Broadly versus Barely Variable Complex Chromophores of Planar Nickel(II) from κ3- N, N′, C and κ3- N, N′, O Donor Platforms

Article abstract of DOI:10.1021/acs.organomet.1c00121

This work reports on complex chromophores of planar nickel(II) of the general formula [(κ3-N,N′,C/O)Ni-ER], which combines a κ2-N,N′-coordinate aromatic N-heterocycle with a pendant phenyl/phenol donor and the variable coligands ER (= carboxylato, phenolato, amido, and arylthiolato). The absorption properties of the planar complex chromophores vary broadly depending on the interplay of the three constituting components. Increasing the manifold of π?-orbitals at the N-heterocyclic moiety through π-extension along 1,10-phenanthroline, tetrahydrodipyridophenazine, and dipyridophenazine has only a small impact on the energies of charge transfer type transitions across the visible (vis) to near-infrared (NIR) spectral region and is due to an inversion among spatially distinct but energetically close unoccupied molecular orbitals. Substantial absorptivity across the visible spectral region rather associates with the increasing capability among carboxylato, phenolato, amido, and arylthiolato coligands to engage in Ni-O/N/S π-bonding. This adds a mechanism for electronic coupling of coligand RE- and π?-orbitals at the N-heterocycle fragment through occupied d orbitals at nickel, enabling light-induced charge transfer. Aryl-S- performs best in this regard, but {Ni-O/N/S}-to-(N-heterocycle-π?) charge transfer efficacy depends on donor-acceptor orbital alignment and appears to be kinetically hindered except for amido coligands owing to polar effects. Irrespective of the nature of coligand RE-, the choice of the pendant phenyl/phenol side arm rules overall absorptivity by setting the energy of occupied frontier molecular orbitals.

Full text of DOI:10.1021/acs.organomet.1c00121

pubs.acs.org/Organometallics
Article
Broadly versus Barely Variable Complex Chromophores of Planar
Nickel(II) from κ3‑N,N′,C and κ3‑N,N′,O Donor Platforms
Riyadh Alrefai, Gerald Hörner, Hartmut Schubert, and Andreas Berkefeld*
Cite This: Organometallics 2021, 40, 1163−1177
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: This work reports on complex chromophores of
planar nickel(II) of the general formula [(κ3-N,N′,C/O)Ni−ER],
which combines a κ2-N,N′-coordinate aromatic N-heterocycle with
a pendant phenyl/phenol donor and the variable coligands ER (=
carboxylato, phenolato, amido, and arylthiolato). The absorption
properties of the planar complex chromophores vary broadly
depending on the interplay of the three constituting components.
Increasing the manifold of π*-orbitals at the N-heterocyclic moiety
through π-extension along 1,10-phenanthroline, tetrahydrodipyr-
idophenazine, and dipyridophenazine has only a small impact on
the energies of charge transfer type transitions across the visible
(vis) to near-infrared (NIR) spectral region and is due to an
inversion among spatially distinct but energetically close unoccupied molecular orbitals. Substantial absorptivity across the visible
spectral region rather associates with the increasing capability among carboxylato, phenolato, amido, and arylthiolato coligands to
engage in Ni−O/N/S π-bonding. This adds a mechanism for electronic coupling of coligand RE− and π*-orbitals at the N-
heterocycle fragment through occupied d orbitals at nickel, enabling light-induced charge transfer. Aryl-S− performs best in this
regard, but {Ni−O/N/S}-to-(N-heterocycle-π*) charge transfer efficacy depends on donor−acceptor orbital alignment and appears
to be kinetically hindered except for amido coligands owing to polar effects. Irrespective of the nature of coligand RE−, the choice of
the pendant phenyl/phenol side arm rules overall absorptivity by setting the energy of occupied frontier molecular orbitals.
visible-to-near-infrared (vis-to-NIR) spectral region.18 Origi-
INTRODUCTION
Exposing a chromophore system to light or a strong electric
field is a common practice to make excited state energy
available to chemical or physical processes. For use such as
■
nally considered for chromophore tuning of octahedral
complexes of Fe(II)18,19 and other six-coordinate 3d metal
ions including Ni(II),20 we herein consider this concept for
organic light-emitting diodes1−3 or photoredox catalysis,4−7
tailoring the properties of chromophore systems of planar
Ni(II). Considering cyclometalated structures with strongly σ-
and π-donating aryl ligands, precedent work in the field of
planar complex chromophores of Ni(II) is limited to the κ3-
N,N,C21 and κ3-N,C,N22,23 type frameworks depicted at the top
of Scheme 1. Notably, the photophysical properties of planar
bipyridine complexes of Ni(II) containing terminal aryl ligands
have been studied in detail regarding their roles in light-
promoted cross-coupling catalysis.7,24−26
platinum group metal complexes from multidentate strong-
field ligands are state-of-the-art chromophore systems owing to
their general robustness, the efficient population of metal-to-
ligand charge transfer triplet excited states (3MLCT) with
lifetimes up to the microsecond time scale, and structural
variability through ligand design.8−11 Alternatively, the
population of excited d−d states can connect to useful
photochemical properties for specific d electron configura-
tions12−14 but generally presents a mechanism for nonradiative
energy dissipation at unrivaled rates.
Strong σ-donor ligands contribute to increasing the lifetime
of a potentially functional excited CT state by rendering
excited d−d states and their population energetically less
favorable.13,15−17 A conceptually related approach considers
adjusting the energies of ligand π-donor orbitals to become
either similar to or higher than those of metal d orbitals.
Ligand−metal p−d π-bonding contributes to decreasing the
energy gap between highest occupied and lowest unoccupied
molecular orbitals, HOMO and LUMO, of the complex
chromophore and thereby imparts absorptivity across the
Building on our recent report on the selective κ3-N,N′,C-
cyclometalation of a 2-phenyl-1,10-phenanthroline platform by
zero-valent Ni,27 we herein focus on three parameters that
govern the chromophore properties of this class of planar
Received: March 2, 2021
Published: April 14, 2021
© 2021 The Authors. Published by
American Chemical Society
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177
1163

Organometallics
pubs.acs.org/Organometallics
Article
of the preparation and molecular structures of Ni(II)
complexes studied in this work. The conversion of 1,10-
phenanthroline A into 10,11,12,13-tetrahydrodipyrido[3,2-
a:2′,3′-c]phenazine B and dipyrido[3,2-a:2′,3′-c]phenazine C
followed reported protocols,29 and Scheme S1 provides the
details. With regard to proligand preparation, we emphasize
two modifications to the reaction conditions for the addition of
the aryl side arm at the 3-position of frameworks B and C that
have led to substantially improved yields. First, the reaction of
the LiCl adducts of (tetrahydro)dipyrido[3,2-a:2′,3′-c]-
phenazines B and C, preformed in THF, with 5-tert-butyl-2-
methoxyphenyllithium improved conversions by a factor of ≥2.
Otherwise, yields of ≤30% resulted as reported for related
building blocks.29 Second, the chemoselectivity for the
oxidative rearomatization of the hydrolyzed products from
aryllithium addition by excess MnO2 correlates inversely with
substrate concentration, so working in a dilute CH2Cl2
solution is key to suppress the formation of substantial
amounts of red colored byproducts. Notably, air suffices as
oxidant in cases of backbone structures B and C.
Scheme 1
Proligands A/B/C0-OPiv undergo selective metalation in
toluene solution under ambient conditions, affording pivaloato
complexes A/B/CNi-OPiv in 80−90% yield and analytical purity
as brick red to red-brown crystalline solids. Ligand metathesis
with potassium salts of the set of R2N−, aryl-O−, and aryl-S−
donors described in Scheme 2 led to products A/B/CNi-ER that
were isolated as crystalline solids. The κ3-N,N′,O-type
complexes as depicted at the bottom of Scheme 1. First, we
focus on modulating the manifold of π*-orbitals at the κ2-
N,N′-bound N-heterocyclic backbone. Second, the effect of the
Ni−ER fragment on ground-state absorption and redox
properties is addressed. This parameter is relevant considering
that 3d metal−carbazolido fragments can support photo-
luminescence.23,28 Third, making use of phenol-derived
proligands has allowed us to compare the effects of κ3-
N,N′,C- versus κ3-N,N′,O-cyclometalation on chromophore
properties.
c o m p l e x , A − O N i - S A r T i p 2 ( S A r T i p 2
=
2 , 6 -
(2,4,6-iPr3C6H3)2C6H3S−) was obtained in the form of dark
red crystals in 82% yield following the protocol reported
previously for A−ONi-NTMS2.27
RESULTS
A general indication for C-metalation is the 13C nuclear
magnetic resonance (NMR) chemical shift of the Ni bound C
atom that falls in the 147 ≤ δC ≤ 150 ppm range. The
respective δC for phenolato complexes A−ONi-ER locates at 162
ppm. The very low solubility of compounds B/CNi-OPiv
impaired recording 13C NMR data, but transmission IR spectra
■
Preparation and Molecular Structures of Nickel
Complex Chromophores. Scheme 2 provides an overview
Scheme 2. Preparation and Molecular Structures of Nickel
Complex Chromophores
a
in KBr show a diagnostic band at υ(CO) = 1612 cm−1 for
̅
the OPiv ligand akin to ANi-OPiv, which corresponds to a
redshift of 145 cm−1 compared with the proligand phenol
ester. Dipolar couplings in 1H−1H NOESY and ROESY NMR
data attest the relative spatial orientation of RE− with respect
to the primary ligand such that a lone-pair np orbital at E
orients toward the square coordination plane at nickel rather
than the π-system of the N-heterocyclic moiety. The saturated
5- and 6-membered peripheral ring structures at the phenazine
moieties of C and B were chosen to disfavor π-stacking and
increase solubility, although this has had a counterproductive
effect on obtaining detailed solid-state structure information
from single-crystal X-ray diffraction (XRD) analysis that is
available only for BNi-OPiv. The solid-state molecular
structures of BNi-OPiv, ANi-NC12H8, and ANi-SArTip2 are
depicted in Figure 1, and Figures S1 and S2 provide details for
C
A−ONi-SArTip2 and a connectivity plot for Ni-SArTip2.
In all cases, the structure metrics agree with planar
coordination at nickel and the relative alignment of the
primary ligand and coligand RE− inferred from solution NMR
data. Taking ANi-NC12H8 as an example, the coordination
plane at nickel and the π-system of the carbazolide enclose an
angle of 69°, confirming previously reported density functional
theory (DFT)-derived structure metrics for an isomeric κ3-
N,C,N-coordinate Ni(II) complex.23 The bond geometry at the
N atom of the carbazolido ligand is trigonal-planar akin to that
a
Structures of A/A−ONi-NTMS2 and ANi-OPiv were reported
previously.27.
1164
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
Å.27 The coplanar alignment of the π-systems at the pendant
aryl and the phenanthroline backbone found in ANi-ER
diminishes slightly in A−ONi-ER as the torsion angle about
the C−C linkage increases from ∼1 to ∼7° in the thiolato and
amido complexes.
Electronic Properties. Ultraviolet/Visible/Near Infrared
(UV/Vis/NIR) Absorption Spectra. The UV/vis/NIR absorp-
tion spectra provide details as to the effects of the N-
heterocyclic backbone, coligand RE− and C- versus O-bonding
of the pendant aryl side arm on chromophore properties.
Starting with the impact of the N-heterocyclic backbone, the
conversion of 1,10-phenanthroline A into 10,11,12,13-
tetrahydrodipyrido[3,2-a:2′,3′-c]phenazine B and fully aro-
matic congener C extends the manifold of π*-type acceptor
MOs successively, and molar absorptivity in the near-UV and
visible spectral regions parallels this order as exemplified for
complexes A/B/CNi-OPiv in Figure 2B. With regard to band
Figure 1. ORTEP plots and selected metrics for solid-state molecular
t
structures of A,BNi-ER, for RE− = BuC(O)O− (top), H8C12N−
(center), and 2,6-(2,4,6-iPr3C6H2)2C6H3S− (bottom). The thermal
displacement ellipsoids are set at 50% probability; H atoms are
excluded for clarity.
27
A
of Ni-NTMS2 but the smaller interplanar angle adopted in
the solid-state structure of ANi-NC12H8 (69 vs 89° in Ni-
NTMS2) indicates that the lone-pair 2p orbital at N tilts
A
toward the π-system of the phenanthroline backbone. The
A
same applies to one of the lone-pair 3p orbitals at S in Ni-
SArTip2 that aligns perpendicularly with respect to the S−
C(aryl) bond and thus tilts even further toward the π-system of
the phenanthroline fragment. Experimental and computational
data described further below indicate that the coplanar
alignment of a lone-pair np-orbital at E and the π-system of
the N-heterocyclic backbone is key to substantial chromophore
absorptivity across the visible spectral region.
Figure 2. UV/vis/NIR absorption solution spectra (THF, 25 °C):
(A) A/B/CNi-SArTip2, SArTip2 = 2,6-(2,4,6-iPr3C6H2)2C6H3S−,
concentrations = 0.8−2.5 × 10−5 M; (B) A/B/CNi-OPiv, OPiv =
tBuC(O)O−, concentrations = 0.8−1.2 × 10−5 M.
energies, absorption profiles are almost congruent for A/BNi-
OPiv, whereas that for CNi-OPiv locates at lower energies.
This trend is more pronounced for thiolato complexes A/B/CNi-
SArTip2 as shown in Figure 2A. Contrary to the high-energy
part of the visible spectral region at λ ≤ 500 nm that shows the
same trend for band energies, band positions in the low-energy
visible region for λ > 500 nm are remarkably similar within
each series A/B/CNi-ER with a given coligand RE− as can be
seen in the insets of Figure 2 for aryl-S− and pivaloate and in
Figure S3 for RE− = TMS2N−.
As noted in Figure 1, the Ni−N/N′ bond lengths differ
significantly in all structures owing to the trans influence of the
strong aryl σ-donor. More notably, there is a distinct effect of
RE− on the Ni−N′ bond length, shortened for carboxylate and
carbazolide and elongated for thiolate and the nonaromatic
amide,27 and may be rationalized by the different σ- and π-
donor capabilities of the RE coligand. The distinct impact of
RE− on the redox and visible absorption properties is described
in the following subsections.
Expanding the five-membered κ2-N′,C-chelate ring moiety in
ANi-ER to the six-membered κ2-N′,O-moiety such as in A−ONi-
SArTip2 has a strong impact on bond metrics. The six-
membered chelate fragment appears to reduce geometric strain
at Ni as the Ni−O bond is shorter by 0.09 Å compared to the
C
C
Comparing the absorption profiles for Ni-OPiv and Ni-
SArTip2 in Figure 2A,B attests to the pronounced effect of
coligand RE− on overall absorptivity of κ3-N,N′,C-type
chromophore systems in the visible spectral region. Taking a
series of C-type ligand complexes as a representative example,
the visible absorption profiles plotted in Figure 3 are similar for
carboxylate (dotted black) and phenolate (solid brown),
although CNi-OPhtBu2 exhibits an additional low-intensity
A
Ni−C bond in Ni-SArTip2. In turn, the Ni−N′ and Ni−S
bonds are elongated by 0.049 and 0.026 Å, respectively, very
similar to the couple ANi-NTMS2 and A−ONi-NTMS2 for
which Δd(Ni−N′) = 0.054 Å and Δd(Ni−NTMS2) = 0.019
1165
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
in the low-energy visible region toward 700 nm. The marked
reduction of absorptivity in the vis-to-NIR region for
chromophores A−ONi-ER indicates a substantial impact of
the pendant aryl/phenol donor on the energy of frontier MOs.
The variation of visible band energies on solvent polarity
attests their charge transfer (CT) character and associates with
a decrease of the permanent dipole moment upon going from
the ground state to the electronically excited state. Taking CNi-
SArTip2 as an example, the band at lowest energy blue-shifts by
∼1000 cm−1 upon substituting pentane (λmax = 690 nm) for thf
(λmax = 654 nm) and 1,2-difluorobenzene (1,2-C6H4F2; λmax
=
647 nm) as shown in Figure 5. The high-energy visible band,
Figure 3. UV/vis/NIR absorption spectra (THF, 25 °C): CNi-ER,
RE− = OPiv (tBuC(O)O−), OPhtBu2 (2,4-tBu2C6H3O−), NTMS2
((Me3Si)2N−), and SArTip2 (2,6-(2,4,6-iPr3C6H2)2C6H3S−). Concen-
trations = 0.7−1.1 × 10−5 M.
band (ε ≈ 1.5 × 103/(M cm)) that reaches the low-energy
visible region at 600−700 nm as can be seen in the inset of
Figure 3. Substitution for amide (solid red) results in an overall
stronger absorptivity and a redshift of the absorption maxima.
Similar to Ni-OPhtBu2, Ni-NTMS2 displays weak absorptiv-
ity in the 600−700 nm region. Finally, substitution for aryl-S−
leads to an absorption profile (solid blue) that spans the visible
region for all complexes A/B/CNi-SArTip2 as shown in Figures 2
and 3.
C
C
Figure 5. UV/vis/NIR absorption spectra (solvents as indicated, 22
C
°C): Ni-SArTip2, SArTip2 = 2,6-(2,4,6-iPr3C6H2)2C6H3S−. Concen-
trations = 8.2−8.7 × 10−6 M.
The overall picture for C-type chromophores applies
likewise to their A-type congeners. Substitution of the
observed at 500 nm in thf solution, is similarly affected,
whereas a negligible shift results for the bands at λ ≤ 400 nm,
which points to electronic transitions between delocalized sets
of MOs in the near-UV. Figure S4 compares UV−vis−NIR
absorption profiles for the series of complexes A/B/CNi-SArTip2
recorded in THF and in 1,2-difluorobenzene solutions. The
negative solvatochromism found for the high-energy visible
bands in the 400−500 nm region (cf. inset, Figure S4)
increases noticeably along A < B < C, in particular in the
transition from B to C. In other words, the change of the
ground-state permanent dipole moment that associates with
the CT transition(s) in the high-energy visible region is largest
A
nonaromatic amide in Ni-NTMS2 for carbazolide results in
a shift of the visible absorption profile to higher energies as
shown in Figure S3. Notably, ANi-NC12H8 exhibits a high-
energy visible band at 430 nm that is reminiscent of pivaloate
complexes A/BNi-OPiv, shown in Figure 2. In contrast to
A
A/BNi-OPiv, Ni-NC12H8 absorbs up to 700 nm, albeit at ∼4
times lower intensity compared with A/BNi-SArTip2. Overall, it
is the π-donor capability of coligand RE− and its relative
alignment toward the N-heterocycle π*-orbitals that relates to
absorptivity in the low-energy visible spectral region, whereas a
lower impact results for high-energy visible bands that depend
primarily on the type of N-heterocyclic moiety.
C
for Ni-SArTip2.
Emission Spectroscopy. Compounds ANi-NC12H8,
A/A−ONi-NTMS2, and A/A−ONi-SArTip2 have been chosen to
address the emission properties of this class of nickel complex
chromophores. Neither of the chromophores luminesces in
dilute (≤50 μM) 2-Me-thf solution (293 K) and glass (77 K)
upon excitation of the CT bands in the visible region or the
π−π*-type transitions in the near-UV. The same applies to the
crystalline materials when irradiated by an UV lamp.
Irradiation at λrad = 265 nm stimulated the appearance of a
very weak emission band at λem ≈ 350 nm (and its second
harmonic at ∼700 nm) as depicted in Figure S12. This
emission appears to relate to a metal ion impurity and/or
defect structure in the borosilicate glass of the sample tubes
because it is strongly reduced in quartz glassware under
otherwise identical conditions.
In stark contrast to the variable absorption properties of aryl-
metalated chromophores A/B/CNi−ER, visible absorption
profiles of phenol congeners A−ONi-SArTip2 and A−ONi-
27
NTMS2 differ primarily in terms of molar absorptivity. As
shown in Figure 4, the UV and visible light absorption profile
of A−ONi-SArTip2 (solid line) is very similar to that of A−ONi-
NTMS2 (dotted line), and both exhibit only weak absorptivity
Cyclic Voltammetry (CV). Scheme 3 provides an overview
of the redox processes supported by compounds A−O/A/B/CNi-
ER as determined by CV, and Table 1 displays the pertinent
data.
First, all complexes undergo two reduction events within the
cathodic potential limit of the electrolyte medium, that is ca.
−2.72 V vs the ferrocene/ferrocenium couple (V vs Fc/Fc+).
Figure 4. UV/vis/NIR absorption spectra (THF, 25 °C): A−ONi-
SArTip2 (solid line; SArTip2 = 2,6-(2,4,6-iPr3C6H2)2C6H3S−) and
A−ONi-NTMS2 (dotted line; included for comparison).27 Concen-
trations ≤ 3.0 × 10−5 M.
1166
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
Scheme 3. Redox Processes of Complexes A−O/A/B/CNi-ER
sibility disallows quantitative comparisons, but the shift of E3
to more positive potentials for increasing potential sweep rates
and sample concentrations indicates that heterogeneous
electron transfer is rate-limiting, which points to a critical
effect of RE− on electronic relaxation during the electron
transfer to and from the electrode. An irreversible chemical
reaction that withdraws the electrochemically generated
complex cation from its Nernst equilibrium would result in
the opposite trend of E3.30 Further oxidation to complex
dications either is electrochemically irreversible or may occur
at potentials more positive than the limit supported by the
electrolyte medium for complexes from the C-type ligand
framework.
The first reduction is electrochemically reversible in all cases,
0
and E1 values vary by ≤0.35 V between N-heterocyclic
moieties A and C. For RE− = aryl-S−, E1 values are very
0
similar for backbones A and B, whereas a shift of ∼0.25 V to
more positive potentials is found for C. An equivalent picture
The third general property considers the effect of the
bonding mode of the pendant aryl donor on reduction
A
results for the couple of amido complexes Ni-NTMS2 and
CNi-NTMS2. As to the impact of coligand RE− on E0 , two
potential E0 and oxidation potential E3. The apparent minor
1
1
impact of the pendant C-/O-aryl on E0 attests that reduction
subclasses can be differentiated: (a) the weak and moderate π-
donors pivaloate (carboxylate) and carbazolide and (b) the
stronger π-donors (Me3Si)2N− and aryl-S−. A higher π-donor
capability appears to render reduction thermodynamically less
favorable, with ΔE0 ≤ 0.07 V for C-type and ΔE0 ≥ 0.15 V
1
is confined to π*-orbitals at the N-heterocyclic backbone. Such
an electronic compartmentalization agrees with a combined
electrolysis/cw-EPR and DFT study of [(κ3-N,N′,C)Ni−Cl]
complexes from a 2,2′-bipyridine type ligand platform.21 With
regard to oxidation, O-bonding associates with an anodic shift
of E3 by ≥0.3 V irrespective of RE− and indicates a substantial
contribution of the phenolato fragment to the HOMO of
A−ONi-ER.
Characterization of Radical Cation Complexes. Com-
plexes ANi-NTMS2, A−ONi-NTMS2, and ANi-SArTip2 have
been chosen as representative examples to further address the
distinct effects of C-/O-bonding and coligand RE− on redox
properties. As described by Scheme 4, stoichiometric oxidation
1
1
for A-type backbones. According to computational data
described further below, the lower impact of coligand RE−
on E0 in case of backbone C results from the inversion of
1
energetically close LUMO and LUMO+1 that associates with
the transition from A/B to C.
Contrary to the first reduction, the second reduction event is
electrochemically reversible at υ ≤ 1000 mV s−1 for
0
(tetrahydro)phenazine platforms B and C only, and E2
correlates with the sizes of their π-manifolds as can be seen
from comparing the first three entries of Table 1. According to
the computational data reported below, the electrochemical
reversibility associates with the localization of the two electrons
in spatially different and electronically weakly coupled acceptor
π-orbitals in case of B and C.
Scheme 4. Chemical Oxidation of A−O/ANi-ER
The second general property is that all complexes can be
oxidized at least once up to the high potential limit, ∼1.7 V, of
the electrolyte medium. Most notably, only the amido
complexes support electrochemically reversible oxidation, and
0
A
A
E3 values for Ni-NTMS2 and Ni-NC12H8 differ by 20 mV.
Electrochemically irreversible oxidation prevails otherwise,
irrespective of the type of N-heterocycle A−C and the pendant
C-/O-aryl group, although peak potentials E3 are remarkably
similar for the O- and S-coligands. Electrochemical irrever-
with [(4-BrH4C6)3N]BArF4 [BArF4 = (3,5-(F3C)2C6H3)4B−] at
T ≤ −60 °C in 1:1 toluene/CH2Cl2 and CH2Cl2 solutions
afforded complex cations for X-band cw-EPR (Figures S6−
S11) and UV/vis/NIR absorption spectroscopic analysis
a
Table 1. Redox Potentials from CV
b
b
b
b
backbone
RE−
E1
E2
E3
Ep,4
A
B
C
C
Tip2ArS−
Tip2ArS−
Tip2ArS−
tBu2PhO−
TMS2N−
PivO−
−2.06
−2.71 (ir)
−2.53
−2.32
−2.24
−2.29
−2.17
−2.62 (ir)
−2.55 (ir)
0.08−0.11 (ir)
0.09−0.11 (ir)
0.04−0.11 (ir)
0.07−0.10 (ir)
0.08
0.09−0.13 (ir)
0.07 (ir)
0.07
0.20 (ir)
0.22−0.26 (ir)
−2.00
−1.75
−1.73
−1.73
−1.68
−1.91
−1.92
−2.08
−2.06
C
C
A
A
A
O−A
O−A
PivO−
0.39 (ir)
0.81 (ir)
c
H8C12N−
TMS2N−
TMS2N−
Tip2ArS−
0.05
0.36
0.30 (ir)
d
1.01 (ir)
≥0.78 (ir)
de
,
f
−1.97 (qr)
−2.28 (ir)
a
b
Conditions: glassy carbon working electrode, 0.1 M nBu4NPF6/1,2-difluorobenzene electrolyte solution, T = 20 °C, 50 ≤ ν ≤ 2000 mV/s. E (V
c
vs Fc/Fc+); ir/qr = electrochemically irreversible/quasireversible, Ep at ν = 100 mV/s. A third irreversible oxidation occurs at Ep(100 mV/s) =
1.04 V. Upon reversal of scan direction, the wave for reoxidation is broadened at all scan rates. Reduction induces a follow-up reactivity leading to
another oxidation event at Ep,ox = −0.65 V. A series of ill-defined oxidation events is observed.
d
e
f
1167
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
(Figure S5). Table 2 lists the principal values of g tensors and
hyperfine coupling A(14N, I = 1, 99.64%).
lower energy to result in the emerald green color of the Ni−
amidyl radical complex cation.
Amido ligand centered oxidation was further corroborated
by using 1,1′-diacetylferrocenium bis(trifluoromethane-
sulfonyl)imide, [Fc″]NTf2, as the oxidant. In addition to the
amidyl-radical complex cation, a Ni(III) complex forms as a
minor coproduct (∼10%) with gx = 2.434 and gy = 2.307, while
gz is masked by the resonances of the Ni−amidyl radical cation
as can be seen in Figure S8. Assuming gx ≈ gy ≫gz equals the
order reported for complexes of 5- and 6-coordinate
Ni(III)33−36 and is herein attributed to O-bonding of the
(F3CSO2)2N-counterion. The emerald green Ni−amidyl
radical cation converts into a yellow, EPR-silent product or
mixture of products at ambient temperature but on distinct
time scales depending on the counterion, namely, minutes for
NTf2 and hours for the weakly coordinating borate.
Table 2. X-Band cw-EPR Spectroscopic Parameters of
a
Complex Cations A−O/ANi-ER
b
c
d
precursor
gx, gy, gz
2.110 [0.018]
giso
Ax, Ay, Az
n.r.
58
ANi-NTMS2
2.066 [0.011] 2.068 [2.068]
2.028 [0.010]
2.046 [0.005]
n.r.
n.r.
n.r.
n.r.
n.r.
n.r.
n.r.
A−ONi-NTMS2,
minor product, I
2.026 [0.005]
2.015 [0.003]
2.057 [0.015]
2.043 [0.006]
2.002 [0.003]
2.272 [0.017]
2.031
[I, 2.029; II, 2.034]
A−ONi-NTMS2,:
major product, II
A
Contrary to the selective oxidation of Ni-NTMS2 in the
e
ANi-SArTip2
2.099 [0.018] [2.141]
2.052 [0.009]
presence of weakly coordinating borate, oxidation of A−ONi-
NTMS2 with [(4-BrH4C6)3N]BArF led to a mixture of two
4
a
Conditions: 1/1 toluene/CH2Cl2, glass (100 K) and solution (213
products that adopt a ∼1:1.5 ratio according to EPR spectra
acquired at 100 K as shown in Figure S9. We attribute the
minor component to the product from oxidation of the
primary ligand, namely, [(κ3-N,N′,O•)Ni(−NTMS2)]+, and the
major component to the product from oxidation of the amido
ligand, namely, [(κ3-N,N′,O−)Ni(•NTMS2)]+. Different from
K), oxidant = [(4-BrC6H4)3N]B(3,5-(F3C)2C6H3)4, T = −78 °C.
b
Distribution of principal g values represented by the “gStrain”
parameter used in EasySpin31 given brackets. [giso = 1/3(gx + gy +
c
d
gz)] from EPR data at 100 K. A(14N, I = 1, 99.64%) in MHz; n.r. =
e
not resolved. Spectrum (Figure S11) shows an additional set of
resonances of a minor S = 1/2 species.
A
oxidized Ni-NTMS2, hyperfine splitting is not resolved for
any of the three components of the A tensors of either redox
isomer, and the anisotropies of g tensors are much smaller.
Assuming that fragment rotation about the Ni−N bond to the
amidyl-radical is less restricted than about the Ni−O bond to
the rigid primary ligand, we assign the Ni−phenoxyl radical
cation to the g tensor with the narrower distribution of
principal components. At 213 K, the subspectra collapse into a
single broad derivative line shown in Figure S10, with giso being
Rhombic g tensors prevail in all three cases, and the principal
g values exceed the free electron g value ge ≈ 2.003, indicating a
variable admixture of Ni-borne orbitals to the singly occupied
frontier MO (SOMO). On the basis of simulations of EPR
spectra depicted in Figures S6 and S8, oxidized ANi-NTMS2 is
best described as a nickel-amidyl radical complex cation with
unpaired electron density locating primarily in an N 2p orbital.
Similar to a Rh-amidyl radical cation,32 location of spin density
in a p orbital relates to the distinct anisotropy and small size of
principal A values, which in this case indicates orientation of Ay
along the N 2p orbital that likely orients along the
coordination plane at Ni. Considering Ay as the only resolved
principal hyperfine component allows reproducing the distinct
curvature of the central derivative line depicted in Figures S6
and S8. Each of the principal g components shows a significant
distribution in frozen solution that is attributed to a
distribution of interplanar angles enclosed by the putatively
trigonal amidyl-radical fragment and the primary ligand
framework and affects π-type Ni−NTMS2 3d−2p orbital
overlap. Warming the sample to 213 K results in a broadened
and featureless derivative line (cf. Figure S7) that persists upon
further warming to ambient temperatures.
equal to the weighted average calculated from giso,I and giso,II
,
which points to the rapid interconversion of the redox isomers
on the experimental time scale. Same as for oxidized ANi-
NTMS2, oxidation products from A−ONi-NTMS2 convert
gradually into an EPR silent product or mixture of products
within hours at room temperature.
A
Oxidation of Ni-SArTip2 at −78 °C affords a red-brown
solution that changes color instantaneously to yellow and
finally green upon warming to ambient temperature. EPR
spectra at 100 K point to the formation of two products as
depicted in Figure S11. The major product is characterized by
a g tensor with gx ≫ gy, gz that each shows substantial
distribution, pointing to a shallow potential well for rotational
motion about the Ni−S bond akin to ANi-NTMS2. We
attribute the marked deviation of one of the principal
components of the g tensor from ge to the localization of
spin density in a 3(d−p) Ni−S π-bond, which implies a
partially covalent character of named bond also in parent
complexes A/B/CNi-SArTip2. The g tensor for the minor
component cannot be fully deduced but anisotropy appears
much smaller and suggests a diminished 3(d−p) Ni−S π-bond
character. We therefore tentatively assign the minor
component to a complex nickel-thiyl radical cation in which
the aryl-S fragment adopts an obtuse interplanar angle such
that spin-density localizes predominantly at sulfur. In line with
this description, EPR spectra recorded at 213 K appear to
consist of an extremely broadened derivative line that hardly
distinguishes from the baseline. Further warming of the sample
Monitoring the oxidation of ANi-NTMS2 at −60 °C by UV/
vis/NIR absorption spectroscopy corroborated the formation
and persistence of the complex amidyl-radical cation. In line
with the aforementioned influence of coligand RE− on the
individual bands in the visible spectral region, the intensity of
the high-energy visible band of the starting material markedly
decreased, whereas the low-energy visible band disappeared in
favor of an intense band that locates at even lower energy as
depicted in Figure S5. The occurrence of the low-energy band
points to a putative CT transition from a lower lying occupied
MO into the SOMO that derived from the former HOMO.
Overall, the marked attenuation of absorptivity in the high-
energy visible region combines with intense absorption at
1168
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
inside the microwave cavity to room temperature caused
instantaneous and irreversible degradation.
strong π-donor ligands. However, the character of the HOMO
is subject to qualitative change when RE− is varied. The
Computational Data. Structures of complexes A−ONi-ER
and A/B/CNi-ER for coligands RE− = OPiv, NC12H8, NTMS2,
and SArTip2 have been optimized with density functional
theory (DFT) methods using the BP86 functional. Details of
the computational settings can be found in the Experimental
Section. In addition, [A−O/A/B/CNi-StBu]+/0/−/2− and A−O/A/B/C-
Ni-SArMe2, with Me2ArS−= 2,6-(H3C)2C6H3S−. have been
optimized as truncated models of A−O/A/B/CNi-SArTip2.
Predicted metrics have been calibrated with experimental
data. An excellent match is generally observed, except for the
artificially elongated Ni−N bond located trans to the Ni−C
bond. Reference calculations with the Ni−N bond length
constrained to the experimental value resulted in marginal
energy differences. Pertinent metrical data is provided in
Tables S1−S3. EPR parameters of cations [ANi-NTMS2]+,
highest donor orbital is clearly metal-borne (dz ) in Ni-OPiv,
A
2
as has been found previously.27 MOs of Ni−O(Piv) 3d−2p π-
bond character are highly stabilized, E < -6 eV, and do not
contribute to the frontier MOs. In consequence, oxidation of
ANi-OPiv is expected to be nickel-centered, and it is noted that
DFT-optimization of [ANi-OPiv]+ involves κ2-O,O′-OPiv
bonding, which stabilizes Ni(III) by increasing the coordina-
tion number. The presence of nominally stronger π-donor
coligands introduces MOs of Ni−E 3d−2/3p π-bond character
at far more positive energy, rendering them accessible to
A
A
optical and electrochemical techniques. In Ni-NTMS2, Ni-
NC12H8, and ANi-SArTip2 (and respective models ANi-
SArMe2 and Ni-StBu) the HOMO indeed manifests largely
A
as a 2/3p orbital localized on E, featuring π-antibonding
interactions with a Ni dxy orbital. Oxidation of these complexes
is expected to be ligand centered, as is found in experiment.
DFT-derived EPR parameters for [ANi-NTMS2]+ and [ANi-
SArTip2]+ (Table S7) corroborate the distinct anisotropy of g
tensors found experimentally, lending strong support to the
DFT assignment of the donor orbitals.
We next turned to the effects of the manifold of π-orbitals at
the N-heterocyclic moiety, using A/B/CNi-StBu as a representa-
tive model for A/B/CNi-SArTip2. Frontier MOs for the series
are shown in Figure 7, and very similar results are obtained all
across the RE− substitution series.
[
A−ONi-NTMS2]+, and [ANi-SArTip2]+ were calculated using
the B3LYP functional. Frontier MO diagrams and related
electronic properties were addressed in single-point calcu-
lations with the TPSSh functional as established in previous
work on planar Ni(II) complexes.22,27,37
We first focus on the impact of coligand RE− on the
electronic structure in ANi-ER; equivalent conclusions hold for
derivatives of B but not for C. Kohn−Sham frontier MOs are
depicted in Figure 6.
Figure 7. DFT-derived Kohn−Sham frontier MOs (TPSSh-D3/
TZVP level of theory). Effect of π-extension on energetic order and
character of A/B/CNi-StBu. Occupied MOs, blue; virtual orbitals, red.
As expected, the donor levels are hardly affected by the
expansion of the π-system from A to C. Intriguingly, also the
overall appearance of the acceptor levels is largely unaffected.
For A, they are centered at the κ2-N,N′-bipyridine moiety and
are energetically close (E = −2.71 and −2.49 eV), and for C,
they change marginally (E = −2.74 and −2.42 eV). Expansion
of the N-heterocyclic π-system adds qualitatively new donor
levels to the MO diagram at E = −2.56 and −3.08 eV for B and
C, respectively. In consequence, the first reduction event
localizes at the κ2-N,N′-bipyridine fragment in A and B at very
similar potentials, whereas a phenazine-centered reduction is
favored by ∼0.3 V in case of C owing to an inversion of the
LUMO character. Computed Mulliken spin densities of the
Figure 6. DFT-derived Kohn−Sham frontier MOs (TPSSh-D3/
TZVP level of theory). Effect of coligand RE− on the energetic order
and the character of ANi-ER. Occupied MOs, blue; virtual orbitals,
red.
Evidently, the nature of RE− does not affect the character of
the LUMO that is centered at the κ2-N,N′-bipyridine moiety in
all cases, albeit it modulates the energy by up to 150 mV. In
agreement with electrochemical data, the range of LUMO
energies coincides with the RE-dependent variation of E0
1
given in Table 1. The computed divergence of LUMO energies
reproduces the phenomenological categories of carboxylate
and carbazolide as weak/moderate and TMS2N− and aryl-S− as
1169
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
anions [A/B/CNi-StBu]−, included in Figure S13, visualize the
qualitative difference between B and C as they stand for the
contrasting character of the parent LUMO. The second
reduction step of B and C yields triplet dianions with an
integer spin residing both in the phenazine and the
phenanthroline moiety. Evidently there is only minor
electronic communication among both moieties, allowing for
the reversible storage/donation of two electrons in electroni-
cally distinct sections of a single acceptor ligand.
Finally, we addressed the substantial impact of O-bonding
on chromophore properties of A−ONi-SArTip2 and A−ONi-
NTMS2. While the character of the LUMO and the Ni−S 3d−
3p π-bond nature of the HOMO are conserved in A−ONi-ER,
the corresponding stabilization of the HOMO inferred from
electrochemical and electronic absorption data is quantitatively
echoed in a DFT-derived MO diagram shown in Figure 8.
Having established level-crossing in the electrochemically
addressable acceptor orbitals between B and C, it is intriguing
to recall the experimental vis/NIR spectra of series A/B/CNi-
SArTip2 in Figures 2 and 3. According to the absorption
spectra, both the energy and the intensity of the lowest-energy
transitions around 650 nm hardly respond to the nature of the
π-system of the N-heterocyclic backbone. This finding is
commensurate with the MO diagram shown in Figure 7 only if
the electroactive LUMO of C is not involved in the diagnostic
low-energy transition. Instead, the low-energy bands are best
interpreted as CT transitions between thiolate-borne donor
levels and phenanthroline centered π*-orbitals, namely, the
LUMO for A and B and LUMO+1 for C. This assignment is
corroborated by the decent match of the experimental vis/NIR
band energies and the HOMO−LUMO(+1) gaps, which
amount to ∼2.0 eV. Similar interpretations hold for the spectra
of amides A/B/CNi-NTMS2. The transitions at longest wave-
lengths here are likewise independent of the extent of the N-
heterocycle π-manifold, featured as weak shoulders (ε < 1000/
(M cm)) at ca. 600 nm, again in agreement with the vertical
HOMO−LUMO(+1) gap as depicted in Figure S15. This
change of the nature of the acceptor orbital from LUMO to
LUMO+1 for increasing π-extension of the N-heterocyclic
moiety and the associated invariability of lowest-energy
electronic transitions are reminiscent of the behavior reported
for planar complexes [(κ3-N,Amido,N)Ni−Cl] from a series of
phenanthridine-containing ligands.38
Figure 8. DFT-derived Kohn−Sham frontier MOs (TPSSh-D3/
TZVP level of theory). Effect of C-/O-metalation on energetic order
and character of HOMO−n, n = 0−2, for A−O/ANi-SArTip2.
The increasing HOMO−LUMO gap derived from DFT
data explains the experimentally observed confinement of
absorptivity to the high-energy visible region. In general,
spectra from series A−ONi-ER are much less prone to donor-
Closer inspection of the orbitals involved in the optical
transitions hints to the origin of the massive intensity
difference of ∼1:4 between the HOMO−LUMO(+1)
transitions of the series of amido and thiolato complexes.
These very transitions are forbidden by symmetry both in the
amides and the thiolates, due to the (almost) orthogonal donor
and acceptor arrangement in the optimized structures,
rendering the transitions silent in TD-DFT. Rotation about
the Ni−E bond will lift these limitations, and it is tempting to
ascribe the recovered intensity of a symmetry-forbidden
transition to the rotation about the Ni−S bond in A/B/CNi-
SArTip2. Indeed, the existence of rotamers was inferred as the
source of speciation in solution-phase EPR spectra. The
rotamer hypothesis was tested using ANi-SArMe2 as a
A
dependent variation than the spectra of parent series Ni-ER.
DISCUSSION
■
A
The lack of luminescence for Ni-NC12H8 and its congeners
contrasts reported emission properties for the complex
fragment [Ni−NC12H8]+ embedded in a symmetric κ3-
N,C,N-bound 2,6-bis(pyrid-2-yl)phenyl framework.23 Since
orbital symmetry and ostensibly minor structural differences
appear to have a pronounced impact on chromophore excited-
state properties, we herein provide a detailed picture of the
ground-state structure−property relationship for square-planar
complex chromophores of Ni(II) in rigid, low-symmetry κ3-
N,N′,C/O-type environments. To this end, we have separately
addressed the effects of (a) the N-heterocyclic moiety as
charge acceptor for light-induced CT, (b) the fragment Ni−ER
as the corresponding charge donor, and (c) C- versus O-
metalation of the pendant aryl-linkage that connects the former
two.
A
truncated model for Ni-SArTip2; the structures are given in
A
Figure S16. Optimization of Ni-SArMe2 gave two close-to-
degenerate minima, for which rotation aligns the N-
heterocyclic π-system coplanar with an antibonding Ni−S
3d−3p π-type orbital combination. Accordingly, TD-DFT
computation gave intense vis−NIR transitions for both rotated
species. The validity of the argument was cross-checked
Effect of N-Heterocyclic Moiety. The electrochemical
and visible absorption spectroscopic data corroborate that the
N-heterocyclic backbone is electronically distinct from the
pendant C-/O-aryl moiety even though their π-systems are
coplanar. The role of a N-heterocycle borne π*-orbital as the
chromophore LUMO matches assignments for structurally
related complexes of Ni21,39 and Pt29,40 and seminal work on
the spin density distribution of radical anion complexes of
A
through TD-DFT analysis of a constrained structure of Ni-
SArMe2, with orthogonally fixed Ni−S−Ar constellation, akin
to the global minima of ANi-SArTip2 and A/B/CNi-StBu. In
keeping with the rotamer hypothesis, orthogonally constrained
ANi-SArMe2 gives no intense vis−NIR transitions.
1170
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
1,10-phenanthroline41,42 and dipyrido[3,2-a:2′,3′-c]-
dissipation through energetically matched overtones of C−H
stretching vibrations of the solvent and/or the ligand
framework may be effective for excited-state quenching.47
Light-promoted CT from the {Ni−ER} fragment into the π*-
manifold of the N-heterocycle results in a partial Ni(III)
character of the electronic excited-state, which can further
increase due to rotational motion about the Ni−E bond or the
coordination of potential donors such as 2-Me-thf. The
bonding of the (F3CSO2)2N-anion inferred from EPR
spectroscopic data adds support to the role of additional
donors in excited-state quenching, and weak interactions
between Ni and the ipso-C atom of the flanking 2,4,6-iPr3-
phenyl groups may contribute in the case of the SArTip2
coligand.48 Finally, tetrahedral distortion at Ni can add a
potential deactivation pathway toward an excited d−d
state,25,26 even though this process may be regarded as
unfavorable considering the overall rigidity of the primary
ligand framework.
phenazine.43−45 The trend among the reduction potentials
E0 attests that extending the conjugation of the acceptor π-
1
system provides the means to modulate the LUMO energy,
albeit in a discontinuous fashion. Appending a pyrazine moiety
to 1,10-phenanthroline A in the form of 10,11,12,13-
tetrahydrodipyrido[3,2-a:2′,3′-c]phenazine B has a marginal
effect on the acceptor strength as the formal potential for the
first reduction E0 changes by +0.06 V. Contrary to the first
1
reduction, extending the π-system has a marked effect on the
second reduction that becomes electrochemically reversible.
Such behavior points to an electronic compartmentalization of
the π-orbital manifold such that first and second reduction
locate at spatially different parts of N-heterocyclic fragment B,
namely, at the κ2-N,N′-bipyridine and 10,11,12,13-tetrahy-
drophenazine moiety, respectively. In principle, this picture is
reminiscent of the manifold of lowest unoccupied MOs of the
parent ligand dipyrido[3,2-a:2′,3′-c]phenazine in complexes of
Ru(II) whose first and second reduction occur at spatially
distinct π*-type MOs but in reverse order, namely, at the
phenazine and the bipyridine moiety, respectively.43 Taking
the π-conjugation one step further in the form of dipyrido[3,2-
Effects of Coligand RE−. The combined effect of coligand
RE− and the N-heterocyclic backbone on chromophore
ground-state electronic properties is the result of a fine balance
of their respective π-donor and acceptor interactions with
occupied d orbitals at nickel. Solid-state and solution structure
analysis confirm that the lone-pair np orbital at E and the π*
orbitals at the κ2-N,N′-heterocyclic moiety adopt variable
relative orientations. A perpendicular alignment prevails for
a:2′,3′-c]phenazine C finally induces a substantial shift of E0
1
(and E0 ) to a more positive potential. Notably, the ΔE01 ≈ 0.3
2
V is reminiscent to the change in reduction potentials reported
for the (bipyridine)2Ru(II) complex fragment of the respective
κ2-N,N-dipyrido[3,2-a:2′,3′-c]phenazine parent ligands of B
and C, albeit its electrochemical reduction is irreversible.46
Rendering ligand reduction energetically more favorable
however associates with a reduced capability for 3d−π*-
back-bonding and an increasing CT character for electronic
transitions that connect to the LUMO. As a consequence, the
respective absorption bands are expected to display a redshift
and an increasingly negative solvatochromism. Inspection of
the UV−vis−NIR absorption data indeed adds support to this
picture in form of the high-energy visible band whose position
shifts substantially to lower energy for CNi-SArTip2 that
exhibits the largest solvatochromism among A/B/CNi-SArTip2.
This indicates a reversal of the order of unoccupied π*-
acceptor states in case of the C-type chromophores, namely,
the phenazine fragment characterizes the LUMO, whereas the
κ2-N,N′-bipyridine forms the LUMO+1. The computational
analysis of unoccupied frontier MOs corroborates the picture
of reversing acceptor state energies in the transition from A
and B to C-type chromophores.
27
ANi-NTMS2 and presumably for CNi-NTMS2, which renders
np(E)−π*(N-heterocycle)-orbital mixing through a Ni d
orbital insignificant. The relative orbital alignment becomes
more favorable in case of ANi-NC12H8, ANi-SArTip2 and
presumably CNi-OPhtBu2. In addition to relative spatial
orientation, composition analysis of DFT-derived frontier
orbitals indicates that the antibonding π-type Ni−E 3d−np
−
orbital overlap becomes increasingly dominant along F3CSO2
(see ref 27) ≈ carboxylato < phenolato ≈ carbazolido < amido
< thiolato. The HOMO of ANi-OPiv carries only 12% ER
character (>75% Ni), whereas the HOMO in ANi-SArTip2 and
ANi-NTMS2 largely localize on ER with contributions of >50%
each on the coordinated S and N atom (<20% Ni). In
consequence, the antibonding π-type Ni−E 3d−np orbital
overlap becomes increasingly dominant. Charge donation
toward nickel increases accordingly and has a marked effect
on the HOMO whose energy approaches that of the LUMO.
The decreasing HOMO−LUMO energy gap relates to
absorption profiles that increasingly expand across the visible
spectral region. In more general terms, the capability of donor
ligands to engage in π-type interactions with occupied metal d
orbitals increases the manifold of chromophore donor states
and aids absorptivity in the vis-NIR spectral region. Molar
absorptivity scales with the degree of coplanar alignment of
donor and acceptor orbitals and their mixing through an
occupied metal d orbital. This aspect adds to the concept of
extending the manifold of π-donor orbitals for symmetry
allowed mixing with 3d orbitals in case of chromophores of
octahedral Fe(II) and Fe(III).18
Observing very similar positions for the high-energy visible
bands of pivaloato and carbazolido as well as for amido and
thiolato complexes connects to their matching reduction
potentials E01 and supports the assignment of a HOMO−n (n
≥ 1)-to-LUMO CT character to the high-energy visible bands
(400−500 nm region) for all chromophores A/B/CNi-ER.
Notably, a similar description was reported for transitions
connecting to the LUMO of a structurally related chromo-
phore of Pt(II) that derives from a C-type κ3-N,N′,C-
framework. The photoluminescence of the Pt(II) system was
ascribed to a triplet excited-state of mixed intraligand π−π*
and MLCT parentage.40 While the primary ligand framework
suits to support photoluminescence as demonstrated by the
Pt(II) congener, we attribute the dominant nonradiative decay
of the Ni(II) chromophores to the combined effect of the CT
character of the transition, the shallow potential well for
rotational motion about the Ni−E bond and possible
coordination of additional donors at nickel. In addition, energy
Independent support for the role of the antibonding 3d−2/
3p π interaction on chromophore electronic properties derives
from electrochemical and bulk oxidation studies. Potential
values for the first oxidation are remarkably similar among the
different donors for a given bonding mode κ3-N,N′,C/O. As
electrochemical oxidation is irreversible for nonamido donors,
EPR data add vital information in form of the distinct
anisotropy (and distribution) of the principal components of
1171
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
the g and A tensors of oxidized amido and thiolato complexes
ANi-ER that corroborate the predominant contribution of E to
the SOMO. Adopting a simplified picture, the order of the π-
bond between the lone-pair 2/3p orbital at E and a doubly
fundamental fashion. While both donor systems are electroni-
cally distinct from the N-heterocycle π-system and therefore
have a minor impact on LUMO energy, chromophore
absorptivity in the low-energy visible region hinges on the
choice of the pendant C-/O-aryl side arm. Computational
analysis indicates that both donors support σ- and π-bonding
to Ni and contribute similarly to occupied frontier MOs for a
given coligand RE− but differ with regard to their impacts on
orbital energies. In full agreement with electrochemical,
absorption spectroscopic, and EPR data, the Ni−phenolate
moiety not only contributes substantially to the HOMO of
A−ONi-NTMS2 but also renders the HOMO energetically more
stable compared to that of C-metalated ANi-NTMS2. While the
LUMO energies appear largely conserved, an energetically
stabilized HOMO and destabilized LUMO+n (n ≥ 2) prevail
for O-metalated chromophore systems, in line with the lack of
low-energy bands in the vis-region.
1
occupied 3d orbital is 0 and increases to /2 upon oxidation,
which is supported by the ligand-radical character inferred
from EPR spectroscopic data in addition to the DFT-derived
structure metrics that predict a contraction of the Ni−E bond
by ≥0.1 Å upon oxidation; see Table S4. This description is
independently supported by the visible absorption profile of
A
the amidyl−radical complex cation formed from Ni-NTMS2,
namely, by the occurrence of an intense visible band at lower
energy compared to the starting material and the persistence of
the high-energy visible band whose intensity decreases. On the
basis of these considerations, we attribute the low-energy
visible bands to a HOMO-to-LUMO+n (n = 0 and 1) type
transition that involves CT from the π-type Ni−E 3d-np bond
into a π*-orbital at the κ2-N,N′-bipyridine fragment, namely,
the LUMO in case of A and B and LUMO+1 for C. As
described above, light-induced CT from the antibonding Ni−E
CONCLUSIONS
■
Structurally dissymmetric complexes of the general formula
[(κ3-N,N′,C/O)Ni−ER] provide a platform toward electroni-
cally variable complex chromophores of planar Ni(II). π-
Bonding of an occupied Ni 3d orbital with an np electron pair
at coligand RE− provides a mechanism for coupling the latter
to the π* orbital manifold at the N-heterocyclic moiety for
light-induced charge transfer, which results in substantial
chromophore absorptivity across the visible spectral region.
Molar absorptivity scales with the relative alignment of donor
and acceptor orbitals and maximizes for a coplanar orientation.
Frontier molecular orbital compositions and relative energies
can be modulated by considering three principal factors. The
first is the π-orbital manifold at the κ2-N,N′-coordinate N-
heterocyclic moiety whose electronic coupling to Ni(II)
through d−π* back-bonding has an impact on the character
of the electronic excitation. Second, the pendant C-/O-aryl
fragment sets the energies of occupied donor and unoccupied
acceptor frontier MOs other than the LUMO. Third, the σ-
and π-donor capability of coligand RE− controls transition
energies and electronic relaxation upon light-induced electron
transfer from the Ni−ER fragment.
1
3d−np π-bond increases the Ni−E π-bond order to /2 and
thereby can contribute to kinetically disfavor the population of
excited d−d states that bear a metal−ligand antibonding
character. A shallow potential well for Ni−E bond rotation
however has a diametric effect and obviously plays a significant
role as inferred from experimental and computational data.
Contrary to ligand-centered oxidation, metal-centered
oxidation to Ni(III) associates with a reorganization of the
coordination sphere through bonding of an additional donor.
As indicated by a Mulliken spin density of ρNi = 0.88 computed
for [ANi-OPiv]+, this scenario likely is relevant for nickel
carboxylate complexes owing to the capability of carboxylates
to κ2-bonding and provides an explanation for the irreversible
character of their electrochemical oxidation. Unfortunately, the
very low solubility of pivaloato complexes is incompatible with
low-temperature solution studies to probe this scenario. The
remarkable redox properties of the monodentate amido ligands
for both C- and O-metalated chromophores line up with
previous reports on pincer-type amidyl-radical complexes of
Rh32 and Ni49 and a terminal amidyl−radical complex of Cu.50
The reversibility of the electrochemical oxidation points to a
minor electronic and structure relaxation for the respective
redox conjugates. We attribute the favorable electrode kinetics
to polar effects that result from matching the nucleophilic
character of the forming N-borne radical51 and the electro-
philic character of the Ni(III) counterpart in preparation of the
transition state. Contrary to amides as ligands, a polarity
mismatch arises for the formation of O- and S-radicals that are
electrophilic species the same as their Ni(III) counterpart and
which is proposed to render the heterogeneous electron
transfer step rate-limiting. Other than slow electrode kinetics,
the pronounced lability of the Ni−thiyl-radical complex cation
compared with its amidyl congener is attributed to the more
diffuse character of S 3p orbitals. The presence of a node in the
radial function diminishes the overlap of a 3p orbital with the
more compact (node-less) 3d orbital of Ni.52 More generally,
the favorable kinetic stability of the unsupported Ni−amidyl
fragment is proposed to represent a key aspect that adds to the
prominent role of the 3d-metal−amido fragment in designing
luminescent base metal chromophore systems.23,28
Understanding the balance of the three contributions leads
us to conclude that rotational motion about the Ni−E bond
admixes Ni(III) character to the electronically excited state
populated by {Ni−ER}-to-{N-heterocycle−π*} charge transfer
and thereby promotes predominant nonradiative decay of
excited-state energy for this particular class of chromophores of
square-planar Ni(II). Further studies in glass matrices of
deuterated solvents are underway to address additional
contributions to excited-state quenching such as through
solvent coordination and coupling to overtones of high-energy
C−H stretching vibrations. As to intramolecular contributions,
future directions include setting coplanar geometric alignment
of the Ni−ER 3d−np π-bond with respect to the manifold of
π* orbitals at the N-heterocycle fragment.
EXPERIMENTAL SECTION
■
Manipulations of air- and moisture-sensitive compounds were carried
out under an atmosphere of dry argon using standard Schlenk or
glovebox techniques. Information on the purification of reagents and
solvents, and preparation of new proligands B/C0-OPiv is provided in
the Supporting Information. The following reagents and starting
materials were prepared according to reported procedures: 1,1′-
Bis(acetyl)ferrocenium bis(trifluoromethanesulfonyl)imide,53 [Fc″]-
Effect of Pendant Aryl Moiety: C- versus O-Metal-
ation. Substituting aryl for phenolate defines a setscrew for
modulating chromophore frontier MO energies in a
1172
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
NTf2, tris(4-bromophenyl)aminium tetrakis(3,5-bis(trifluoromethyl)-
51/52), 1.37 (br d, J = 6.8, 12H, H53/54/57/58), 1.34 (s, 9H, H20/
21/22), 1.02 (d, J = 6.8, 12H, H49/50/55/56) ppm. 13C NMR (151
MHz, C6D6, 25 °C): δ = 160.4 (C7), 147.8 (C2), 147.7 (C23), 147.6
(C16), 147.6 (2C, C24/28), 147.6 (C1), 147.6 (4C, C30/34/36/40),
147.4 (C4), 147.4 (2C, C32/38), 145.7 (C12), 144.1 (C8), 141.5
(C6), 141.1 (2C, C29/35), 137.0 (C14), 134.5 (C10), 131.2 (2C,
C25/27), 128.4 (C13; masked by solvent), 127.4 (C5), 126.0 (C17),
125.7 (C9), 125.5 (C18), 124.1 (C15), 123.4 (C26), 121.2 (br, 4C,
C31/33/37/39), 119.2 (C3), 117.4 (C11), 34.9 (C42/45), 34.7
(C19), 32.0 (3C, C20−22), 31.7 (4C, C41/43/44/46), 26.9 (4C,
C47/48/51/52 or C53/54/57/58), 24.6 (4C, C49/50/55/56), 24.5
(4C, C53/54/57/58 or C47/48/51/52) ppm.
phenyl)borate,54 [(4-BrC6H4)3N]BArF , 5-tert-butyl-2-methoxyphe-
4
n y l l i t h i u m , 2 7 N a N ( S i M e 3 ) 2 , 5 5 K C 7 H 7 , 5 6 2 , 6 -
(2,4,6-iPr3C6H2)2C6H3SH,57 Tip2ArSH, ANi-OPiv, A−O/ANi-NTMS2,
27
−
and [{A−ONi-thf}2(μ-Tf)]OTf [Tf = κ-O,O′−F3CSO2 , OTf =
−
F3CSO3 ].
Preparation of ANi-NC12H8. A toluene suspension of carbazole
(23 mg, 0.138 mmol) and red KC7H7 (18 mg, 0.14 mmol) was stirred
until only pale yellow KNC12H8 was observed. Toluene was removed
under vacuum, and ANi-OPiv (65 mg, 0.14 mmol) was added. The
solids were dissolved in 1,2-difluorobenzene (15 mL) to give a dark
yellow mixture that was stirred for 2 h. A dark yellow solution was
separated from KOPiv by cannula filtration and dried under vacuum
to afford an ochre solid. The solid was extracted with toluene (total of
35 mL), and the extract solutions were filtered. The combined
solutions were concentrated to ∼5 mL and stored at 5 °C overnight.
Dark yellow crystals were deposited, separated, washed with pentane,
and finally dried under vacuum. Yield: 46 mg (0.09 mmol, 62%).
Elemental analysis calcd. (%) for C34H27N3Ni: C, 76.15; H, 5.07; N,
7.84. Found: C, 75.97; H, 5.08; N, 7.81. Unit cell parameters: a =
12.5383(4), b = 17.1643(5), c = 19.9812(6) Å; α = 66.877(2), β =
Preparation of A−ONi-SArTip2. Addition of OEt2 (10 mL) to a
mixture of colorless Tip2ArSK, obtained as a solid from Tip2ArSH (97
mg, 0.19 mmol) and KC7H8 (25 mg, 0.19 mmol) in toluene (2 g)
solution, and pale green [{A−ONi-thf}2(μ-Tf)]OTf (110 mg, 0.19
mmol) afforded a dark red solution within a few minutes of stirring.
The solution was filtered after stirring for 1 h. The residual solid was
extracted with OEt2 (2 × 10 mL) until no more red product could be
removed, and the solvent was removed under vacuum. Dissolution of
the residual dark red solid in pentane (10 mL) afforded a dark red
solution from which crystals separated upon reducing the volume (3
mL) and cooling at −28 °C overnight. A pale red supernatant was
removed by cannula filtration, and the crystalline product was dried
under vacuum. Yield: 140 mg (0.16 mmol, 82%) of dark red
microcrystalline A−ONi-SArTip2. The microcrystals were dissolved in
C6H6 (∼1 mL) and crystallized through layer diffusion of (Me3Si)2O,
which deposited dark red prisms of A−ONi-SArTip2 × 0.5C6H6.
Crystals for XRD analysis were grown from layer diffusion of
(Me3Si)2O into a dark red solution in CDCl3. Elemental analysis
calcd. (%) for C58H68N2NiOS × 0.5C6H6: C, 78.03; H, 7.62; N, 2.98;
S, 3.41. Found: C, 77.99; H, 7.33; N, 3.01; S, 2.82. Unit cell
parameters: a = 37.0050(7), b = 37.0050(7), c = 14.5311(3) Å; α = β
= γ = 90°; V = 19898.5(9) Å3, Z = 16, I41/a. Deposition number
CCDC 2060408. UV−vis (THF, 24 °C): λ (ε/103/M−1 cm−1) = 267
(39.8; shoulder), 287 (28.9; shoulder), 306 (28.2), 319 (22;
shoulder), 376 (26), 479 (9.4), 522 (7.0; shoulder), 699 (0.2) nm.
1H NMR (600 MHz, CDCl3, 25 °C): δ = 8.31 (dd, J = 1.2, 5.4, 1H,
H16), 8.22 (d, J = 9.2, 1H, H11), 8.07 (d, J = 8.9, 1H, H10), 7.99 (dd,
J = 1.2, 8.1, 1H, H14), 7.71 (d, J = 2.4, 1H, H3), 7.48 (d, J = 8.6, 1H,
H17), 7.42 (d, J = 8.6, 1H, H18), 7.35 (t, J = 7.6, 1H, H26), 7.17 (d, J
= 7.5, 2H, H25/27), 7.12 (dd, J = 2.4, 8.9, 1H, H5), 7.00 (br s, 4H,
H31/33/37/39), 6.93 (dd, J = 5.4, 8.1, 1H, H15), 6.17 (d, J = 8.9,
1H, H6), 3.03 (br s, υ1/2 ∼ 60 Hz, 4H, H41/43/44/46), 2.73 (sep, J =
6.9, 2H, H42/45), 1.35 (s, 9H, H20/21/22), 1.02 (d, J = 7.1, 12H,
H48/51), 1.35 and 1.01 (br s each, 24H, H47/49/50/52) ppm. 13C
NMR (151 MHz, CDCl3, 25 °C): δ = 161.9 (C1), 151.1 (C7), 150.4
(C16), 146.9 (2C, C24/28), 146.7 (4C, C30/34/36/40), 146.2 (2C,
C32/38), 144.3 (C23), 143.1 (C12), 143.0 (C8), 140.3 (2C, C29/
35), 136.8 (C4/14), 133.9 (C10), 129.6 (C5), 129.5 (2C, C25/27),
127.1 (C13), 125.6 (C18), 125.0 (C9), 124.7 (C17), 124.0 (C6),
78.741(2), γ = 87.577(2)°; V = 3875.8(2) Å3, Z = 4, P1
̅
. Deposition
number CCDC 2060410. UV−vis (THF, 24 °C): λ (ε/103/M−1
cm−1) = 277 (58.7), 300 (28.8), 311 (27.6), 349 (16.4; shoulder),
364 (21.6), 385 (8.9; shoulder), 429 (7.8), 486 (3.8), 616 (1.0) nm.
1H NMR (600 MHz, C6D6, 25 °C): δ = 8.59 (m, 4H, H25/28/31/
34), 7.62 (ddd, J = 1.0, 6.9, 8.0, 2H, H27/33), 7.43 (ddd, J = 1.0, 6.9,
8.0, 2H, H26/32), 7.41 (d, J = 2.2, 1H, H3), 7.23 (d, J = 8.6, 1H,
H10), 7.18 (dd, J = 1.2, 4.9, 1H, H16), 6.98 (d, J = 8.5, 1H, H11),
6.90 (dd, J = 1.2, 8.1, 1H, H14), 6.86 (d, J = 8.8, 1H, H17), 6.68 (d, J
= 8.8, 1H, H18), 6.56 (dd, J = 2.2, 8.1, 1H, H5), 6.50 (d, J = 8.1, 1H,
H6), 5.99 (dd, J = 4.9, 8.1, 1H, H15), 1.23 (s, 9H, H20/21/22) ppm.
13C NMR (151 MHz, C6D6, 25 °C): δ = 164.3 (C7), 151.1 (2C,
C23/29), 150.5 (C16), 149.7 (C1), 148.4 (C4), 147.9 (C2), 146.4
(C12), 146.3 (C8), 141.1 (C6), 136.1 (C14), 135.7 (C10), 129.6
(C5), 128.7 (C13; masked by solvent), 127.7 (2C, C24/30), 125.8
(C15), 125.7 (C9), 125.6 (C17), 124.6 (2C, C27/33), 124.5 (C18),
120.8 (2C, C28/34 or C25/31), 120.1 (C3), 118.0 (C11), 116.3 (2C,
C26/32), 116.3 (2C, C25/31 or C28/34), 34.7 (C19), 31.8 (3C,
C20−22) ppm.
Preparation of ANi-SArTip2. Addition of OEt2 (10 mL) to a
mixture of colorless Tip2ArSK, obtained from Tip2ArSH (55 mg,
0.106 mmol) and KC7H8 (14 mg, 0.11 mmol) in toluene (2 g), and
brick-red ANi-OPiv (50 mg, 0.11 mmol) afforded a dark green
solution immediately upon stirring. The solvent was removed under
vacuum after stirring for 2.5 h. The residual green solid was extracted
successively with toluene, and the green extracts were filtered. No
crystallization was observed from either toluene or pentane but from
OEt2 (5 mL) within 7 days at −28 °C. Dark green microcrystals were
separated from a pale green supernatant by cannula filtration and
dried under vacuum. Yield: 55 mg (0.06 mmol, 59%). Single crystals
for XRD structure analysis were obtained from a C6D6 solution
through slow layer diffusion of (Me3Si)2O at room temperature,
affording dark green ANi-SArTip2 × 0.5C6D6. Elemental analysis
calcd. (%) for C58H68N2NiS: C, 78.81; H, 7.75; N, 3.17; S, 3.63.
Found: C, 78.83; H, 7.43; N, 3.17; S, 3.08. Unit cell parameters: a =
34.9672(11), b = 15.9198(5), c = 19.7110(6) Å; α = 90, β =
112.474(2), γ = 90°; V = 10139.2(6) Å3, Z = 8, C2/c. Deposition
number CCDC 2060409. UV−vis (THF, 24 °C): λ (ε/103/M−1
cm−1) = 306 (26.2), 314 (25.0; shoulder), 327 (19.8; shoulder), 368
(10.3; shoulder), 455 (6.5), 606 (3.3; shoulder), 636 (3.4) nm. 1,2-
C6H4F2 (22 °C): λ (ε/103/M−1 cm−1) = 310 (22.7), 366 (9.8;
122.9 (C26), 122.2 (C15), 121.4 (C11), 121.3 (C3), 120.2 (br, υ1/2
=
48 Hz, 4C, C31/33/37/39), 117.0 (C2), 34.0 (2C, C42/45), 34.0
(C19), 31.5 (3C, C20−22), 30.4 (br, υ1/2 = 60 Hz, 4C, C41/43/44/
46), 25.7 and 23.4 (br, 4C, C47/49/50/52), 24.0 (2C, C48/51) ppm.
Preparation of BNi-OPiv. A suspension of proligand B0-OPiv (90
mg, 0.17 mmol) in toluene (2.5 g) was transferred by pipet to a
toluene solution (2.5 g) of Ni(cod)2 (48 mg, 0.17 mmol) in the
glovebox, affording a dark green solution. Separation of a brown-red
crystalline precipitate was noted within 1 h at room temperature, and
stirring was continued for another 16 h after which the supernatant
was removed by cannula filtration, the residual crystalline mass
washed twice with pentane and dried under dynamic vacuum. Yield:
1
shoulder), 454 (5.7), 589 (3.1; shoulder), 632 (3.5) nm. H NMR
B
(600 MHz, C6D6, 25 °C): δ = 8.19 (dd, J = 1.3, 5.0, 1H, H16), 7.45
(d, J = 7.7, 2H, H25/27), 7.34 (d, J = 8.0, 1H, H6), 7.27 (d, J = 2.4,
1H, H3), 7.25 (t, J = 7.6, 1H, H26), 7.20 (br s, 4H, H31/33/37/39),
7.19 (dd, J = 1.5, 8.3, 1H, H14), 7.14 (d, J = 8.5, 1H, H10), 7.04 (dd,
J = 2.2, 8.0, 1H, H5), 6.92 (d, J = 8.5, 1H, H11), 6.77 (m, 2H, H17/
18), 6.54 (dd, J = 5.0, 8.1, 1H, H15), 3.48 (sep, J = 6.8, 4H, H41/43/
44/46), 2.68 (sep, J = 6.9, 2H, H42/45), 1.43 (br s, 12H, H47/48/
81 mg (0.14 mmol, 81%) of Ni-OPiv ×0.5C7H8 as reddish-brown,
hairlike crystals. Elemental analysis calcd (%) for C33H34N4NiO2 ×
0.5C7H8: C, 70.32; H, 6.14; N, 8.99. Found: C, 69.55; H, 6.24; N,
9.10. Unit cell parameters: a = 8.2710(6), b = 12.0751(8), c =
15.0435(11) Å; α = 83.787(4), β = 83.284(4), γ = 86.569(4)°; V =
1481.67(18) Å3, Z = 1, P1
̅
. Deposition number CCDC 2060407.
UV−vis (THF, 28 °C): λ (ε/103/M−1 cm−1) = 326 (27.3), 350 and
1173
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
365 (25.5), 431 (7.7), 446 (6.8; shoulder), 513 (3.1), 554 (1.2;
H15), 2.16 (s, 9H, H46/47/48), 1.53 (s, 9H, H49/50/51), 1.33 (s,
9H, H35/36/37), 1.26 (s, 6H, H29/30), 1.25 (s, 6H, H33/34), 0.77
(s, 6H, H31/32). 13C NMR (151 MHz, C6D6, 25 °C): δ = 165.8
(C38), 165.1 (C7), 158.3 (C22), 157.8 (C23), 150.7 (2C, C16/18),
150.2 (C1), 148.3 (2C, C4/17), 147.6 (C12), 146.7 (C2), 143.5
(C20), 142.8 (C19), 140.1 (C8), 139.6 (C41), 139.4 (C6), 136.0
(C43), 134.0 (C14), 132.6 (C10), 128.4 (C13; masked by solvent),
129.3 (C5), 126.3 (C15), 125.0 (C9), 124.5 (C39), 124.3 (C40),
123.3 (C21), 123.1 (2C, C24/42), 119.9 (C3), 118.3 (C11), 49.5
(C26), 48.7 (C27), 48.6 (C25), 36.7 (C45), 35.0 (C28), 34.9 (C44),
33.0 (3C, C49−51), 32.1 (3C, C35−37), 31.1 (3C, C46−48), 28.5
(2C, C33/34), 28.4 (2C, C29/30), 22.0 and 21.9 (2C, C31/32).
UV−vis (THF, 26 °C): λ (ε/103/M−1 cm−1) = 297 (62), 310 (49;
shoulder), 376 (50), 385 (46.5; shoulder), 472 (10.5), 513 (7.2;
shoulder), 606 (2.0; shoulder) nm.
shoulder) nm. IR (KBr, transmission): υ = 1612 (carbonyl stretch)
̅
cm−1.
Preparation of BNi-SArTip2. BNi-OPiv × 0.5C7H8 (60 mg, 96
μmol) was added to colorless solid KSArTip2 that was prepared from
HSArTip2 (53 mg, 100 μmol) and KC7H8 (14 mg, 100 μmol) in
toluene (2 g) solution. Addition of OEt2 (10 mL) to the solids affords
a dark green solution within few minutes of stirring. The solution was
filtered and the residual solids were extracted with OEt2 (2 × 10 mL)
until no more green product could be removed. The volume of the
combined OEt2 solution was reduced until crystallization occurred,
and the mixture was stored at −30 °C overnight. The supernatant was
removed at −30 °C, and the residual dark green crystalline solid was
dried under dynamic vacuum. Yield: 66 mg (67 μmol, 70%) dark
green crystalline solid. Elemental analysis calcd (%) for C64H74N4NiS:
C, 77.64; H, 7.53; N, 5.66; S, 3.24. Found: C, 77.64; H, 7.17; N, 5.64;
Preparation of CNi-SArTip2. Addition of OEt2 (15 mL) to a
mixture of colorless Tip2ArSK, obtained from Tip2ArSH (44 mg, 86
μmol) and KC7H8 (11 mg, 86 μmol) in toluene (4 g), and ochre ANi-
OPiv (60 mg, 86 μmol) afforded a red-purple solution upon stirring
for 0.5 h. The solvent was removed under vacuum after stirring for 5
h. The residual dark red solid was extracted successively with OEt2,
and the red-purple extracts were filtered, combined, and dried under
vacuum. The residual solid was dissolved in (Me3Si)2O (10 mL), and
the volume was slowly reduced to ∼5 mL, leading to the separation of
dark red crystals. Crystallization was completed at 4 °C overnight;
dark red microcrystals were separated from a pale purple supernatant
by cannula filtration and dried under vacuum. Yield: 68 mg (58 μmol,
68%) of purple CNi-SArTip2 × 1/3(Me3Si)2O. Dark red single
crystals for XRD analysis were separated from a C6D6/(Me3Si)2O
(0.5/3 mL) solution kept at −35 °C for 3 months. Elemental analysis
calcd (%) for C73H86N4NiS × 1/3C6H18OSi2: C, 77.36; H, 7.96; N,
4.81; S, 2.75. Found: C, 77.35; H, 7.59; N, 4.81; S, 2.63. Unit cell
parameters: a = 14.7177(6), b = 19.1313(8), c = 31.7880(14) Å; α =
99.367(2), β = 101.410(2), γ = 93.719(2)°; V = 8612.92 Å3, Z = 4,
1
S, 3.09. H NMR (600 MHz, C6D6): δ = 9.22 (d, J = 8.1, 1H, H16),
8.51 (dd, J = 1.1, 5.2, 1H, H14), 8.41 (d, J = 8.3, 1H, H10), 7.51 (d, J
= 7.7, 2H, H31/33), 7.43 (d, J = 8.0, 1H, H6), 7.35 (br s, 4H, H37/
39/43/45), 7.33 (t, J = 7.7, 1H, H32), 7.13 (d, J = 2.0, 1H, H3), 7.10
(dd, J = 7.9, 5.1, 1H, H15), 7.03 (dd, J = 2.2, 8.1, 1H, H5), 6.65 (d, J
= 8.5, 1H, H11), 3.62 (br sep, 4H, H47/48/50/51), 2.88 (m, 2H,
H25/26/27/28), 2.79 (br sep, 2H, H49/52), 2.55 (m, 2H, H25/26/
27/28), 1.55 (m, 4H, H25/26/27/28), 2.34−0.64 (very br s, 12H,
H53−56, H59−62), 1.48 (br s, 12H, H53−56, H59−62), 1.31 (s, 9H,
H22/23/24), 1.14 (br d, 12H, H57/58/63/64) ppm. 13C NMR (151
MHz, C6D6): δ = 160.9 (C7), 153.6 and 153.5 (C19/21), 149 (2C,
C30/34), 147.9 (C14), 147.7 (4C, C36/40/42/46), 147.6 (3C, C1
and C38/44), 147.3 (C4), 147.2 (C2), 146.6 (2C, C35/41), 146.5
(C18), 144.9 (C17), 141.8 (C6), 141.0 (C29), 137.2 (C12), 137.1
(C8), 136.3 (C16), 131.4 (2C, C31/33), 130.4 (C10), 128.8 (C15),
127.6 (C5), 127.1 (C13), 124.5 (C32), 123.3 (C9), 121.5 (4C, C37/
39/43/45), 119 (C3), 116.2 (C11), 35.0 (2C, C49/52), 34.7 (C21),
33.4 (C25), 32.4 (C26), 31.9 (3C, C22−24), 31.8 (4C, C47/48/50/
51), 26.8 and 25.0 (8C, C53−56, C59−62), 24.8 (4C, C57/58/63/
64), 23.4 and 23.2 (2C, C25/28) ppm. UV−vis (THF, 25 °C): λ (ε/
103/M−1 cm−1) = 282 (57.6), 301 (44; shoulder), 327 (39.8;
shoulder), 350 (35; shoulder), 371 (23.9; shoulder), 458 (11.1), 524
(4.4; shoulder), 623 (4.5; shoulder), 648 (4.6) nm. 1,2-C6H4F2 (22
°C): λ (ε/103/M−1 cm−1) = 288 (44.1; shoulder), 328 (34.5), 346
(29.5), 372 (18.5; shoulder), 454 (9.3), 589 (4.0; shoulder), 637
(4.6) nm.
P1
̅
. 1H NMR (600 MHz, C6D6, 25 °C): δ = 9.10 (dd, J = 1.3, 8.0, 1H,
H16), 8.76 (dd, J = 1.1, 5.0, 1H, H14), 8.67 (d, J = 8.3, 1H, H10),
8.15 (s, 1H, H25), 8.03 (s, 1H, H28), 7.49 (d, J = 7.6, 2H, H40/42),
7.36 (d, J = 8.0, 1H, H6), 7.30 (br s, 4H, H46/48/52/54), 7.30 (t, J =
7.6, 1H, H41), 7.26 (d, J = 2.1, 1H, H3), 7.05 (dd, J = 2.2, 8.2, 1H,
H5), 6.97 (m, 2H, H11 and H15), 3.58 (sep, J = 6.7, 4H, H56/58/
59/61), 2.74 (sep, J = 6.8, 2H, H57/60), 1.57 (br s, υ1/2 = 55 Hz,
12H, H68/69/72/73), 1.42 (br d, J = 7.0, 12H, H62/63/66/67), 1.29
(s, 9H, H20/21/22), 1.28 (s, 6H, H32/33), 1.23 (s, 6H, H36/37),
1.10 (d, J = 6.8, 12H, H64/65/70/71), 0.78 (s, 6H, H35/35), 0.12 (s,
6H, 1/3(Me3Si)2O) ppm. 13C NMR (151 MHz, C6D6, 25 °C): δ =
162.5 (C7), 157.7 (C27), 157.2 (C26), 149.7 (C14), 148.5 (C12),
148.4 (2C, C39/43), 147.8 (C1), 147.7 (4C, C45/49/51/55), 147.5
(4C, C2/4/47/53), 147.4 (C38), 147.1 (C8), 143.2 (C23), 142.8
(C24), 141.6 (C6), 141.0 (2C, C44/50), 139.7 (2C, C17/18), 134.7
(C16), 131.4 (2C, C40/42), 131.1 (C10), 128.2 (C13; masked by
solvent), 127.6 (C5), 127.1 (C15), 124.3 (C9), 123.9 (C41), 123.1
(C28), 122.9 (C25), 121.3 (4C, C46/48/52/54), 119.3 (C3), 117.6
(C11), 49.4 (C30), 48.4 (2C, C29/31), 34.9 (2C, C57/60), 34.6
(C19), 31.9 (3C, C20−22), 31.8 (4C, C56/58/59/61), 28.5 (2C,
C32/33), 28.3 (2C, C36/37), 26.8 (4C, C62/63/66/67), 24.8 (4C,
C68/69/72/73), 24.7 (4C, C64/65/70/71), 21.9 (2C, C35/35), 2.41
(1/3(Me3Si)2O) ppm. UV−vis (THF, 24 °C): λ (ε/103/M−1 cm−1) =
299 (70.1), 317 (57.1; shoulder), 380 (44.2; shoulder), 394 (51.4),
447 (7.6; shoulder), 500 (9.2), 654 (4.1) nm. 1,2-C6H4F2 (24 °C): λ
(ε/103/M−1 cm−1) = 302 (56.7), 323 (42.5), 380 (39.7), 394 (39.7),
448 (6.1), 488 (7.1), 645 (4.0) nm. n-pentane (24 °C): λ (ε/103/M−1
cm−1) = 236 (77.8), 280 (66.7), 291 (71.7), 303 (67.4), 380 (43.4),
396 (44.5), 451 (6.5), 512 (8.9), 677 (3.3) nm.
Preparation of CNi-OPiv. The procedure is analogous to that
B
described above for Ni-OPiv, except for the following changes: C0-
OPiv, 120 mg (0.19 mmol); Ni(cod)2, 52 mg (0.19 mmol), toluene
(6.2 g). Yield: 113 mg (0.16 mmol, 86%) of hairlike ochre crystals.
Elemental analysis calcd (%) for C42H46N4NiO2: C, 72.32; H, 6.65;
N, 8.03. Found: C, 71.94; H, 6.70; N, 8.03. UV−vis (THF, 28 °C): λ
(ε/103/M−1 cm−1) = 295 (61.5), 310 (45.5; shoulder), 375 (48.7),
385 (42.4; shoulder), 468 (9.5), 510 (6.5; shoulder), 560 (2.2;
shoulder) nm. IR (KBr, transmission): υ = 1612 (carbonyl stretch)
̅
cm−1.
Preparation of CNi-OPhtBu2. CNi-OPiv, prepared as described
above from C0-OPiv (50 mg; 78 μmol) and Ni(cod)2 (22 mg; 79
μmol), was combined with KOPhtBu2, prepared from 2,4-di-tert-
butylphenol (16 mg, 78 μmol) and KC7H7 (10 mg, 78 μmol) in
toluene (7 g), in 1,2-difluorobenzene (10 mL) to afford a dark red
solution within a few minutes of stirring. Solvent was removed after 1
h. The residual solid was washed with pentane (40 mL), dissolved in
1,2-difluorobenzene, and crystallized by slow layer diffusion of
(Me3Si)2O. Reddish hairlike crystals separated within days and were
separated by cannula filtration, washed with pentane, and dried under
vacuum. Yield: 24 mg (30 μmol, 38%). Elemental analysis calcd (%)
for C51H58N4NiO: C, 76.40; H, 7.29; N, 6.99. Found: C, 74.22; H,
7.24; N, 6.86. 1H NMR (600 MHz, C6D6, 25 °C): δ = 9.04 (d, J = 8.4,
1H, H10), 8.91 (dd, J = 1.3, 8.0, 1H, H14), 8.45 (d, J = 8.2, 1H,
H39), 8.34 (dd, J = 1.5, 5.0, 1H, H16), 8.26 (s, 1H, H21), 8.14 (s,
1H, H24), 7.76 (d, J = 2.6, 1H, H42), 7.47 (d, J = 8.0, 1H, H6), 7.46
(d, J = 2.0, 1H, H3), 7.45 (dd, J = 2.5, 8.5, 1H, H40), 7.08 (d, J = 8.5,
1H, H11), 6.97 (dd, J = 2.1, 8.0, 1H, H5), 6.61 (dd, J = 4.9, 8.2, 1H,
Preparation of CNi-NTMS2. To a suspension of CNi-OPiv in
toluene (10 g), prepared as described above from C0-OPiv (75 mg;
0.12 mmol) and Ni(cod)2 (33 mg; 0.12 mmol), was added NaNTMS2
(22 mg, 0.12 mmol) dissolved in thf (8 g). After stirring for 3 h,
solvents were removed under vacuum, and the solid residue washed
with pentane (50 mL), extracted with hexanes (3 × 75 mL). The
extracts were filtered, combined, concentrated until a crystalline solid
1174
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
separated, and stored at −40 °C to deposit brick red microcrystals.
The supernatant was removed by cannula filtration, and the
microcrystalline product was dried under vacuum. Yield: 10 mg
(0.01 mmol, 11%). 1H NMR (600 MHz, C6D6, 25 °C): δ = 9.12 (d, J
= 7.6, 1H, H14), 9.05 (dd, J = 1.2, 4.9, 1H, H16), 9.02 (d, J = 8.3, 1H,
H10), 8.24 (s, 1H, H21), 8.21 (s, 1H, H24), 8.10 (d, J = 7.9, 1H,
H6), 7.49 (d, J = 1.8, 1H, H3), 7.47 (dd, J = 2.2, 8.1, 1H, H5), 7.18
(d, J = 8.7, 1H, H11), 6.97 (dd, J = 4.7, 8.1, 1H, H15), 1.38 (s, 9H,
H35/36/37), 1.25 (s, 6H, H29/30), 1.24 (s, 6H, H33/34), 0.85 (s,
JH−Si = 6.1, 18H, (Me3Si)2N), 0.75 (s, 6H, H31/32). 13C NMR (151
MHz, C6D6, 25 °C): δ = 165.5 (C7), 158.2 (C22), 157.8 (C23),
151.2 (C16), 148.5 (C8 or C12), 148.4 (C12 or C8), 148.3 (C1),
148.0 (C4), 147.5 (C2), 143.5 (C20), 142.9 (2C, C6/19), 140.5
(C17), 139.9 (C18), 134.5 (C14), 132.5 (C10), 128.4 (2C, C5/13;
masked by solvent), 126.2 (C15), 125.3 (C9), 123.2 (C24), 123.1
(C21), 120.3 (C3), 118.4 (C11), 49.5 (C26), 48.7 (C25 or C27),
48.6 (C27 or C25), 35.0 (C28), 32.1 (3C, C35−37), 28.5 (2C, C29
or C30, C33 or C34), 28.4 (2C, C30 or C29, C34 or C33), 22.0 (2C,
C31/32), 7.81 (6C, J = 53.6, (Me3Si)2N). UV−vis (THF, 24 °C): λ
(ε/103/M−1 cm−1) = 292 (89.2), 315 (67.1; shoulder), 358 (46.1;
shoulder), 379 (67.1), 389 (68.5), 492 (12.6), 543 (9.0; shoulder)
nm.
gStrainx,y,z = 0.015, 0.006, and 0.003; peak-to-peak line width lwpp
= 1.0 + 0 mT (Gaussian + Lorentzian); Ax,y,z(14N, nat. abundance) =
0, 0, 0 MHz; rel. abundance = 1.5; mw freq = 9.30774 GHz; range =
310−340 mT. Simulation parameters, 213 K, [(κ3-N,N′,O•)Ni−
NTMS2]BArF : giso = 2.029; gisoStrain = 0.001; peak-to-peak line
4
width lwpp = 0.9 + 0 mT (Gaussian + Lorentzian); Aiso(14N; nat.
abundance) = 0 MHz; [(κ3-N,N′,O)Ni(•NTMS2)]BArF : giso = 2.033;
4
gStrainiso = 0.001; peak-to-peak lwpp = 0.9 + 0 mT (Gaussian +
Lorentzian); Aiso(14N; nat. abundance) = 0 MHz; rel abundances 1.0/
1.0; mw freq = 9.294284 GHz; range = 310−340 mT.
A
In Situ Oxidation of Ni-SArTip2. The procedure was the same
A
A
as described for Ni-NTMS2, except for the following changes: Ni-
SArTip2 (3.5 mg, 4.0 μmol) in toluene (1.74 g), [(4-BrC6H4)3N]-
BArF (5.4 mg, 4.0 μmol) in CH2Cl2 (2.0 mL). Figure S11 provides
4
details for X-band cw-EPR spectrum at 100 K. Simulation parameters
for major component: gx,y,z = 2.272, 2.099, and 2.052; gStrainx,y,z
=
0.017, 0.018, and 0.009; peak-to-peak line width lwpp = 2.0 + 0 mT
(Gaussian + Lorentzian); mw freq. = 9.312679 GHz; range = 270−
350 mT.
Computational Details. Electronic structure calculations were
carried out through DFT methods using the ORCA program
package.58 For all optimizations, triple-ξ-valence TZVP59 basis sets
were used with the generalized gradient approximated functional
BP86.60 Pertinent metrical data is arranged in Tables S1−S3.
Molecular orbitals and electronic properties were extracted from
single-point calculations in the optimized geometries with the global
hybrid functional TPSSh61 and triple-ξ-valence TZVP basis sets.
Grimme’s third-generation D3 correction of dispersion was
applied,62,63 and medium effects were approximated in a dielectric
continuum approach (COSMO), parametrized for THF.64 Cartesian
coordinates of the DFT optimized structures are assembled in form of
a separate .xyz file (Supporting Information); frontier molecular
orbital landscapes are shown in Figures 6−8. For each complex, the
60−80 lowest optical electronic transitions were assessed with ORCA-
implemented TD-DFT methods within the Tamm−Dancoff approx-
imation.
In situ Oxidation of ANi-NTMS2. To a red solution of ANi-
NTMS2 (2 mg, 4 μmol) in toluene (1.74 g) in a 25 mL septum
capped Schlenk flask cooled at −78 °C (acetone/dry ice) was added a
blue solution of [(4-BrC6H4)3N]BArF (5 mg, 4 μmol) at ≤ −40 °C
4
in CH2Cl2 (2.0 mL; in Ar supplied 4 mL screw cap vial with septum)
to afford an emerald green solution instantaneously upon mixing. A
∼0.3 mL portion of this solution was transferred through a precooled
double-tipped cannula into a septum-capped EPR sample tube at −78
°C and was immediately frozen at 77 K for X-band cw-EPR
characterization. Figures S6 and S7 provide details for spectra
recorded at 100 and 213 K, respectively. Simulation parameters, 100
K: gx,y,z = 2.110, 2.068, and 2.0279; gStrainx,y,z = 0.018, 0.010, and
0.0098; Ax,y,z(14N, nat. abundance) = 0, 53, 0 mT; peak-to-peak line
width lwpp = 1.0 + 0 mT (Gaussian + Lorentzian), mw freq =
9.31026 GHz, range = 290−350 G. Simulation parameters, 213 K: giso
= 2.069; gStrainiso = 0.013; A(14N, nat. abundance) = 18 mT; peak-to-
peak line-width lwpp = 2.1 + 1.0 mT (Gaussian + Lorentzian); mw
freq = 9.30053 GHz; range = 300−340 mT.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.1c00121.
■
sı
Oxidation of ANi-NTMS2 (4.8 mg, 9.1 μmol) with [Fc″]NTf2 (4.8
mg, 8.7 μmol) followed above procedure. Simulation parameters for
ANi(NTMS2)(κ−O-N(SO2CF3)2) (Figure S8): gx,y = 2.434, 2.307, gz
masked by primary product; gStrainx,y = 0.029 and 0.017; peak-to-
peak line width lwpp = 1.5 + 0 mT (Gaussian + Lorentzian); rel.
abundance 0.13; mw freq = 9.311058 GHz; range = 260−350 G.
Vis−NIR spectroscopic monitoring the oxidation of ANi-NTMS2
was carried out as follows. A quartz glass cuvette (1 cm) was charged
with CH2Cl2 (2.6 mL), and reference spectra were recorded at T =
20, −20, −40, and −60 °C. A 50 μL aliquot of a stock solution of ANi-
NTMS2 (5 mg, 9 μmol) in CH2Cl2 (1.0 mL) was added, and spectra
were recorded at the above temperatures for fitting the temperature
dependence of sample concentration. A 83 μL aliquot of a stock
solution of [(4-BrC6H4)3N]BArF4 (2.5 mg, 1.9 μmol) in CH2Cl2 (2.0
mL), equal to 0.33 equiv of oxidant, was added in three successive
steps at −60 °C and spectra were recorded each time. Finally, excess 5
μL of oxidant stock solution were added. Spectra were corrected for
blank solvent background and are plotted in Figure S5; vis bands
(−60 °C): λ (ε/103/M−1 cm−1) = 446 (2.5), 504 (2.7), 637 (2.2) nm.
In Situ Oxidation of A−ONi-NTMS2. The procedure was the same
as described for ANi-NTMS2, except for the following changes:
A−ONi-NTMS2 (3 mg, 5 μmol) in toluene (2.2 g), [(4-BrC6H4)3N]-
General information on preparation/purification of
proligands C/B0-OPiv, reagents, solvents, additional
UV/vis and IR absorption, and EPR spectroscopic,
XRD and electrochemical data (PDF)
DFT optimized molecular structure data (XYZ)
Accession Codes
CCDC 2060407−2060410 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
AUTHOR INFORMATION
Corresponding Author
■
Andreas Berkefeld − Institut fu
Eberhard Karls Universität Tu
Germany; orcid.org/0000-0002-2757-7841;
̈
r Anorganische Chemie,
bingen, 72076 Tubingen,
̈
̈
BArF (7 mg, 5 μmol) in CH2Cl2 (2.0 mL). Figures S9 and S10
4
Email: andreas.berkefeld@uni-tuebingen.de
provide details for X-band cw-EPR spectra at 100 and 213 K,
respectively. Simulation parameters, 100 K: Minor component [(κ3-
Authors
Riyadh Alrefai − Institut fu
Karls Universität Tubingen, 72076 Tu
N,N′,O•)Ni−NTMS2]BArF , gx,y,z = 2.046, 2.026, and 2.015;
4
̈
r Anorganische Chemie, Eberhard
bingen, Germany
gStrainx,y,z = 0.005, 0.005, and 0.003; peak-to-peak line width lwpp
= 1.0 + 0 mT (Gaussian + Lorentzian); Ax,y,z(14N, nat. abundance) =
0, 0, 0 MHz; rel. abundance = 1.0; major component [(κ3-
̈
̈
Gerald Hörner − Anorganische Chemie IV, Universität
Bayreuth, 95440 Bayreuth, Germany
N,N′,O)Ni(•NTMS2)]BArF : gx,y,z = 2.057, 2.043, and 2.002;
4
1175
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
(14) Hockin, B. M.; Li, C.; Robertson, N.; Zysman-Colman, E.
Photoredox catalysts based on earth-abundant metal complexes.
Catal. Sci. Technol. 2019, 9, 889−915.
Hartmut Schubert − Institut fu
̈
r Anorganische Chemie,
bingen, 72076 Tubingen,
Eberhard Karls Universität Tu
Germany
̈
̈
(15) Kjær, K. S.; Kaul, N.; Prakash, O.; Chábera, P.; Rosemann, N.
W.; Honarfar, A.; Gordivska, O.; Fredin, L. A.; Bergquist, K.-E.;
Häggström, L.; Ericsson, T.; Lindh, L.; Yartsev, A.; Styring, S.; Huang,
P.; Uhlig, J.; Bendix, J.; Strand, D.; Sundström, V.; Persson, P.;
Lomoth, R.; Wärnmark, K. Luminescence and reactivity of a charge-
transfer excited iron complex with nanosecond lifetime. Science 2019,
363, 249.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.1c00121
Notes
The authors declare no competing financial interest.
(16) Williams, J. A. G. The coordination chemistry of
dipyridylbenzene: N-deficient terpyridine or panacea for brightly
luminescent metal complexes? Chem. Soc. Rev. 2009, 38, 1783−1801.
(17) Dixon, I. M.; Alary, F.; Boggio-Pasqua, M.; Heully, J.-L.
Reversing the relative 3MLCT−3MC order in Fe(ii) complexes using
cyclometallating ligands: a computational study aiming at luminescent
Fe(ii) complexes. Dalton Trans. 2015, 44, 13498−13503.
(18) Mukherjee, S.; Torres, D. E.; Jakubikova, E. HOMO inversion
as a strategy for improving the light-absorption properties of Fe(ii)
chromophores. Chem. Sci. 2017, 8, 8115−8126.
(19) Braun, J. D.; Lozada, I. B.; Kolodziej, C.; Burda, C.; Newman,
K. M. E.; van Lierop, J.; Davis, R. L.; Herbert, D. E. Iron(ii)
coordination complexes with panchromatic absorption and nano-
second charge-transfer excited state lifetimes. Nat. Chem. 2019, 11,
1144−1150.
ACKNOWLEDGMENTS
■
A.B. acknowledges Dr. Wolfgang Leis, Universität Tu
̈
bingen,
for recording emission spectroscopic data, Deutsche For-
schungsgemeinschaft (DFG; German Research Foundation)
for financial support of this work (grant number BE4424/7-1),
and SCIOM-Syn GmbH & Co. KG, Schonungen, Germany,
for the generous donation of 1,1,2,2,3,3-hexamethylindane-5,6-
diamine. The CryoProbe Prodigy BBO NMR probe head used
in this work was purchased with the support of an
instrumentation program by the DFG, grant number INST
37/929-1 FUGG. A.B. dedicates this work to Prof. Dr.
Wolfgang Kaim on the occasion of his 70th birthday.
̈
Herzlichen Gluckwunsch!
(20) Braun, J. D.; Lozada, I. B.; Herbert, D. E. In Pursuit of
Panchromatic Absorption in Metal Coordination Complexes:
Experimental Delineation of the HOMO Inversion Model Using
Pseudo-Octahedral Complexes of Diarylamido Ligands. Inorg. Chem.
2020, 59, 17746−17757.
(21) Sandleben, A.; Vogt, N.; Hörner, G.; Klein, A. Redox Series of
Cyclometalated Nickel Complexes [Ni((R)Ph(R′)bpy)Br]+/0/−/2−
(H−(R)Ph(R′)bpy = Substituted 6-Phenyl-2,2′-bipyridine). Organo-
metallics 2018, 37, 3332−3341.
(22) Kletsch, L.; Hörner, G.; Klein, A. Cyclometalated Ni(II)
Complexes [Ni(N∧C∧N)X] of the Tridentate 2,6-di(2-pyridyl)phen-
ide Ligand. Organometallics 2020, 39, 2820−2829.
(23) Wong, Y.-S.; Tang, M.-C.; Ng, M.; Yam, V. W.-W. Toward the
Design of Phosphorescent Emitters of Cyclometalated Earth-
Abundant Nickel(II) and Their Supramolecular Study. J. Am. Chem.
Soc. 2020, 142, 7638−7646.
(24) Sun, R.; Qin, Y.; Ruccolo, S.; Schnedermann, C.; Costentin, C.;
Nocera, D. G. Elucidation of a Redox-Mediated Reaction Cycle for
Nickel-Catalyzed Cross Coupling. J. Am. Chem. Soc. 2019, 141, 89−
93.
(25) Ting, S. I.; Garakyaraghi, S.; Taliaferro, C. M.; Shields, B. J.;
Scholes, G. D.; Castellano, F. N.; Doyle, A. G. 3d-d Excited States of
Ni(II) Complexes Relevant to Photoredox Catalysis: Spectroscopic
Identification and Mechanistic Implications. J. Am. Chem. Soc. 2020,
142, 5800−5810.
(26) Wenger, O. S. Photoactive Nickel Complexes in Cross-
Coupling Catalysis. Chem. - Eur. J. 2021, 27, 2270−2278.
(27) Berkefeld, A.; Fröhlich, M.; Kordan, M.; Hörner, G.; Schubert,
H. Selective metalation of phenol-type proligands for preparative
organometallic chemistry. Chem. Commun. 2020, 56, 3987−3990.
(28) Creutz, S. E.; Lotito, K. J.; Fu, G. C.; Peters, J. C. Photoinduced
Ullmann C−N Coupling: Demonstrating the Viability of a Radical
Pathway. Science 2012, 338, 647.
REFERENCES
■
(1) Tang, C. W.; VanSlyke, S. A. Organic electroluminescent diodes.
Appl. Phys. Lett. 1987, 51, 913−915.
(2) Tang, C. W.; VanSlyke, S. A.; Chen, C. H. Electroluminescence
of doped organic thin films. J. Appl. Phys. 1989, 65, 3610−3616.
(3) Yersin, H.; Rausch, A. F.; Czerwieniec, R.; Hofbeck, T.; Fischer,
T. The triplet state of organo-transition metal compounds. Triplet
harvesting and singlet harvesting for efficient OLEDs. Coord. Chem.
Rev. 2011, 255, 2622−2652.
(4) Shaw, M. H.; Twilton, J.; MacMillan, D. W. C. Photoredox
Catalysis in Organic Chemistry. J. Org. Chem. 2016, 81, 6898−6926.
(5) Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Visible Light
Photoredox Catalysis with Transition Metal Complexes: Applications
in Organic Synthesis. Chem. Rev. 2013, 113, 5322−5363.
(6) Milligan, J. A.; Phelan, J. P.; Badir, S. O.; Molander, G. A. Alkyl
Carbon−Carbon Bond Formation by Nickel/Photoredox Cross-
Coupling. Angew. Chem., Int. Ed. 2019, 58, 6152−6163.
(7) Welin, E. R.; Le, C.; Arias-Rotondo, D. M.; McCusker, J. K.;
MacMillan, D. W. C. Photosensitized, energy transfer-mediated
organometallic catalysis through electronically excited nickel(II).
Science 2017, 355, 380.
(8) Haque, A.; Xu, L.; Al-Balushi, R. A.; Al-Suti, M. K.; Ilmi, R.;
Guo, Z.; Khan, M. S.; Wong, W.-Y.; Raithby, P. R. Cyclometallated
tridentate platinum(ii) arylacetylide complexes: old wine in new
bottles. Chem. Soc. Rev. 2019, 48, 5547−5563.
(9) Ibrahim-Ouali, M.; Dumur, F. Recent Advances on Metal-Based
Near-Infrared and Infrared Emitting OLEDs. Molecules 2019, 24,
1412.
(10) Cebrián, C.; Mauro, M. Recent advances in phosphorescent
platinum complexes for organic light-emitting diodes. Beilstein J. Org.
Chem. 2018, 14, 1459−1481.
(11) Longhi, E.; De Cola, L. Iridium(III) Complexes for OLED
Application. Iridium(III) in Optoelectronic and Photonics Applications
2017, 205−274.
(29) McGuire, R.; Wilson, M. H.; Nash, J. J.; Fanwick, P. E.;
McMillin, D. R. Lewis Acid and Base Sensing by Platinum(II)
Polypyridines. Inorg. Chem. 2008, 47, 2946−2948.
(30) Nicholson, R. S.; Shain, I. Theory of Stationary Electrode
Polarography. Single Scan and Cyclic Methods Applied to Reversible,
Irreversible, and Kinetic Systems. Anal. Chem. 1964, 36, 706−723.
(31) Stoll, S.; Schweiger, A. EasySpin, a comprehensive software
package for spectral simulation and analysis in EPR. J. Magn. Reson.
2006, 178, 42−55.
(12) Dorn, M.; Kalmbach, J.; Boden, P.; Päpcke, A.; Gómez, S.;
Förster, C.; Kuczelinis, F.; Carrella, L. M.; Buldt, L. A.; Bings, N. H.;
̈
Rentschler, E.; Lochbrunner, S.; González, L.; Gerhards, M.; Seitz,
M.; Heinze, K. A Vanadium(III) Complex with Blue and NIR-II Spin-
Flip Luminescence in Solution. J. Am. Chem. Soc. 2020, 142, 7947−
7955.
(13) Wenger, O. S. Photoactive Complexes with Earth-Abundant
Metals. J. Am. Chem. Soc. 2018, 140, 13522−13533.
1176
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177

Organometallics
pubs.acs.org/Organometallics
Article
(32) Bu
̈
ttner, T.; Geier, J.; Frison, G.; Harmer, J.; Calle, C.;
tzmacher, H. A Stable Aminyl
(51) Roberts, B. P. Polarity-reversal catalysis of hydrogen-atom
abstraction reactions: concepts and applications in organic chemistry.
Chem. Soc. Rev. 1999, 28, 25−35.
Schweiger, A.; Schönberg, H.; Gru
̈
Radical Metal Complex. Science 2005, 307, 235.
(33) Collins, T. J.; Nichols, T. R.; Uffelman, E. S. A square-planar
nickel(III) complex of an innocent ligand system. J. Am. Chem. Soc.
1991, 113, 4708−4709.
(52) Kaupp, M. The role of radial nodes of atomic orbitals for
chemical bonding and the periodic table. J. Comput. Chem. 2007, 28,
320−325.
(53) Mews, N. M.; Berkefeld, A.; Hörner, G.; Schubert, H.
Controlling Near-Infrared Chromophore Electronic Properties
through Metal−Ligand Orbital Alignment. J. Am. Chem. Soc. 2017,
139, 2808−2815.
(34) Hanss, J.; Kruger, H.-J. First Isolation and Structural
̈
Characterization of a Nickel(III) Complex Containing Aliphatic
Thiolate Donors. Angew. Chem., Int. Ed. 1998, 37, 360−363.
(35) Jacobs, S. A.; Margerum, D. W. Solution properties of
bis(dipeptide)nickelate(III) complexes and kinetics of their decom-
position in acid. Inorg. Chem. 1984, 23, 1195−1201.
(54) Rosokha, S. V.; Stern, C. L.; Ritzert, J. T. π-Bonded molecular
wires: self-assembly of mixed-valence cation-radical stacks within the
nanochannels formed by inert tetrakis[3,5-bis(trifluoromethyl)-
phenyl]borate anions. CrystEngComm 2013, 15, 10638−10647.
(36) Thomas, F. Ligand-centred oxidative chemistry in sterically
hindered salen complexes: an interesting case with nickel. Dalton
Trans. 2016, 45, 10866−10877.
(37) Kurz, H.; Schötz, K.; Papadopoulos, I.; Heinemann, F. W.;
Maid, H.; Guldi, D. M.; Köhler, A.; Hörner, G.; Weber, B. A
Fluorescence-Detected Coordination-Induced Spin State Switch. J.
Am. Chem. Soc. 2021, 143, 3466−3480.
(55) Wannagat, U.; Niederprum, H. Beitrage zur Chemie der
̈
̈
Silicium-Stickstoff-Verbindungen, XIII. Silylsubstituierte Alkaliamide.
Chem. Ber. 1961, 94, 1540−1547.
(56) Bailey, P. J.; Coxall, R. A.; Dick, C. M.; Fabre, S.; Henderson, L.
C.; Herber, C.; Liddle, S. T.; Lorono-González, D.; Parkin, A.;
̃
Parsons, S. The First Structural Characterisation of a Group 2 Metal
Alkylperoxide Complex: Comments on the Cleavage of Dioxygen by
Magnesium Alkyl Complexes. Chem. - Eur. J. 2003, 9, 4820−4828.
(57) Niemeyer, M.; Power, P. P. Donor-Free Alkali Metal Thiolates:
Synthesis and Structure of Dimeric, Trimeric, and Tetrameric
Complexes with Sterically Encumbered Terphenyl Substituents.
Inorg. Chem. 1996, 35, 7264−7272.
(38) Mandapati, P.; Giesbrecht, P. K.; Davis, R. L.; Herbert, D. E.
Phenanthridine-Containing Pincer-like Amido Complexes of Nickel,
Palladium, and Platinum. Inorg. Chem. 2017, 56, 3674−3685.
(39) Klein, A.; Hurkes, N.; Kaiser, A.; Wielandt, W. π-Stacking
Modulates the Luminescence of [(dppz)Ni(Mes)Br] (dppz =
dipyrido[3,2-a:2′,3′-c]phenazine, Mes = 2,4,6-trimethylphenyl). Z.
Anorg. Allg. Chem. 2007, 633, 1659−1665.
(58) Neese, F. The ORCA program system. Wiley Interdiscip. Rev.:
Comput. Mol. Sci. 2012, 2, 73−78.
(40) Lazzaro, D. P.; McGuire, R.; McMillin, D. R. Regiospecific
Quenching of a Photoexcited Platinum(II) Complex at Acidic and
Basic Sites. Inorg. Chem. 2011, 50, 4437−4444.
(59) Schäfer, A.; Horn, H.; Ahlrichs, R. Fully optimized contracted
Gaussian basis sets for atoms Li to Kr. J. Chem. Phys. 1992, 97, 2571−
2577.
(60) Becke, A. D. Density-functional exchange-energy approxima-
tion with correct asymptotic behavior. Phys. Rev. A: At., Mol., Opt.
Phys. 1988, 38, 3098−3100.
(61) Staroverov, V. N.; Scuseria, G. E.; Tao, J.; Perdew, J. P.
Comparative assessment of a new nonempirical density functional:
Molecules and hydrogen-bonded complexes. J. Chem. Phys. 2003, 119,
12129−12137.
(62) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and
accurate ab initio parametrization of density functional dispersion
correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010,
132, 154104.
(41) Kaim, W. Electron transfer to 1,10-phenanthroline by grignard
reagents. the spin distribution in organomagnesium radical complexes.
J. Organomet. Chem. 1981, 222, c17−c20.
(42) Kaim, W. Electron transfer to complex ligands. Radical anions
and organomagnesium radical complexes of 2,2′-bipyridines and 1,10-
phenanthrolines. J. Am. Chem. Soc. 1982, 104, 3833−3837.
(43) Fees, J.; Kaim, W.; Moscherosch, M.; Matheis, W.; Klima, J.;
Krejcik, M.; Zalis, S. Electronic structure of the ″molecular light
switch″ bis(bipyridine)dipyrido[3,2-a:2′,3′-c]phenazineruthenium-
(2+). Cyclic voltammetric, UV/visible and EPR/ENDOR study of
multiply reduced complexes and ligands. Inorg. Chem. 1993, 32, 166−
174.
(44) Schoonover, J. R.; Bates, W. D.; Meyer, T. J. Application of
Resonance Raman Spectroscopy to Electronic Structure in Metal
Complex Excited States. Excited-State Ordering and Electron
Delocalization in Dipyrido[3,2-a:2′,3′-c]phenazine (dppz): Com-
plexes of Re(I) and Ru(II). Inorg. Chem. 1995, 34, 6421−6422.
(45) Butsch, K.; Gust, R.; Klein, A.; Ott, I.; Romanski, M. Tuning
the electronic properties of dppz-ligands and their palladium(ii)
complexes. Dalton Trans. 2010, 39, 4331−4340.
(63) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping
function in dispersion corrected density functional theory. J. Comput.
Chem. 2011, 32, 1456−1465.
(64) Klamt, A.; Schuurmann, G. COSMO: a new approach to
̈ ̈
dielectric screening in solvents with explicit expressions for the
screening energy and its gradient. J. Chem. Soc., Perkin Trans. 2 1993,
799−805.
(46) Delaney, S.; Pascaly, M.; Bhattacharya, P. K.; Han, K.; Barton, J.
K. Oxidative Damage by Ruthenium Complexes Containing the
Dipyridophenazine Ligand or Its Derivatives: A Focus on
Intercalation. Inorg. Chem. 2002, 41, 1966−1974.
̌
(47) Wang, C.; Otto, S.; Dorn, M.; Kreidt, E.; Lebon, J.; Srsan, L.;
Di Martino-Fumo, P.; Gerhards, M.; Resch-Genger, U.; Seitz, M.;
Heinze, K. Deuterated Molecular Ruby with Record Luminescence
Quantum Yield. Angew. Chem., Int. Ed. 2018, 57, 1112−1116.
(48) Nguyen, T.; Panda, A.; Olmstead, M. M.; Richards, A. F.;
Stender, M.; Brynda, M.; Power, P. P. Synthesis and Characterization
of Quasi-Two-Coordinate Transition Metal Dithiolates M(SAr*)2 (M
= Cr, Mn, Fe, Co, Ni, Zn; Ar* = C6H3-2,6(C6H2-2,4,6-iPr3)2. J. Am.
Chem. Soc. 2005, 127, 8545−8552.
(49) Adhikari, D.; Mossin, S.; Basuli, F.; Huffman, J. C.; Szilagyi, R.
K.; Meyer, K.; Mindiola, D. J. Structural, Spectroscopic, and
Theoretical Elucidation of a Redox-Active Pincer-Type Ancillary
Applied in Catalysis. J. Am. Chem. Soc. 2008, 130, 3676−3682.
(50) Mankad, N. P.; Antholine, W. E.; Szilagyi, R. K.; Peters, J. C.
Three-Coordinate Copper(I) Amido and Aminyl Radical Complexes.
J. Am. Chem. Soc. 2009, 131, 3878−3880.
1177
https://doi.org/10.1021/acs.organomet.1c00121
Organometallics 2021, 40, 1163−1177