
+
the electrocatalytic activity of [Cu-BPIMe
]
and [Ni-BPIMe]
observed for [Cu-BPIMe] .
toward CO
2
reduction were investigated in MeCN. Fig. 6 shows
+
+
a
Table 1 Summary of redox potentials for BPIMeH, [Cu-BPIMe] , and [Ni-BPIMe
]
in MeCN under N
2
E
pc1
E
pa1
E
1/2 (ΔE
p
)
E
pc2
E
pa2
p
E1/2 (ΔE )
E
pc3
E
pa3
p
E1/2 (ΔE )
BPIMe
H
−1.76
−1.82
−1.84
−1.43
−1.39
−1.77
−1.59 (0.33)
−1.60 (0.43)
−1.82 (0.07)
−2.33
−2.32
−2.26
−2.17
−2.20
−2.02
−2.25 (0.16)
−2.26 (0.12)
−2.14 (0.24)
+
[
[
Cu-BPIMe
Ni-BPIMe
]
]
−0.49
−1.14
−0.28
−1.05
−0.38 (0.21)
−1.09 (0.09)
+
a
p
ΔE = Epa − Epc.
+
+
. Scan rate: 100 mV s−1
Fig. 6 CV data of (a) 0.5 mM [Cu-BPIMe
]
and (b) 0.5 mM [Ni-BPIMe
]
under N
2
- and CO
2
-saturated MeCN with 0.1 M TBAPF
6
.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
a comparison of these two complexes under N
2
and CO
2
the major product of CO
2
reduction with only small amounts
atmospheres where some degree of electrocatalysis was of H detected. Table 2 shows a summary of FE data as a func-
2
observed in both cases. Scan rate dependent CVs for both com- tion of electrolysis time for both complexes. Clearly,
+
plexes revealed that increases in current at potentials <−2.0 V [Cu-BPIMe] is more electrocatalytic, supported by the CV data
1
/2
+
are catalytic, exhibiting higher normalized current (i/v ) at of Fig. 6. The decrease in FE for H in the case of [Cu-BPIMe]
2
+
lower scan rates (Fig. S10†). For [Cu-BPIMe] , a large catalytic is the result of continued electrolysis with no more appreciable
wave is found with Epc = −2.7 V in addition to new features increased in H . Likewise, the near constant FE for CO pro-
observed with E1/2 = −0.7 V and Epc = −1.8 V. These latter fea- duction at 11(±2)% in the case of [Ni-BPIMe
tures may indicate coordination of CO to the copper complex. steady-state production of CO during the electrolysis period. In
In the case of [Ni-BPIMe] , a small anodic shift and enhance- comparison, poor FE (<3%) for CO evolution was also observed
2
+
]
points to a
2
+
56
ment in current at the first ligand reduction was observed by a nickel bis(aldimino)pyridine pincer complex. Given the
+
along with loss of the second ligand-based reduction feature low FE for CO production for [Ni-BPIMe] , this complex was not
and an exponential rise in current at potentials <−2.0 V. It is explored further with electrocatalytic studies.
+
worth noting that the free ligand also displays some catalytic
behavior for CO
indicate that ligand reduction triggers catalysis.
The FE for CO production in the case of [Cu-BPIMe] makes
2
reduction (Fig. S11†). Altogether, these data a notable increase from 54(±2)% within the first ∼15 min to 70
(±1)% after this time point. Fig. 7a shows the charge accumu-
Controlled potential electrolysis (CPE) experiments were lation during CPE where a sudden increase in charge was
performed for both complexes to determine reduced products observed around the same time and coincided with the for-
and faradaic efficiencies (FE). Using a GC rod electrode and mation of a turbid solution. These observations suggested the
+
+
0
.5 mM [Cu-BPIMe] or [Ni-BPI ] in CO -saturated MeCN, an formation of deposited material on the electrode surface
Me 2
2
applied potential of −2.45 V or −2.50 V, respectively, was held which was also catalytic for CO reduction to CO. After finish-
for upwards of 1 h. For both complexes, CO was found to be ing the CPE experiment, the GC rod electrode appeared to
have a brass colored, shiny deposition on the surface
(
Fig. S14†). The electrode was gently rinsed with acetone and
scanned over the reductive potential window in fresh (absence
Table 2 Summary of faradaic efficiencies (FE) for 0.5 mM complex in
a
+
CO
2
-saturated MeCN
of [Cu-BPIMe] ) CO -saturated MeCN. The increase in current
2
after the rinse test revealed deposition of a catalytically active
material onto the electrode surface (Fig. S12†). Charge vs. time
data obtained from CPE (−2.45 V) of the rinsed electrode are
overlaid in Fig. 7a for comparison. The deposited material
shows an immediate increase in charge with a similar slope to
that observed after 15 min of electrolysis in the presence of
+
b
+ c
[
Cu-BPIMe
]
[Ni-BPIMe]
Time
min)
FE for CO
(%)
FE for H
(%)
2
FE for CO
(%)
FE for H
(%)
2
(
5
1
1
2
2
3
3
4
6
52
54
55
6
2
<1
0
5
0
5
0
5
5
0
13
9
0
0
0
+
[Cu-BPIMe] . Furthermore, the FE for CO with the rinsed elec-
+
trode was consistently higher (75(±7)%) than [Cu-BPIMe
]
in
71
<1
solution with no evidence for H production (Table S10†). The
9
2
71
69
<1
<1
deposited material eventually delaminated from the electrode
surface, as evidenced by the leveling off of accumulated charge
13
0
+
over time. CPE of [Ni-BPIMe
]
also found evidence for a de-
a
b
c
GC rod electrode, 0.1 M TBAPF
6
.
−2.45 V. −2.5 V.
posited material at the electrode surface (Fig. S14†) which was
+
+
Fig. 7 Accumulated charge collected as a function of time during CPE of (a) 0.5 mM [Cu-BPIMe
] (−2.45 V) and (b) 0.5 mM [Ni-BPIMe] (−2.50 V) in
CO -saturated MeCN with 0.1 M TBAPF using a GC rod electrode. Overlaid for each plot is the accumulated charge measured at the same applied
2
6
potentials for the GC rod electrode after electrolysis. Electrodes were rinsed and CPE performed in a fresh CO
absence of either complex.
2
-saturated MeCN solution in the
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
catalytic for reduction of CO
2
to CO (Fig. 7b, Table S10†). FE’s what was observed for free BPIMeH, while the reduction was
+
for CO production were found to be 83(±3)% over the first more reversible for [Ni-BPIMe] . Both complexes were observed
2
5 min and began to decrease over time as the material dela- to decompose during CPE experiments, however, the
+
minated from the electrode surface.
decomposition process appeared to be faster for [Cu-BPIMe] ,
+
Electrocatalysis with [Cu-BPIMe] was further explored in the as evidenced by the sudden increase in charge and FE for CO
+
presence of 5% H
common source for protons needed in the reduction of CO
CO as well as in the competitive formation of H . In the could indicate slow decomposition of the complex.
2
O added to the MeCN electrolyte. Water is a after 15 min. For [Ni-BPIMe] , a much lower amount of accu-
2
to mulated charge and FE for CO was observed during CPE. This
2
absence of added H
2
O, protons are believed to be derived from
Deposition of a heterogeneous catalysts is always a concern
trace water in the solvent and from the complex itself, which when studying molecular electrocatalysts. In the case of [Cu-
+
contains two coordinated water molecules in the solid state. BPIMe] , there are two relatively clear regimes for catalysis
CPE experiments were performed under these conditions and based on CPE data and measured FE’s. In the first 15 min, we
+
product yields monitored over time. At an applied potential of believe catalysis is largely homogeneous and [Cu-BPIMe
]
is
−
2.3 V, the production of CO and H were constant throughout responsible for production of CO at 54(±2)% FE with the other
2
the electrolysis period at 21(±4)% and 15(±2)% (Table S11†). 46% of current possibly contributing to decomposition.
These data indicate a competition between CO and H for- Experiments with 5% H O also clearly indicated homogeneous
2
2
mation with CO production decreasing with added H O. catalytic activity for the production of hydrogen and formate,
2
−
Furthermore, the presence of formate (HCO
2
) was detected by showing neither of these products to be present after CPE
1
H NMR at the end of the CPE experiment with an FE of 18% experiments with the rinsed electrode. After 15 min, catalysis
Fig. S13†). The presence of formate suggests that a hydride inter- shifts from homogeneous to heterogeneous as indicated by
(
58–61
mediate may be present for the condition of added H
This would also support the competitive formation of H
2
O.
the increase in charge and jump in FE for CO to 70(±1)%. This
increase in FE would be consistent with deposition of copper
2
.
6
2–68
Eletrodeposition of an unknown material was also observed metal, a well-known CO reduction catalyst.
Although CO
O and had a similar brass colored is commonly observed as a product of copper metal catalysis,
appearance as the deposited material in the absence of H it is most notable for its ability to yield hydrocarbons and C–C
Fig. S14†). CPE experiments using the rinsed electrode bonded products in addition to operation in aqueous
revealed a nearly constant FE for CO of 87(±4)% across the conditions.
electrolysis period with no detectable formation of H or Many examples of homogeneous molecular catalysts for
2
for the condition of 5% H
2
2
O
(
2
formate (Table S11†). The FE for CO is slightly higher than CO reduction to CO have been reported in the literature. The
2
that observed in the absence of water, however, we believe the most extensively studied examples include Fe porphyrins, Ni
0
/+
I
deposited material to be the same in both cases.
cyclams, and [M(bpy)(CO)
3
X]
complexes were M = Re or
I
69–73
Further experiments to understand the nature of surface Mn , bpy is 2,2′-bipyridine, and X is an anion or solvent.
deposition were conducted with GC disk electrodes so they Onset potentials for CO
2
reduction with these catalysts range
+
/0
II
+/0
could be analyzed by SEM and EDS following CPE experi- from −1.2 V vs. Fc
(Ni (cyclam)Cl
2
) to −2.2 V vs. Fc
III
70,74
0/+
ments. SEM measurements on the GC disk electrode after 1 h (Fe (tetraphenylporphyrin)Cl).
of CPE (−2.67 V) in CO
brass colored deposition on the electrode surface. SEM images depending on the functionalization of the bpy ligand.
[M(bpy)(CO) X]
com-
3
2
-saturated MeCN clearly showed a plexes fall in the middle of this range around −1.5 to −1.7 V
6
1,75–79
+
+
of the electrode surface supported this observation (Fig. S15†) The present examples of [Cu-BPIMe] and [Ni-BPIMe] display
while EDS measurements revealed the deposited material con- onset potentials of −2.1 and −2.5 V respectively. These values
tained a significant amount of copper (Table S12†). Notably, are on the high end of the range reported above and we believe
no evidence for deposition of nitrogen, which would be they could be tuned in the positive direction through
−
−
derived from the BPIMe ligand, was observed. For compari- functionalization of the BPIMe ligand with electron withdraw-
son, EDS measurements for a GC disk subjected to CPE in the ing groups. Given the mixed homogeneous/heterogeneous
+
presence of [Ni-BPIMe] showed the presence of nickel on the catalytic behavior observed here, it is difficult to determine
electrode surface in addition to carbon and oxygen. The detec- metrics for homogeneous catalysis such as turnover frequency
tion of carbon is unavoidable based on the nature of the GC and turnover number to make further comparisons with other
electrode, so we cannot say whether a carbonaceous species is catalysts in the literature.
deposited in either case, however, the measured oxygen
content for both electrodes was greater than that observed for BPIMe] is notable as molecular copper complexes are not com-
2
Finally, the CO reduction activity observed with [Cu-
+
a clean GC electrode. This means we cannot rule out the pres- monly observed to display such reactivity. In addition, the few
ence of metal oxides (e.g. Cu
2
O or NiO), but the color of the reported examples do not produce CO but instead produce
2
−
80–82
depositions strongly suggested metallic species.
oxalate (C
2
O
4
),
a
C–C bonded product.
In these
+
+
The stability of [Cu-BPIMe] and [Ni-BPI ] may be related examples, the coordination environment around the copper is
Me
to the reversibility of the ligand reduction. CV data shown in distorted tetrahedral whereas the present example is trigonal
Fig. 5 and Table 1 indicate that the first ligand-based bipyramidal. These results may suggest a structural selectivity
+
reduction was irreversible in the case of [Cu-BPIMe] , similar to for CO reduction in copper coordination complexes.
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020

View Article Online
Dalton Transactions
Paper
7
G. Mancano, M. J. Page, M. Bhadbhade and B. A. Messerle,
Inorg. Chem., 2014, 53, 10159–10170.
Conclusions
In summary, this paper describes the synthesis and character-
ization of two pincer complexes of Cu(II) and Ni(II) with the bis
8 S. Hosokawa, J. Ito and H. Nishiyama, Organometallics,
2012, 31, 8283–8290.
(
pyridylimino) isoindoline derived BPIMeH ligand. Solid state
9 A. A. Danopoulos, N. Tsoureas, J. A. Wright and
M. E. Light, Organometallics, 2004, 23, 166–168.
X-ray crystallographic studies suggest that the position of
steric methyl groups ortho to the pyridine nitrogen are impor- 10 E. Jürgens and D. Kunz, Eur. J. Inorg. Chem., 2017, 2017,
tant to preventing further coordination in the equatorial 233–236.
plane. Electrochemical studies of the free ligand, [Cu-BPIMe] , 11 M. Hernández-Juárez, M. Vaquero, E. Álvarez, V. Salazar
+
+
and [Ni-BPIMe] under N
2
and CO
2
atmospheres reveal catalytic
and A. Suárez, Dalton Trans., 2012, 42, 351–354.
behavior for the reduction of CO in all cases, where ligand- 12 K. Umehara, S. Kuwata and T. Ikariya, Inorg. Chim. Acta,
2
based reduction appears to be the initiation step for catalysis.
2014, 413, 136–142.
Both complexes were found to produce CO as the reduced 13 B. K. Langlotz, H. Wadepohl and L. H. Gade, Angew. Chem.,
+
product with [Cu-BPIMe] displaying higher catalytic activity
Int. Ed., 2008, 47, 4670–4674.
than [Ni-BPIMe] . Both complexes are also shown to decom- 14 V. C. Gibson, C. Redshaw and G. A. Solan, Chem. Rev.,
pose into heterogeneous catalysts at the electrode surface. 2007, 107, 1745–1776.
+
Based on physical appearance and EDS measurements, we 15 W. O. Siegl, J. Org. Chem., 1977, 42, 1872–1878.
believe these catalysts to be metallic copper and nickel, 16 D. Benito-Garagorri and K. Kirchner, Acc. Chem. Res., 2008,
respectively, but cannot rule out the presence of metal oxides.
41, 201–213.
The degradation of the complexes is believed to be the result 17 D. Benito-Garagorri, E. Becker, J. Wiedermann, W. Lackner,
of fluxional behavior in the ligand coordination of BPIMe−
pointing to the importance of ligand rigidity in the design
,
M. Pollak, K. Mereiter, J. Kisala and K. Kirchner,
Organometallics, 2006, 25, 1900–1913.
of first-row transition metal catalysts in low coordination 18 A. A. Kassie, P. Duan, M. B. Gray, K. Schmidt-Rohr,
environments.
P. M. Woodward and C. R. Wade, Organometallics, 2019, 38,
419–3428.
3
19 J. J. Gair, Y. Qiu, R. L. Khade, N. H. Chan, A. S. Filatov,
Conflicts of interest
Y. Zhang and J. C. Lewis, Organometallics, 2019, 38, 1407–
1
412.
0 M. A. Robinson, S. I. Trotz and T. J. Hurley, Inorg. Chem.,
967, 6, 392–394.
There are no conflicts to declare.
2
2
2
2
2
2
2
2
1
1 R. R. Gagne, R. S. Gall, G. C. Lisensky, R. E. Marsh and
L. M. Speltz, Inorg. Chem., 1979, 18, 771–781.
2 R. R. Gagne, W. A. Marritt, D. N. Marks and W. O. Siegl,
Inorg. Chem., 1981, 20, 3260–3267.
3 C. S. Sevov, S. L. Fisher, L. T. Thompson and M. S. Sanford,
J. Am. Chem. Soc., 2016, 138, 15378–15384.
4 M. L. Buil, M. A. Esteruelas, S. Izquierdo, A. I. Nicasio and
E. Oñate, Organometallics, 2020, 39, 2719–2731.
5 T. Váradi, J. S. Pap, M. Giorgi, L. Párkányi, T. Csay,
G. Speier and J. Kaizer, Inorg. Chem., 2013, 52, 1559–1569.
6 J. Kaizer, G. Baráth, G. Speier, M. Réglier and M. Giorgi,
Inorg. Chem. Commun., 2007, 10, 292–294.
Acknowledgements
SS, AMS, MRM, and BHF acknowledge support from Auburn
University as well as the National Science Foundation through
a CAREER Award CHE-1945160. STS and JWJ thank the
National Science Foundation for support through a CAREER
Award CHE-1848478. The authors thank Dr John D. Gorden,
Dr Melissa Boersma, and Alexandria Bredar for assistance with
X-ray crystallography, ESI-MS, and SEM/EDS measurements,
respectively.
7 M. Bröring and C. Kleeberg, Chem. Commun., 2008, 2777–
References
2778.
1
2
3
4
5
6
F. Bertini, M. Glatz, B. Stöger, M. Peruzzini, L. F. Veiros, 28 F.-W. Liu, J. Bi, Y. Sun, S. Luo and P. Kang, ChemSusChem,
K. Kirchner and L. Gonsalvi, ACS Catal., 2019, 9, 632–639. 2018, 11, 1656–1663.
S. Chakraborty, P. Bhattacharya, H. Dai and H. Guan, Acc. 29 D. C. Sauer, H. Wadepohl and L. H. Gade, Inorg. Chem.,
Chem. Res., 2015, 48, 1995–2003. 2012, 51, 12948–12958.
M. Albrecht and G. van Koten, Angew. Chem., Int. Ed., 2001, 30 M. Szávuly, R. Csonka, G. Speier, R. Barabás, M. Giorgi and
0, 3750–3781. J. Kaizer, J. Mol. Catal. A: Chem., 2014, 392, 120–126.
M. E. van der Boom and D. Milstein, Chem. Rev., 2003, 103, 31 A. L. Müller, T. Bleith, T. Roth, H. Wadepohl and
759–1792. L. H. Gade, Organometallics, 2015, 34, 2326–2342.
K. Talukdar, A. Issa and J. W. Jurss, Front. Chem., 2019, 7, 32 K.-N. T. Tseng, J. W. Kampf and N. K. Szymczak,
30. Organometallics, 2013, 32, 2046–2049.
J. Choi, A. H. R. MacArthur, M. Brookhart and 33 W. R. Hagen, Biomolecular EPR Spectroscopy, CRC Press,
A. S. Goldman, Chem. Rev., 2011, 111, 1761–1779. 2008.
4
1
3
This journal is © The Royal Society of Chemistry 2020
Dalton Trans.

View Article Online
Paper
Dalton Transactions
3
3
3
3
3
4 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. 59 B. M. Ceballos, C. Tsay and J. Y. Yang, Chem. Commun.,
Crystallogr., 2008, 64, 112–122. 2017, 53, 7405–7408.
5 DOC-M86-EXX229 V1 APEX3 Crystallography Software Suite 60 K. M. Waldie, A. L. Ostericher, M. H. Reineke,
User Manual. Bruker AXS Inc., Madison, Wisconsin, 2016.
6 K. Gogoi, S. Saha, B. Mondal, H. Deka, S. Ghosh and
B. Mondal, Inorg. Chem., 2017, 56, 14438–14445.
A. F. Sasayama and C. P. Kubiak, ACS Catal., 2018, 8, 1313–
1324.
61 S. S. Roy, K. Talukdar and J. W. Jurss, ChemSusChem, 2020,
13, 1–10.
7 R. Csonka, G. Speier and J. Kaizer, RSC Adv., 2015, 5,
1
8401–18419.
62 F. S. Roberts, K. P. Kuhl and A. Nilsson, Angew. Chem., Int.
Ed., 2015, 127, 5268–5271.
8 L. Gianelli, V. Amendola, L. Fabbrizzi, P. Pallavicini and
G. G. Mellerio, Rapid Commun. Mass Spectrom., 2001, 15, 63 S. Nitopi, E. Bertheussen, S. B. Scott, X. Liu,
2
347–2353.
A. K. Engstfeld, S. Horch, B. Seger, I. E. L. Stephens,
K. Chan, C. Hahn, J. K. Nørskov, T. F. Jaramillo and
I. Chorkendorff, Chem. Rev., 2019, 119, 7610–7672.
3
9 H. Lavanant, H. Virelizier and Y. Hoppilliard, J. Am. Soc.
Mass Spectrom., 1998, 9, 1217–1221.
4
0 Y.-Z. Hao, Y. Xu, X.-H. Wei and H.-B. Tong, Acta 64 A. J. Garza, A. T. Bell and M. Head-Gordon, ACS Catal.,
Crystallogr., Sect. E: Struct. Rep. Online, 2007, 63, o1549–
o1550.
1 T. Csay, B. Kripli, M. Giorgi, J. Kaizer and G. Speier, Inorg.
Chem. Commun., 2010, 13, 227–230.
2 O. P. Anderson, A. la Cour, A. Dodd, A. D. Garrett and
M. Wicholas, Inorg. Chem., 2003, 42, 122–127.
3 A. M. Herrera, R. J. Staples, S. V. Kryatov, A. Y. Nazarenko
and E. V. Rybak-Akimova, Dalton Trans., 2003, 846–856.
4 M. B. Meder and L. H. Gade, Eur. J. Inorg. Chem., 2004,
2018, 8, 1490–1499.
65 D. Kopač, B. Likozar and M. Huš, Appl. Surf. Sci., 2019, 497,
143783.
66 J. Zhao, S. Xue, J. Barber, Y. Zhou, J. Meng and X. Ke,
J. Mater. Chem. A, 2020, 8, 4700–4734.
67 C. S. Chen, A. D. Handoko, J. H. Wan, L. Ma, D. Ren and
B. S. Yeo, Catal. Sci. Technol., 2014, 5, 161–168.
68 C. W. Li and M. W. Kanan, J. Am. Chem. Soc., 2012, 134,
7231–7234.
4
4
4
4
4
4
4
2
004, 2716–2722.
69 C. Costentin, S. Drouet, M. Robert and J.-M. Savéant,
Science, 2012, 338, 90–94.
70 M. Beley, J.-P. Collin, R. Ruppert and J.-P. Sauvage, J. Chem.
Soc., Chem. Commun., 1984, 1315–1316.
71 J. Hawecker, J.-M. Lehn and R. Ziessel, J. Chem. Soc., Chem.
Commun., 1983, 536–538.
5 J. Kaizer, J. Pap, G. Speier and L. Párkányi, Z. Kristallogr. -
New Cryst. Struct., 2004, 219, 141–142.
6 B. L. Truesdell, T. B. Hamby and C. S. Sevov, J. Am. Chem.
Soc., 2020, 142, 5884–5893.
7 J. D. Dang and T. P. Bender, Inorg. Chem. Commun., 2013,
3
0, 147–151.
72 C. Riplinger, M. D. Sampson, A. M. Ritzmann, C. P. Kubiak
and E. A. Carter, J. Am. Chem. Soc., 2014, 136, 16285–16298.
4
4
8 R. N. Jones, Chem. Rev., 1947, 41, 353–371.
9 K. Hanson, L. Roskop, P. I. Djurovich, F. Zahariev, 73 E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja,
M. S. Gordon and M. E. Thompson, J. Am. Chem. Soc.,
010, 132, 16247–16255.
0 H.-M. Wen, Y.-H. Wu, Y. Fan, L.-Y. Zhang, C.-N. Chen and
Z.-N. Chen, Inorg. Chem., 2010, 49, 2210–2221.
Chem. Soc. Rev., 2008, 38, 89–99.
2
74 M. Hammouche, D. Lexa, M. Momenteau and
J. M. Saveant, J. Am. Chem. Soc., 1991, 113, 8455–8466.
75 K. Talukdar, S. Sinha Roy, E. Amatya, E. A. Sleeper, P. Le
Magueres and J. W. Jurss, Inorg. Chem., 2020, 59, 6087–
6099.
5
5
1 M. Corbett and B. F. Hoskins, J. Am. Chem. Soc., 1967, 89,
1
530–1532.
5
5
2 E. Garribba and G. Micera, J. Chem. Educ., 2006, 83, 1229.
3 B. J. Hathaway and A. A. G. Tomlinson, Coord. Chem. Rev.,
76 A. Zhanaidarova, A. L. Ostericher, C. J. Miller, S. C. Jones
and C. P. Kubiak, Organometallics, 2019, 38, 1204–1207.
77 J. Hawecker, J.-M. Lehn and R. Ziessel, J. Chem. Soc., Chem.
Commun., 1984, 328–330.
78 M. H. Rønne, D. Cho, M. R. Madsen, J. B. Jakobsen,
S. Eom, É. Escoudé, H. C. D. Hammershøj, D. U. Nielsen,
S. U. Pedersen, M.-H. Baik, T. Skrydstrup and K. Daasbjerg,
J. Am. Chem. Soc., 2020, 142, 4265–4275.
1
970, 5, 1–43.
5
4 M. Sheng, N. Jiang, S. Gustafson, B. You, D. H. Ess and
Y. Sun, Dalton Trans., 2015, 44, 16247–16250.
5 T. H. T. Myren, A. M. Lilio, C. G. Huntzinger,
J. W. Horstman, T. A. Stinson, T. B. Donadt, C. Moore,
B. Lama, H. H. Funke and O. R. Luca, Organometallics,
5
2
019, 38, 1248–1253.
79 K. T. Ngo, M. McKinnon, B. Mahanti, R. Narayanan,
D. C. Grills, M. Z. Ertem and J. Rochford, J. Am. Chem. Soc.,
2017, 139, 2604–2618.
5
5
5
6 R. Narayanan, M. McKinnon, B. R. Reed, K. T. Ngo,
S. Groysman and J. Rochford, Dalton Trans., 2016, 45,
1
5285–15289.
80 U. R. Pokharel, F. R. Fronczek and A. W. Maverick, Nat.
Commun., 2014, 5, 5883.
7 J. D. Cope, N. P. Liyanage, P. J. Kelley, J. A. Denny,
E. J. Valente, C. E. Webster, J. H. Delcamp and T. K. Hollis, 81 B. J. Cook, G. N. Di Francesco, K. A. Abboud and
Chem. Commun., 2017, 53, 9442–9445. L. J. Murray, J. Am. Chem. Soc., 2018, 140, 5696–5700.
8 A. Taheri and L. A. Berben, Chem. Commun., 2016, 52, 82 R. Angamuthu, P. Byers, M. Lutz, A. L. Spek and
768–1777. E. Bouwman, Science, 2010, 327, 313–315.
1
Dalton Trans.
This journal is © The Royal Society of Chemistry 2020