The Influence of Redox-Innocent Donor Groups in Tetradentate Ligands Derived from o-Phenylenediamine: Electronic Structure Investigations with Nickel

Article

pubs.acs.org/IC

Cite This: Inorg. Chem. XXXX, XXX, XXX −XXX

The Inﬂuence of Redox-Innocent Donor Groups in Tetradentate

Ligands Derived from o‑Phenylenediamine: Electronic Structure

Investigations with Nickel

Kyle D. Spielvogel,^†Ezra J. Coughlin,^§Hayley Petras,^†Javier A. Luna,^†Austin Benson,^†

Courtney M. Donahue,^†Amani Kibasa,^†Kyounghoon Lee,^†Ryan Salacinski,^†Suzanne C. Bart,^§

Scott K. Shaw,*^,†James J. Shepherd,*^,†and Scott R. Daly*^,†

^†Department of Chemistry, The University of Iowa, E331 Chemistry Building, Iowa City, Iowa 52242-1294, United States

^§H.C. Brown Laboratory, Department of Chemistry, Purdue University, West Lafayette, Indiana 47907, United States

S

* Supporting Information

ABSTRACT: The continued development of redox-active ligands requires an understanding as to how ligand modiﬁcations

and related factors aﬀect the locus of redox activity and spin density in metal complexes. Here we describe the synthesis,

characterization, and electronic structure of nickel complexes containing triaryl NNNN (1) and SNNS (2) ligands derived from

o-phenylenediamine. The tetradentate ligands in 1 and 2 were investigated and compared to those in metal complexes with

compositionally similar ligands to determine how ligand-centered redox properties change when redox-active ﬂanking groups

are replaced with redox-innocent NMe₂or SMe. A derivative of 2 in which the phenylene backbone was replaced with ethylene

(3) was also prepared to interrogate the importance of o-phenylenediamine for ligand-centered redox activity. Cyclic

voltammograms collected for 1 and 2 revealed two fully reversible ligand-centered redox events. Remarkably, several quasi-

reversible ligand-centered redox waves were also observed for 3 despite the absence of the o-phenylenediamine subunit.

Oxidizing 1 and 2 with silver salts containing diﬀerent counteranions (BF₄, OTf⁻, NTf₂) allowed the electrochemically

generated complexes to be analyzed as a function of diﬀerent oxidation states using single-crystal X-ray diﬀraction (XRD), EPR

spectroscopy, and S K-edge X-ray absorption spectroscopy. The experimental data are corroborated by DFT calculations, and

together, they reveal how the location of unpaired spin density and electronic structure in singly and doubly oxidized salts of 1

and 2 varies depending on the coordinating ability of the counteranions and exogenous ligands such as pyridine.

−

by the fully oxidized o-diiminoquinone (Chart 1).¹⁴Since the

INTRODUCTION

■

pioneering work of Holm^15,16and important electronic

structure investigations by Wieghardt,¹⁷o-phenylenediamine

has been incorporated into conjugated multidentate ligands as

Metal complexes with redox-active ligands are renowned for

their ability to enable and promote new types of reactivity,^1,2

especially with metals that prefer to undergo one-electron

redox transformations or display limited redox activity on their

own. Given their crucial role in mediating electron-transfer

reactions in metalloproteins and their prominent use in the

development of homogeneous catalysts with earth abundant

metals,³⁻¹³understanding how redox-active ligand modiﬁca-

tions aﬀect the electronic structure and reactivity of metal

complexes remains an important goal.

Chart 1. One-Electron Redox Transformations with o-

Phenylenediamine

o-Phenylenediamine is a classic redox-active ligand. While

bound to metals, it can undergo two reversible one-electron

oxidations to form the o-diiminosemiquinone radical followed

Received: June 5, 2019

A

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

a means to tune ligand redox properties, improve complex

stability, and accommodate diﬀerent metal coordination

environments. This is perhaps most well-known with

tetradentate salophen ligands,^18,19but another approach is to

use direct N-aryl linkages to access additional ligand-centered

redox transformations and alter the locus of ligand-centered

redox events. For example, Wieghardt and co-workers showed

that square planar Cu(II) and Zn(II) complexes with triaryl

ONNO ligands undergo ligand-centered redox reactions that

involve both the o-phenylenediamine backbone and the

ﬂanking phenolate arms (Chart 2).¹⁷In contrast, Thomas

of o-phenylenediamine and related redox-active ligands,^25,26

which further complicates redox assignments and interpreta-

tion. It has also been shown that thioethers can be redox-

noninnocent in diﬀerent ligand frameworks such as thioether

crowns,²⁷⁻²⁹which has raised questions as to whether the

thioether substituents in L2 are indeed redox-innocent.

However, as we will show, the thioether substituents in L2

do not appear to be participating in the ligand-centered redox

events.

To simplify our electrochemical investigations and help

identify the locus of ligand-centered redox activity, we

prepared a series of Ni(II) complexes with the aforementioned

SNNS ligand L2, as well as a new triaryl tetradentate ligand

containing ﬂanking NMe₂substituents (L1; Chart 2). Nickel

was selected because it does not have readily accessible redox

couples to interfere with assignments in our electrochemical

window of interest, and the more contracted 3d orbitals on

nickel are expected to delocalize less into the π manifold of

redox-active ligands relative to 4d orbitals on ruthenium. The

tetradentate ligand H₂(^MeNNNN^Me)-DAB (L1) was prepared

and investigated alongside our previously reported

H₂(^MeSNNS^Me)-DAB ligand (L2) to compare substituents

with diﬀerent hard/soft donor strength, evaluate potential

diﬀerences in redox noninnocence, and provide a more direct

comparison to the NNNN complexes described by Thomas et

al.²⁰While dithiolate SNNS complexes without the o-

phenylenediamine subunit are known with Ni,³⁰we are not

aware of any examples with similar triaryl ligands and ﬂanking

thiol substituents to evaluate against our thioether-substituted

ligand L2. Finally, to investigate the importance of the o-

phenylenediamine backbone for ligand-centered redox activity,

we prepared a known derivative of the SNNS ligand L2 in

which the phenylene backbone was replaced with an ethylene

linker (L3; Chart 2). Here we present our synthetic results

alongside comparative X-ray diﬀraction data, electrochemical

studies, spectroscopic analyses, and density functional theory

(DFT) calculations to elucidate the electronic structure of

these complexes as a function of diﬀering oxidation states,

ligand modiﬁcations, and coordinating ancillary ligands.

Chart 2. Triaryl Tetradentate Ligands Described Previously

and Abbreviations for Ligands Reported Here

and co-workers more recently reported that isostructural

Ni(II) NNNN complexes with ﬂanking aniline groups undergo

four redox transformations that localize on the aniline

groups.²⁰

Flanking OH and NH₂donor substituents in triaryl

tetradentate ligands, as shown in Chart 2, are considered

noninnocent because they actively participate in redox

transformations with concomitant proton loss and change in

O or N hybridization. We recently began investigating how

replacing these ﬂanking donors with redox-inactive (i.e., redox-

innocent) substituents alters ligand-centered redox activity. We

postulated that redox-innocent donor groups would help direct

redox activity to the o-phenylenediamine backbone, which is

desirable for our eﬀorts to understand how metal−ligand

cooperativity (MLC) across metal−nitrogen bonds aﬀects

redox activity at the ligand and metal. However, our previous

MLC studies with ruthenium complexes containing the

H₂(^MeSNNS^Me)-DAB ligand (L2, Chart 2) yielded redox

waves in the cyclic voltammograms that were diﬃcult to assign

based on comparison to electrochemical data reported for

similar systems.²¹⁻²⁴These diﬃculties stemmed in part from

the fact that ruthenium has a Ru³⁺/Ru²⁺redox couple that can

fall within the range of redox potentials expected for o-

phenylenediamine ligands.²¹⁻²⁴Moreover, ruthenium 4d

orbitals are often highly delocalized with frontier π orbitals

RESULTS AND DISCUSSION

■

Synthesis and Characterization of Ni(II) Complexes.

The triaryl H₂(^MeNNNN^Me)-DAB ligand (L1) was prepared by

palladium-catalyzed Buchwald−Hartwig cross-coupling of o-

phenylenediamine with 2 equiv of 2-bromo-N,N′-dimethylani-

line, using a procedure similar to that described for the

H₂(^MeSNNS^Me)-DAB ligand (L2).³¹L1 was isolated in 62%

yield after puriﬁcation by column chromatography and

1

subsequent crystallization. H and ¹³C NMR spectroscopy,

electron impact mass spectrometry (EI-MS), microanalysis,

infrared spectroscopy, and single-crystal X-ray diﬀraction

(XRD) conﬁrmed the identity of L1. The chemically

1

equivalent NMe₂groups appeared as a singlet in the H and

¹³C NMR spectra at δ 2.58 and 44.16 ppm, respectively, and

Scheme 1. Syntheses and Structures of 1−3

B

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

a

Table 1. Selected Bond Distances (Å) and Angles (deg) from Single-Crystal XRD Data

1

1[BF₄]

1[OTf]₂

2

2[BF₄]

2[NTf₂]₂

2[OTf]₂

3

Ni−N1

Ni−N2

Ni−X1

Ni−X2

Ni−O

1.830(3)

1.843(4)

2.025(4)

1.989(3)

1.853(3)

1.845(3)

1.981(3)

2.008(3)

2.002(3)

1.992(2)

2.112(2)

2.120(3)

2.160(2)

2.145(2)

1.512(4)

1.309(3)

1.306(3)

1.406(3)

1.400(3)

81.18(9)

112.42(9)

1.847(4)

1.845(4)

2.156(2)

2.149(1)

1.914(6)

1.791(5)

2.081(2)

2.240(2)

1.872(2)

1.871(2)

2.1933(7)

2.1880(6)

2.847(3)

2.794(2)

1.484(3)

1.321(3)

1.327(3)

1.413(3)

1.415(3)

84.12(8)

97.47(2)

2.005(3)

2.000(3)

2.3787(9)

2.391(1)

2.147(2)

2.069(2)

1.505(5)

1.307(5)

1.402(5)

1.405(5)

80.7(1)

1.845(7)

1.850(7)

2.183(2)

2.176(3)

C1−C2

N1−C1

N2−C2

N1−C11

N2−C21

N−Ni−N

X−Ni−X

1.433(6)

1.392(6)

1.392(5)

1.380(6)

1.383(5)

84.5(1)

1.449(4)

1.359(4)

1.354(4)

1.401(4)

1.400(4)

84.2(1)

1.414(8)

1.401(6)

1.400(6)

1.384(7)

1.376(7)

86.2(2)

1.45(1)

1.311(7)

1.37(1)

1.380(9)

1.42(1)

86.1(2)

99.61(8)

1.54(1)

1.44(1)

1.45(1)

1.37(1)

1.35(1)

85.2(3)

97.2(1)

105.4(1)

104.8(1)

96.95(6)

109.82(3)

a

X = SMe or NMe₂. All crystal structures follow the labeling scheme shown in Figure 1.

Figure 1. Molecular structures of 1 (left) and 2 (right) from single-crystal XRD studies. Ellipsoids are drawn at the 50% probability level. Hydrogen

atoms were omitted from the ﬁgure.

Figure 2. CV traces collected on 1 mM solutions of 1−3 in THF at 100 mV/s scan rates at room temperature with 0.1 M TBAPF₆. WE = glassy

carbon, CE = Pt wire, RE = quasi Pt. (left) Scans collected out to ca. −3 V vs Fc⁺/Fc. (right) CV scans collected with a truncated electrochemical

window. Arrows signify the initial scan start point and scan direction.

the exchangeable N−H proton resonance was observed as a

broad resonance in the ¹H NMR spectrum at δ 6.65 ppm. The

remaining peak positions and integrations supported the triaryl

framework of the ligand. The SNNS congener of L2 in which

the phenylene backbone is replaced with an ethylene linker

referred to here as H₂(^MeSNNS^Me)-DAE (L3)was prepared

using a modiﬁed procedure reported by Bayer and

Breitmaier.³²

The Ni(II) complexes 1−3 were prepared by reﬂuxing a 1:1

mixture of the corresponding ligand (L1−L3) and NiBr₂in

THF overnight (Scheme 1). In the absence of base, each

reaction yielded a light green precipitate that was not

characterized further due to its insolubility in organic solvents.

Addition of 2 equiv of NaO^tBu to the green precipitate,

however, resulted in color changes over the course of 12 h to

yield dark brown Ni(^MeNNNN^Me)-DAB (1), dark green

Ni(^MeSNNS^Me)-DAB (2), and dark blue Ni(^MeSNNS^Me)-

DAE (3), respectively. All three compounds were crystallized

in moderate yields (54−60%), and their formulation was

conﬁrmed by elemental analysis and NMR spectroscopy. The

¹H and ¹³C NMR spectra for 1 in THF-D₈revealed singlets for

the chemically equivalent NMe₂substituents at δ 3.05 and

50.34 ppm. The ¹H NMR resonances for the SMe substituents

in 2 and 3 appeared more upﬁeld at δ 1.71 and 1.87 ppm,

C

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

a

Table 2. Redox Potentials of 1−3

Ni features

ligand features

b

c

b

d

b

1

2

3

potential

ΔE peak

potential

ΔE peak

potential

ΔE peak

−2.52

0.18

−2.87

−0.36

0.18

−0.19

0.28

0.23

0.18

0.22

0.27

c

b

−2.18

irreversible

c

d

−2.72

−2.04

−0.39

0.08

0.47

irreversible

a

CV traces collected on 1 mM solutions in THF at 100 mV/s scan rates at room temperature with 0.1 M TBAPF₆. WE = glassy carbon, CE = Pt

wire, RE = quasi Pt. ΔE values are reported for reversible features (for comparison, ΔE = 0.21 V for Fc⁺/Fc). All potentials and ΔE values are

b

c

d

expressed in V and referenced to Fc⁺/Fc. Reported as E_1/2for reversible waves. Reported as E_pc. Reported as E_pa.

respectively. While diﬀerent stereoisomers of 2 and 3 are

expected based on the possible orientations of the two SMe

groups,³¹they could not be discerned presumably due to rapid

exchange on the NMR time scale. The ¹³C NMR spectra for 2

and 3 also revealed broadened resonances for the SMe groups

at δ 24.16 and δ 24.56 ppm, respectively.

−0.36 V, which reﬂects the decreased donor ability of the SMe

groups in 2 vs the NMe₂groups in 1. In contrast, the second

ligand oxidations at 0.23 and 0.22 V are nearly identical.

We previously reported two similar features in the CV trace

of Ru(^MeSNNS^Me)(PPh₃) in THF (where ^MeSNNS^Me= L2),

albeit at more negative potentials (−0.78 and −0.28 V vs Fc⁺/

Fc).³¹An irreversible reduction wave was also observed at

−2.46 V in the CV of Ru(^MeSNNS^Me)(PPh₃) similar to that

observed for 2. Based on our analysis here, we can revise our

tentative redox assignments made previously for the reversible

one-electron waves at −0.78 and −0.28 V in Ru(^MeSNNS^Me)-

(PPh₃)³¹as ligand-centered redox events assigned to L2.

We also collected the CV of 3 to evaluate the importance of

the phenylene backbone in the redox chemistry of 1 and 2.

Scanning the full electrochemical window of 3 in THF yields a

complicated CV with irreversible reduction waves at E_pcvalues

of −2.72 and −2.04 V similar to those observed in the CV of 2

(Figure 2). Limiting the scanned electrochemical window as

described above yields a cleaner CV trace with several quasi-

reversible waves. The redox activity of the ligand in 3 is not

eliminated completely by removing the phenylene backbone,

although reversibility was clearly attenuated, as shown by lower

peak current densities in the associated redox features. This

suggests the isolated, ﬂanking aryl groups still participate in

ligand-centered redox reactions, even with innocent ﬂanking

donor substituents such as thioethers. Intuitively, it also

provides compelling evidence that including the aromatic

backbone between the two peripheral rings strengthens

electrochemical communication of redox-active triaryl ligands

in 1 and 2.

Single-crystal XRD studies revealed the anticipated square

planar coordination geometries for all three complexes in the

solid state. Selected bond distances and angles are presented in

Table 1. The Ni−N bond distances range from 1.830(3) to

1.850(7) Å, consistent with covalent Ni−N bond distances

reported previously with anionic amido ligands with ﬂanking

aryl substituents.^20,33,34The Ni−SMe and Ni−NMe₂distances

also fall within their expected ranges at 2.149(1)−2.183(2) and

1.989(3)−2.025(4) Å, respectively.^35,36Bond distances

associated with the ligand will be discussed below with the

structures of the chemically oxidized complexes (vide infra).

Electrochemical Studies. Cyclic voltammograms (CVs)

of 1−3 were collected in THF to determine how changes in

the neutral donor atoms (NMe₂in 1 vs SMe in 2) and

backbone linker (phenylene in 2 vs ethylene in 3) aﬀect the

redox chemistry of the Ni complexes (Figure 2). CVs of 1 and

2 were also collected in DCM (the solvent used for EPR

studies below), and the results were consistent with those

obtained in THF (Figures S52 and S53, Supporting

Information). The CV of NMe₂-substituted 1 shows three

fully reversible one-electron redox events with E_1/2values of

0.23, −0.36, and −2.52 V vs Fc⁺/Fc. The most negative wave

at −2.52 V is assigned as the Ni²⁺/Ni¹⁺redox couple based on

comparison to literature reports.^35,37−40The two waves at

−0.36 and 0.23 V are assigned to ligand-centered redox

features and are similar to two of the four ligand-centered

redox potentials for the Ni(^HNNNN^H) complex shown in

Chart 2 (−0.41 and 0.02 V).²⁰Using innocent NMe₂

substituents here in 1, compared to redox-active NH₂

substituents in a similar system reported by Thomas et al.,

eliminates their additional ligand-centered redox waves at

−1.75 and −1.04 V.²⁰

Chemical Oxidations of 1 and 2. To investigate the

chemical and electronic structures of the electrochemically

generated species observed in the CV traces, 1 and 2 were

chemically oxidized using silver salts. Similar studies were

attempted with 3, but we were unable to isolate pure

compound from these mixtures. Singly oxidized complexes

were generated by mixing one-to-one equivalents of 1 or 2

with AgBF₄in CH₂Cl₂. After ﬁltration to remove Ag(0), dark

green crystals of 1[BF₄] and 2[BF₄] were obtained by layering

The CV trace of 2 reveals similar ligand-centered redox

events as 1 but diﬀers in that the most negative redox couple

assigned as Ni²⁺/Ni¹⁺is irreversible for 2. Scanning the full

electrochemical window in THF reveals two irreversible one-

electron reduction waves at E_pc−2.18 and −2.87 V along with

additional features in the oxidation scan attributed to

electrochemically active degradation products. Limiting the

scans to only 0.6 V vs Fc⁺/Fc (Figure 2, right column)

reveals two reversible one-electron exchange waves with E_1/2

values of −0.19 and 0.22 V assigned to ligand-centered redox

events. The ﬁrst ligand oxidation at −0.19 V is shifted 170 mV

more positive compared to the same feature in the CV of 1 at

1

the CH₂Cl₂ﬁltrate with Et₂O. H and ¹³C NMR spectra

collected on dissolved crystalline material of 1[BF₄] and

2[BF₄] conﬁrmed that the complexes were NMR silent,

consistent with the formation of paramagnetic ligand-centered

radicals. Only ¹¹B and ¹⁹F resonances associated with the

−

outer-sphere BF₄anions were observed. 1[BF₄] revealed

sharp resonances at δ −1.36 and −151.7 ppm in the ¹¹B and

¹⁹F NMR spectra, respectively. 2[BF₄] showed a similar

resonance in the ¹¹B NMR spectrum at δ −1.12 ppm, but the

¹⁹F NMR spectrum revealed an uncharacteristically shifted and

broadened resonance at δ −149.4 ppm. The diﬀerence in ¹⁹F

D

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

NMR data for 1[BF₄] and 2[BF₄] suggests that outer-sphere

BF₄anions interact more with the paramagnetic complex

lower energies in the spectrum of 2 (598 and 800 nm,

respectively).

−

containing the thioether-substituted ligand L2. This observa-

tion was corroborated by EPR studies described in the

following section (see below).

The UV−vis spectrum of 3 helped to elucidate the origin of

some of the spectral features for 1 and 2. The highest-energy

absorption band of 1 and 2 (210 and 215 nm, respectively) is

absent in the spectrum of 3, suggesting that this feature is

associated with the o-phenylenediamine subunit. The highest

energy absorption feature for 3 is signiﬁcantly lower at 240 nm,

which appears to correspond with the additional shoulder

observed in the UV−vis spectrum of 2 at ca. 230 nm. The

absence of this shoulder in the spectrum of 1 suggests that

these absorptions are associated with LLCT transitions

involving the SMe groups on 2 and 3. The other notable

diﬀerence in the spectra of 3 is the more intense d−d transition

at 671 nm which displays a slight high-energy shoulder (Figure

3). This suggests that the two d−d transitions observed in the

spectra of 1 and 2 are less resolved in the spectrum of 3.

We next compared the change in the UV−vis-NIR spectra of

1 and 2 as they were chemically oxidized (Figure 4). Oxidation

Attempts to oxidize 1[BF₄] and 2[BF₄] further using AgBF₄

were not productive because it aﬀorded complex salts that

were insoluble in all organic solvents we tested. Instead, the

doubly oxidized complexes were prepared by treating CH₂Cl₂

solutions of 1 or 2 with 2 equiv of AgOTf or AgNTf₂. After

ﬁltration, dark purple blocks of 1[OTf]₂and 2[OTf]₂and

green needles of 2[NTf₂]₂were grown by layering the CH₂Cl₂

ﬁltrate with Et₂O. Unlike the NMR silent singly oxidized salts,

several broad and paramagnetically shifted ¹H NMR

resonances were observed at δ 2.40, 3.01, 12.73, 14.94,

1

21.13, and 39.44 ppm for 1[OTf]₂. Broad H resonances are

also observed for 2[OTf]₂and 2[NTf₂]₂.

¹⁹F NMR spectroscopy was again used to conﬁrm the

presence of the ﬂuorinated outer-sphere anions. The ¹⁹F NMR

spectrum of 1[OTf]₂revealed a broad peak at δ −38.5 ppm

that was shifted signiﬁcantly from where free triﬂate is typically

observed in diamagnetic salts (δ −74 to −77 ppm).⁴¹Similarly,

a broad ¹⁹F resonance is observed for 2[OTf]₂at δ −41.7 ppm.

Like the NMR studies of singly oxidized 2[BF₄], the change in

¹⁹F NMR shifts for 1[OTf]₂and 2[OTf]₂suggested that the

triﬂate anions are bound in solution, as observed in solid-state

XRD studies (see below). In contrast, the ¹⁹F NMR spectrum

of 2[NTf₂]₂revealed only a slight shift in the ¹⁹F resonance at

δ −72.2 ppm compared to that typically reported for free

triﬂimidate at δ −76.0 ppm,^42,43consistent with the decreased

coordination ability of triﬂimidate relative to triﬂate.

UV−vis-NIR Spectroscopy. UV−vis data were collected

for 1−3 to analyze their electronic structure diﬀerences as a

function of ligand modiﬁcations and counteranions (Figure 3).

Figure 4. UV−vis-NIR data collected for the singly and doubly

oxidized complexes. (A) Comparison of 1[BF₄] and 1[OTf₂]₂. (B)

Comparison of 2[BF₄], 2[OTf]₂, and 2[NTf₂]₂. Data were collected

on 50 μM (UV) and 250 μM (vis-NIR) solutions in DCM at 298 K.

Figure 3. Comparison of UV−vis spectra for neutral complexes 1, 2,

and 3. The inset shows the magniﬁed visible region with the dipole

forbidden d−d transitions. Data were collected on 50 μM (UV) and

800−1200 μM (vis) solutions in DCM at 298 K.

of 1 attenuates the most intense charge transfer transitions in

the spectra of 1[BF₄] and 1[OTf]₂, and similar observations

are made for oxidation of 2. A prominent absorption observed

at 395 nm for 1[BF₄] and 2[BF₄] decreases in intensity for the

doubly oxidized complexes. The dipole-forbidden d−d

transitions increase in intensity but are located at similar

energies as those observed for neutral 1 and 2. The NIR region

of singly oxidized complexes revealed the appearance of new

low-energy peaks at 1124, 1336, and 1674 nm for 1[BF₄] and

1155, 1373, and 1690 for 2[BF₄]. These are commonly

assigned as intervalence charge transfer (IVCT) bands and are

indicative of a delocalized ligand-centered radical (Figure 4).⁴⁴

The UV−vis spectrum of 1 shows three intense charge transfer

bands at 210, 278, and 355 nm that are assigned to either

ligand−ligand charge transfer (LLCT) or metal−ligand charge

transfer (MLCT) transitions. Complex 2 also shows three

similar UV−vis absorption features at similar energies of 215,

296, and 350 nm, albeit with increased intensity relative to

those in 1. The spectrum of 2 contains an additional shoulder

at 230 nm that was not observed in the spectrum of 1. Two

weak absorptions assigned to dipole-forbidden d−d transitions

were observed at 558 and 704 nm for 1, and these shift to

E

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

As expected, the IVCT bands do not appear in the NIR spectra

of the doubly oxidized complexes.

The Ni−O distances were 2.160(2) and 2.145(2) Å in

1[OTf]₂and 2.147(2) and 2.069(2) Å in 2[OTf]₂. These

structures conﬁrm the ¹⁹F NMR results that showed larger

than expected chemical shifts based on comparison to free

triﬂate. In contrast, the structure of 2[NTf₂]₂showed no

signiﬁcant interaction with the outer-sphere triﬂimidate anions,

again consistent with the ¹⁹F NMR data. The closest

triﬂimidate atom to nickel was an oxygen with a Ni−O

distance of 2.794(2) Å.

The average Ni−N bond distances associated with the o-

phenylenediamine backbone are similar in 1 and 1[BF₄] at

1.837 and 1.849 Å, respectively, but increase to 1.997 Å as the

coordination number changes from four to six (Table 1). A

nearly identical trend with similar values was observed for 2,

2[BF₄], and 2[OTf]₂. Comparison of the average Ni−N

distances in six-coordinate 2[OTf]₂at 2.003 Å to those in four-

coordinate 2[NTf₂]₂at 1.872 Å indicates that larger changes in

bond distance are primarily due to changes in coordination

number. This is further supported by comparison of the

average Ni−S distances: only a small stepwise increase is

observed for 2 (2.153 Å), 2[BF₄] (2.161 Å), and 2[NTf₂]₂

(2.191 Å), but this increases more abruptly to 2.385(1) Å in

six-coordinate 2[OTf]₂.

Structural Analysis of the Oxidized Complexes. XRD

data were collected on single crystals of the oxidized complexes

so their bond distances and structures could be compared as a

function of oxidation state. Singly oxidized 1[BF₄] and 2[BF₄]

revealed monomeric structures with outer-sphere BF₄⁻anions.

Notably, the structure of the thioether-substituted congener

2[BF₄] features two Ni complexes that were stacked close

together (Figure 5). The intermetallic Ni···Ni distance of

We next analyzed how oxidation and ligand substituents

aﬀect the four-coordinate geometry around Ni (Table 3). All

Figure 5. Molecular structure of 2[BF₄] from single-crystal XRD

studies. Ellipsoids are drawn at the 50% probability level. Hydrogen

−

atoms and outer-sphere BF₄were omitted from the ﬁgure.

Table 3. Deviations from Square Planar to Tetrahedral

Geometry for Four-Coordinate Complexes

3.264(1) Å is longer than the Ni−Ni distance in a structurally

similar [Ni(^HNNNN^H)]₂²⁺dimer reported by Thomas and co-

workers (2.784 Å).²⁰The close Ni···Ni distance in 2[BF₄]

relative to monomeric 1[BF₄] is attributed to the additional

conformational ﬂexibility of the methyl substituents attached

to sulfur in L2, which can rotate away from one face of the Ni

complex to accommodate stacking between the two units. In

contrast, the methyl substituents on the more rigid NMe₂

groups in 1[BF₄] project axially on both sides of the square

plane and block similar π−π stacking.

θ (deg)

τ_sq‑tet(θ/90°)

tetrahedral

sq. planar

1

1[BF₄]

2

90

0

1

0

6.6

3.1

12.1

13.4

3.9

0.07

0.03

0.13

0.15

0.04

2[BF₄]

2[NTf₂]₂

XRD data collected for the doubly oxidized complexes

1[OTf]₂and 2[OTf]₂revealed octahedral coordination

geometries with trans triﬂate anions bound to Ni (Figure 6).

of the four-coordinate complexes are roughly square planar,

and we used the angle θ between the two planes deﬁned by

N−Ni−N and S−Ni−S to quantify the distortion toward

Figure 6. Molecular structures of 1[OTf]₂(left) and 2[OTf]₂(right) from single-crystal XRD studies. Ellipsoids are drawn at the 50% probability

level. Hydrogen atoms were omitted from the ﬁgure.

F

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

tetrahedral (Figure 7). As shown in Table 3, the neutral and

singly oxidized complexes with L2 (X = SMe) show greater

shown in Table 1, the NC−CN bond distances are slightly

shorter in 2[NTf₂]₂at 1.484(3) Å compared to 1.505(5) Å in

2[OTf]₂. Likewise, all of the N−C bond distances are slightly

longer in the structure of 2[NTf₂]₂. As we will show, these

subtle diﬀerences in ligand bond distances foreshadow more

pronounced diﬀerences in the electronic structure of 2²⁺when

anions bind to Ni.

Magnetic Susceptibility Measurements. To determine

the number of unpaired electrons present in the oxidized

complexes, their magnetic susceptibilities were determined

using the Evans method.^46,47The results are compiled in Table

4 alongside the theoretical spin-only values expected for

Figure 7. Angle used to quantify deviation from square planar

geometry.⁴⁵

deviations from planarity relative to those in L1 (X = NMe₂)

presumably due to the more weakly coordinating and

asymmetric thioether groups. Single oxidation of 1 and 2

had only a subtle change on the degree of distortion in the

structure of 1[BF₄] and 2[BF₄], but the structure of doubly

oxidized 2[NTf₂]₂showed a signiﬁcant decrease in θ

consistent with increased sp²character on N in the o-

phenylenediamine backbone due to multiple ligand-centered

oxidations.

The change in ligand bond distances for 1 and 2 matches

those expected for stepwise oxidation of o-phenylenediamine

(as shown in Chart 1) and contrasts the changes reported

previously for Ni(^HNNNN^H) complexes with redox-active

ﬂanking NH substituents. The dominant resonance structures

for 1, 1⁺, and 1²⁺are shown in Figure 8 based on the observed

changes in bond distances from the single-crystal XRD data.

Upon oxidation, the average N−C bond distances associated

with the o-phenylenediamine subunit decrease from 1.392(8)

(1) to 1.357(6) (1[BF₄]) to 1.308(4) Å (1[OTf]₂)

corresponding to an increase in double bond character (Figure

8a). A complementary increase is observed in the backbone

NC−CN bond distance from 1.433(6) to 1.449(4) to

1.512(4) Å corresponding to an increase in single bond

character (Figure 8a). Similar trends are observed in the

stepwise oxidation of 2 (Table 1). For comparison, the neutral

Ni(^HNNNN^H) diradical complexes depicted in Figure 8b

undergo stepwise ligand-centered oxidations on the peripheral

aryl groups.

Table 4. Average Evans Method Magnetic Susceptibility

Values for Singly and Doubly Oxidized Complexes of 1 and

2 Reported in Units of Bohr Magnetons (μ_B) from

a

Triplicate Measurements

singly oxidized

doubly oxidized

1[OTf]₂

1[BF₄]

1[NTf₂]

2[BF₄]

2[NTf₂]

1.6(1)

1.7(1)

1.5(2)

1.73

2.9(2)

2[OTf]₂

2[NTf₂]₂

theoretical (S = 1)

2.5(2)

1.9(1)

2.83

theoretical (S = 1/2)

a

Theoretical spin-only values are included for comparison. Measure-

ments were collected in DCM at 298 K with a protio/deuterated

DCM capillary insert and in a range of 2−10 μM.

doublet (S = 1/2) and triplet (S = 1) electron conﬁgurations.

The singly oxidized complexes are all close to the spin-only

value of 1.73 μ_B, as would be expected for the o-

diiminosemiquinone. However, the doubly oxidized complexes

show a greater range of magnetic susceptibilities. The triﬂate

complexes 1[OTf]₂and 2[OTf]₂are both close to the spin-

only value for an S = 1 conﬁguration at 2.9(2) and 2.5(2) μ_B

(theoretical 2.83 μ_B), but the triﬂimidate complex 2[NTf₂]₂

was signiﬁcantly lower at 1.9(1) μ_B, which is closer to the S =

1/2 value. Variable temperature Evans method experiments

with 2[NTf₂]₂revealed Curie−Weiss behavior between the

temperatures of −80 and 30 °C where the μ_effspin value

increases to a maximum of 2.15 μ_Bat the lowest temperature

−1

In general, bond distances for the doubly oxidized complexes

1²⁺and 2²⁺suggest formation of the o-diiminoquinone, as

shown in Figure 8a. However, one of the more interesting

structural comparisons concerns the ligand bond distances in

four-coordinate 2[NTf₂]₂and six-coordinate 2[OTf]₂. As

(−80 °C). Plotting χ_Magainst temperature shows that

2[NTf₂]₂displays antiferromagnetic behavior (Weiss constant:

θ < 0) (Figure S51).⁴⁸

EPR Spectroscopy. EPR data were collected to investigate

the distribution of unpaired electron spins in all of the oxidized

Figure 8. Comparison of dominant Lewis structures based on single-crystal XRD data for sequential oxidation of (a) 1 with ﬂanking NMe₂

substituents and (b) Ni(^HNNNN^H) complexes with ﬂanking NH substituents from ref 19. The tert-butyl substituents attached to the ﬂanking aryl

groups (as shown in Chart 2) were omitted from the ﬁgure in part b.

G

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Figure 9. (left) X-band EPR spectra of native 1[BF₄] and 2[BF₄] at 109 K. (right) X-band EPR spectra of 1[BF₄] and 2[BF₄] at 109 K in the

presence of excess pyridine. Experimental spectra were collected in neat DCM. Simulated data are shown in red.

complexes. Initial experiments were performed by dissolving

crystalline material into neat DCM. The room-temperature

spectra of singly oxidized 1[BF₄] and 2[BF₄] revealed an

isotropic signal at g_iso= 2.008 consistent with the presence of a

ligand-centered radical. However, upon cooling to 109 K,

subtle diﬀerences between the two compounds were observed.

The low-temperature (LT) EPR data of 1[BF₄] show a slightly

distorted isotropic signal with g_iso= 2.007. The LT-EPR data of

2[BF₄] revealed a similar axial signal at g ∼ 2 but also a low-

intensity spectroscopic feature around g ∼ 2.2 (Figure 9). The

signals around g ∼ 2 are typical of the S = 1/2 ground state for

systems containing an o-iminosemiquinone radical, whereas

the signal at g ∼ 2.2 is commonly assigned to Ni(III).⁴⁹⁻⁵²No

hyperﬁne splitting is observed with the o-phenylenediamine

nitrogen atoms, similar to that reported for square planar

Ni(III) salophen complexes.⁵²

Figure 10. EPR spectrum of 2[BF₄] at 109 K experimental spectra

(black) and simulated spectra (red). Spectrum collected in 2-

MeTHF/DCM (1:3).

Previous reports of closely related nickel complexes with

tetradentate Schiﬀ-base ligands describe how a ligand-derived

singly occupied molecular orbital (SOMO) can transition to a

metal-based SOMO via valence tautomerization.^53,54This type

of tautomerization can be further deﬁned as temperature-

dependent (where population of the SOMO changes as a

function of temperature, thereby resulting in relative changes

in diﬀerent signal intensities) or as coordination-induced (where

coordination to the metal destabilizes a metal-based orbital to

yield a new spectroscopic signal).⁵⁴⁻⁵⁶

To investigate possible temperature dependent tautomerism,

variable-temperature EPR spectroscopic experiments of

2[BF₄] were conducted in 2-MeTHF/DCM (1:3) solutions.

Upon cooling to 109 K, the Ni(III) signals at g_iso= 2.215

became more well resolved and showed satellite peaks to either

side of the isotropic signal (Figure 10). However, warming the

sample and collecting spectra showed that metal and organic

signal intensities decrease uniformly (Figure S35), ruling out

temperature-dependent valence tautomerism. This suggested

that the new spectroscopic EPR feature was coordination-

induced, presumably due to interactions between the nickel

center in 2[BF₄] and the BF₄⁻anion. Evidence of this Ni···F−

BF₃interaction in the EPR spectrum of 2[BF₄] and not

1[BF₄] is consistent with the decreased steric proﬁle of SMe vs

NMe₂and the increased Lewis acidity of Ni in 2[BF₄]. As

described in the comparison to the XRD data, the ﬂuxional

SMe substituents in 2[BF₄] can rotate away from one axial face

to allow the BF₄⁻anion to interact with Ni in solution. Indeed,

simulations required modeling a ¹⁹F hyperﬁne splitting

interaction (A = 777 MHz) at g = 2.186 and an isotropic

Ni(III) feature at g_iso= 2.215, indicating an interaction

between the counteranion and the complex cation (Figure 10).

The magnitude of the hyperﬁne splitting with axial ¹⁹F is

consistent with EPR data reported for other Ni(III) complexes

with Ni−F bonds.⁵⁷

To further test our hypothesis regarding the coordination-

induced valence tautomerism, we performed two experiments:

H

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Table 5. Calculated g Values from Simulated EPR Spectra of Singly Oxidized Complexes Collected in DCM at 109 K

Ni(III) region

ligand radical region

b

compound

g_xx

g_yy

g_zz

g_iso

g_xx

g_yy

g_zz

g_iso

1[BF₄]

1-py[BF₄]

1[NTf₂]

1-py[NTf₂]

2[BF₄]

2-py[BF₄]

2[NTf₂]

2.007

2.003

2.007

2.008

2.010

2.008

c

2.287

2.275

2.240

2.213

2.167

2.231

2.025

2.005

1.993

c

2.164

2.206

2.217

2.215

2.256

2.231

2.025

2.016

2.025

2.029

2.005

2.008

2.003

2.006

1.992

1.985

1.992

1.990

d

a

2.224

c

2.263

2. 261

2.222

2.170

c

2-py[NTf₂]

2.266

2.184

2.205

2.024

2.012

2.004

a

b

c

d

Values determined at RT. Calculated average values. Experiments were conducted in the presence of 3000 equiv of pyridine. Collected in 2-

e

MeTHF/DCM. Fluorine splitting was modeled as a separate S = 1/2 species with g_iso= 2.186 and A = 11.62, 8.45, and 777.41 MHz.

−

•

[Ni^II(L )]⁺+ n(py) F [Ni^III(L)](py)_n

(1) We replaced the BF₄anion with more noncoordinating

(1)

−

triﬂimidate (NTf₂), and (2) we recollected the EPR data on

The ﬁnal question for the results with pyridine concerns the

number or pyridine ligands that bind to Ni (n in eq 1). It is

known that pyridine addition to singly oxidized Ni complexes

containing redox-active ligands typically yields octahedral

complexes with two pyridines bound to the metal,^52,58−62

although examples of ﬁve-coordinate complexes with one

bound pyridine are also known.⁶²Addition of pyridine to

solutions of 2[BF₄] and 2[NTf₂] resulted in an immediate

color change, but we were unable to isolate and structurally

characterize the resulting products to determine the number of

metal-bound pyridine ligands. However, UV−vis titration

studies of 2[NTf₂] with excess pyridine suggested that only

one pyridine binds to nickel (Figure 11). While we cannot

1[BF₄] and 2[BF₄] in the presence of excess pyridine, as has

been described for other singly oxidized Ni complexes with

redox-active ligands.^52,58−62The singly oxidized triﬂimidate

complexes 1[NTf₂] and 2[NTf₂] were prepared as described

for the BF₄salts by treating 1 and 2 with 1 equiv of AgNTf₂in

DCM. Filtering the reaction mixture and layering with Et₂O

aﬀorded dark green crystals of 1[NTf₂] and 2[NTf₂]. Similar

1

to 1[BF₄] and 2[BF₄], the triﬂimidate salts were H NMR

silent but produced sharp and unshifted ¹⁹F signals (δ −79.2

and −78.8 ppm, respectively),⁴²suggesting little to no

interaction with the native oxidized complexes. Subsequent

EPR investigations led to only isotropic resonances centered at

g ∼ 2 at both RT and 109 K for 1[NTf₂] and 2[NTf₂] (Figures

S38 and 39 and S43 and S44, respectively). No Ni(III) signals

were observed, thereby conﬁrming the lack of signiﬁcant

interactions between the complex cations and the non-

coordinating triﬂimidate anions in solution.

Next, excess dry pyridine was added to neat DCM solutions

of 1[BF₄], 1[NTf₂], 2[BF₄], and 2[NTf₂] and their EPR

spectra were collected. At room temperature, the EPR

spectrum for 2[BF₄] with excess pyridine yielded a resolved

isotropic signal at g ∼ 2.20, suggesting pyridine coordination to

the Ni (Figure S42). In contrast, the room-temperature EPR

spectrum for 1[BF₄] with pyridine still shows only an isotropic

resonance at g ∼ 2 (Figure S37). However, decreasing the

temperature to 109 K revealed the appearance of a strong

signal at g_iso= 2.231 for 1[BF₄]. A smaller isotropic signal is

observed in both low-temperature spectra with pyridine, but

the resonance at g ∼ 2.22 for 2[BF₄] is more intense relative to

the smaller isotropic resonance observed for 1[BF₄] (Figure

9). Results for 1[NTf₂] and 2[NTf₂] mimic the results for

1[BF₄] and 2[BF₄] (Figures S38 −S41 and S43 −S46). Overall,

the diﬀerences in pyridine binding are again consistent with

the decreased steric proﬁle of SMe and the increased Lewis

acidity of Ni in 2⁺, as has been described in pyridine binding

studies with diﬀerent singly oxidized Ni complexes with redox-

active salen ligands.⁶¹Furthermore, the Ni(III) EPR

resonances for the pyridine coordinated spectra indicate g_⊥>

g_∥, suggesting that coordination involves the axial d_z²orbital on

Ni (Table 5). In general, this is well precedented for square

planar complexes,^52,49,63although exceptions do exist.⁶⁴

The EPR results suggest that the singly oxidized complexes

undergo coordination-induced valence tautomerism according

to the following expression (eq 1):⁶¹

Figure 11. UV−vis pyridine titration studies of 2[NTf₂] in THF at

298 K. Inset: change in the molar absorptivity of the peak at 399 nm.

The solid line represents the nonlinear least-squares ﬁt for 1:1

complex formation.

completely rule out that more than one pyridine binds to the

metal at lower temperatures, especially in the presence of a

large excess of pyridine, DFT calculations corroborate our

UV−vis titration and suggest that only one pyridine binds.

Notably, attempts to optimize structures of 2⁺with two

pyridine ligands bound to Ni resulted in one of the pyridine

ligands being expelled out of the coordination sphere to form a

ﬁve-coordinate square pyramidal Ni complex with Ni−pyridine

distances of 2.183 and 4.553 Å (Figure S55).

1

Motivated by the paramagnetically shifted H NMR spectra

and magnetic susceptibility results for doubly oxidized

1[OTf]₂, 2[OTf]₂, and 2[NTf₂]₂(Table 4), EPR spectro-

scopic data were collected on these compounds to determine if

I

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

EPR signals associated with their unpaired spins could be

observed. In this context, we note that high-spin S = 1 Ni

complexes with metal-centered spin density often have signals

that are too broad to be observed without a high-ﬁeld EPR

(HFEPR) spectrometer.⁶⁵⁻⁶⁷However, room-temperature

EPR data collected in 2-MeTHF/DCM (1:3) showed isotropic

signals centered at g ∼ 2. Subsequent low-temperature data

collected at 109 K revealed broad features at g_iso= 2.016,

2.014, and 2.013 for 1[OTf]₂, 2[OTf]₂, and 2[NTf₂]₂,

respectively, with some hyperﬁne structure (Figure 12 and

Density Functional Theory Calculations. DFT calcu-

lations were used to evaluate the electronic structure and spin

density variations of 1 and 2 upon stepwise oxidation and in

the presence of axially coordinating ligands. All gas-phase

geometry optimizations were performed using B3LYP/6-

31G(d,p). The calculated bond distances and angles are in

relatively good agreement with the experimental structures and

follow the same general trends (Table S5). For example, the

calculated Ni−N distances in the optimized structure of

1[BF₄] of 1.837 and 1.838 Å are very close to the experimental

values of 1.853(3) and 1.845(3) Å (Table 1). Likewise, the

calculated Ni−N bond distances in doubly oxidized 1[OTf]₂

increase to 1.999 and 2.008 Å, as observed in the experimental

structure (1.992(2) and 2.002(3) Å; Table 1).

Calculations on the doubly oxidized complexes suggest that

the triplet ground states (S = 1) are lower in energy compared

to the singlet ground states (S = 0), corroborating

experimental observations of unpaired spin in the doubly

oxidized complexes. The triplet states calculated for 1[OTf]₂

and 2[OTf]₂, for example, were 18 and 23 kcal/mol lower in

energy than their respective closed-shell singlet states.

Furthermore, we saw no evidence suggesting a more stable

open-shell singlet state for 1[OTf]₂and 2[OTf]₂, as indicated

by minimal spin contamination of the singlet and S²being

close to zero.⁷⁰However, in the case of 2[NTf₂]₂, the energy

of the triplet is calculated to be within DFT error of the higher

energy open-shell singlet (0.6 kcal/mol). For reference, the

error in calculated singlet−triplet gaps using conventional DFT

is typically around 2−3 kcal/mol.^71,72The thermally accessible

singlet state for 2[NTf₂]₂may account for its decreased

solution magnetic moment compared to 1[OTf]₂and 2[OTf]₂

(Table 4) and observed temperature dependence (i.e.,

increasing spin value with decreasing temperature).

The calculations for the singly oxidized complexes 1[BF₄]

and 2[BF₄] revealed that the S = 1/2 spin density is primarily

ligand-centered and delocalized over the π-conjugated triaryl

framework (Figure 13). In order to determine if the metal···F−

BF₃⁻interactions had any eﬀect on spin density, especially with

the more exposed axial coordination site in 2⁺, calculations

were performed with BF₄⁻located at diﬀerent positions around

the metal complexes. Indeed, a small amount of spin density

was delocalized from Ni onto one of the ﬂuorine atoms on

Figure 12. EPR spectra of 1[OTf]₂and 2[NTf₂]₂in 2-MeTHF/

DCM (1:3) at 109 K (g_iso= 2.016 and 2.013, respectively).

Experimental data (black) and simulated trace (red).

Figure S49). Additionally, no half-ﬁeld triplet signal implicating

an organic diradical was observed in the g = 4 region.

Simulations and least-squares ﬁts of the low-temperature EPR

data corroborated the high-spin conﬁguration and required

modeling with small zero-ﬁeld splitting parameters of D = 2.9

× 10⁻⁴, 7.3 × 10⁻⁴, and 7.2 × 10⁻⁴cm⁻¹and E = 1.8 × 10⁻⁴,

9.9 × 10⁻⁵, and 8.9 × 10⁻⁵cm⁻¹for 1[OTf]₂, 2[OTf]₂, and

2[NTf₂]₂, respectively.

Collectively, these results closely match EPR data reported

previously for bimetallic biradical complexes containing two

redox-active ligand centers separated by a structural linker.^68,69

It has been shown that the linkers in these systems serve to

isolate and spatially localize the two radicals on the two

diﬀerent metallic fragments, thereby leading to very weak

electron coupling and remarkably small zero-ﬁeld splitting on

the order of 10⁻⁴cm⁻¹.^68,69By way of comparison, it remains

unclear as to why the radicals would be so localized in our

complex to give similar EPR spectra, especially when

considering the signiﬁcant diﬀerences in calculated spin

density, electronic structure, and corroborating X-ray absorp-

tion spectroscopy results described in the following sections.

Additional experiments to account for these EPR data are

ongoing.

−

BF₄when positioned close to Ni (Figure S54), but no

obvious change in spin density was observed on the L2 ligand.

A similar result was observed in calculations on a hypothetical

structure of 2-py[BF₄] with pyridine bound to Ni in the axial

position; a small amount of spin density was transferred from

Ni to the nitrogen on pyridine, but there was no obvious

change in calculated spin density on the conjugated L2

backbone (Figure 13). Together, these results corroborate the

additional structure in the low-temperature EPR spectrum of

2[BF₄] in 2-MeTHF/DCM (Figure 10). The minimum

observed at 3125 G in the EPR spectrum of 2-py[BF₄] splits

into two peaks in the spectrum of 2[BF₄], consistent with

hyperﬁne splitting due to coupling with one ¹⁹F nucleus on

−

BF₄.

We next performed the same spin density calculations for

the doubly oxidized 1²⁺and 2²⁺with both triﬂate and

triﬂimidate counteranions. Interestingly, the optimized DFT

structure of 2[NTf₂]₂resulted in the triﬂimidate counterion

binding to Ni in contrast to the experimental XRD structure

that shows no such binding. The Ni−O distances decreased

from 2.794(2) and 2.847(3) Å in the XRD structure of

J

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Figure 13. Spin density plots for singly and doubly oxidized complexes of 2. Spin density isovalues are shown at 0.004 for comparison between

each species. Hydrogen atoms on each complex were omitted from the ﬁgure.

Figure 14. MO correlation diagram for neutral and oxidized complexes of 2 with diﬀerent ancillary ligands and anions. Data shown for the oxidized

complexes are from DFT-optimized structures with the exception of 2[NTf₂]₂, which are from a calculated structure constrained to match the XRD

coordinates to prevent NTf₂⁻coordination to Ni. Select Kohn−Sham orbitals calculated for 2 are shown on the left and color-coded in the diagram

to show how they transform across the series (metal-based b₁= blue, a₁= red; ligand-based b₂and a₂= gold). Mulliken symbol assignments were

made by assuming C_2vpoint group symmetry and using the coordinate system shown. The principal C₂-axis used to assign the Mulliken symbols is

labeled as y.

2[NTf₂]₂to 2.171 and 2.223 Å in the optimized DFT structure

(Table S5). To compare the spin density diﬀerences between

the two structures, a DFT calculation was performed on

2[NTf₂]₂with the structure constrained to match the

experimentally derived coordinates from the XRD data (Figure

13). Overall, the DFT calculations revealed signiﬁcant

diﬀerences in spin density that depend on coordination of

the counteranion to Ni. In the absence of axial coordination

with triﬂimidate in 1[NTf₂]₂and 2[NTf₂]₂(as observed in the

experimental solid-state structures), the excess spin density

localizes almost exclusively on the ligand (Figure 13).

However, in stark contrast, triﬂate coordination to Ni in

1[OTf]₂and 2[OTf]₂causes the ligand-centered spin density

to localize on the metal and metal-bound ligand atoms.

Likewise, spin density calculations on the optimized gas-phase

structure of 2[NTf₂]₂with bound triﬂimidate also show spin

localization on the metal, albeit to a lesser extent than that

observed for 2[OTf]₂. A similar calculation on 2[OTf]₂with

the structure constrained to match the coordinates obtained

from XRD studies was performed, but no remarkable

diﬀerences were observed when compared to the optimized

structure (Figure 13). Overall, the qualitative diﬀerences in

−

calculated OTf⁻and NTf₂metal binding are consistent with

K

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

Figure 15. Top left (A): Comparison of normalized S K-edge XAS data for 2 (black), 2[BF₄] (red), and 2[NTf₂]₂(blue). Bottom left (B):

Comparison of S K-edge XAS data for 2[NTf₂]₂(blue) and 2[OTf₂]₂(gold). Right: Qualitative correlation diagram with select MOs for 2, 2[BF₄],

and 2[NTf₂]₂. MO labels correspond to those used in Figure 14.

empirical observations from our XRD and NMR studies, as

well as XAS results described in the following section.

calculated spin density localizes on Ni upon axial triﬂate

coordination, as shown in Figure 13. The two unpaired

electrons are located in the metal-based b₁and a₁MOs in

2[OTf]₂, whereas they are located in the ligand-based b₂and

a₂MOs in 2[NTf₂]₂.

To understand the underlying relationship between the

unpaired spin density and the associated electronic structure,

we analyzed the calculated molecular orbitals (MOs) for 1 and

2 as a function of oxidation state and coordinating ancillary

ligands. MO correlation diagrams for both series of complexes

with 1 and 2 showed similar variations, and a representative

MO diagram is provided by the L2 complexes in Figure 14.

Beginning with the description of closed-shell 2, the frontier

MOs most relevant to our analysis are those labeled as b₁

(LUMO), b₂(HOMO), a₂(HOMO−1), and a₁(HOMO−2).

MO b₁is assigned as Ni−N and Ni−S σ* with the Ni 3d_xy,

whereas the lower energy a₁in 2 is best described as

Sulfur K-Edge XAS Studies. To provide a better

comparison between experiment and theory using deﬁnitively

known and well-deﬁned solid-state structures, S K-edge X-ray

absorption spectroscopy data were collected on solid samples

of 2 and 3, as well as oxidized salts of 2 (Figure 15). S K-edge

XAS, in general, measures dipole-allowed S 1s → np

transitions,^73,74and transitions to unoccupied metal−sulfur

antibonding MOs can be observed in the pre-edge region of

the XAS spectra when they contain appreciable S 3p character

mixing in the antibonding wave function. Notably, d⁸square

planar Ni complexes with dithiolenes and other sulfur ligands

have been routinely studied using S K-edge XAS to assign

oxidation state and quantify related electronic structure

variations.^73,75,76

The S K-edge XAS spectra of 2 and 3 revealed only subtle

diﬀerences (Figure S6), and the spectrum of 2 in Figure 15A is

representative of both spectra. A pre-edge shoulder was

observed at 2473.2 eV (as deﬁned by analysis of the second

derivative trace), and several higher energy features were

located at 2474.1 and 2475.1 eV on top of the rising edge. The

ﬁrst pre-edge shoulder at 2473.2 eV was assigned as 1s → Ni−

S σ* (b₁in Figures 14 and 15), consistent with previous

studies and DFT calculations described in the previous

section.^75,77Higher energy features are assigned to transitions

involving primarily ligand-based MOs containing S character.

Oxidation of 2 to 2[BF₄] caused the pre-edge feature to

decrease in energy to 2472.9 eV and become more resolved

with respect to the adjacent feature, consistent with the

decrease in calculated b₁MO energy in Figure 14. However,

no additional pre-edge transitions are observed in the singly

oxidized spectra despite the generation of an additional

2

nonbonding Ni 3d_z. MOs b₂and a₂are primarily ligand-

centered π-orbitals with smaller amounts of Ni 3d character.

As shown in Figure 14, 2 undergoes a ligand-centered

oxidation by removing an electron from the HOMO (b₂) to

form 2[BF₄]. The same occurs with 2-py[BF₄], but the

2

nonbonding d_zin a₁becomes chemically engaged with

pyridine in the axial coordination site. This raises the energy

of the a₁MO with respect to the HOMO and appears to

account for why the EPR data for 2-py[BF₄] and 1-py[BF₄]

begin to show electron spin on Ni in the presence of

coordinating ligands, especially when cooled to 109 K.

Like the singly oxidized complexes, there are distinguishing

diﬀerences in MO energies for the doubly oxidized complexes

that depend on the axial coordination ability of the

counteranions. Calculations on the XRD structure of 2[NTf₂]₂

(i.e., no axial coordination) revealed that the second electron is

removed from the ligand-centered a₂MO to give rise to a

triplet electron conﬁguration with an ∼1 eV gap between the

singly occupied α b₂MO and the next singly occupied orbital.

However, triﬂate coordination in 2[OTf]₂raises the energy of

2

the a₁MO (3d_z) and simultaneously decreases the energy of

the b₁MO (3d_xy). This appears to account for why the

L

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

electron hole. This indicates that the electron was removed

from an MO lacking signiﬁcant S 3p character, which is

corroborated by the DFT calculations that suggest the electron

is removed from the ligand-based b₂MO, as shown in Figures

14 and 15.

switching from Ni^II(L^•) to Ni^III(L) upon pyridine binding to

the metal. Similar eﬀects were observed with BF₄, as shown in

−

the low-temperature EPR spectra of 2[BF₄]. DFT calculations

showed that these experimental observations align well with

recent electronic structure investigations on related square

planar Co complexes⁷⁸by revealing that axial coordination

Further oxidation of singly oxidized 2[BF₄] to doubly

oxidized 2[NTf₂]₂yields a S K-edge XAS spectrum with the

same general proﬁle as 2[BF₄]. The only signiﬁcant diﬀerence

is that the normalized pre-edge peak intensities in the spectrum

of 2[NTf₂]₂are lower due to additional high-energy transitions

from sulfur in the triﬂimidate anion that aﬀect the data

normalization (see the full spectrum in Figure S10). In

contrast, the S K-edge spectrum of 2[OTf]₂reveals a

signiﬁcant diﬀerence when compared to 2[NTf₂]₂(Figure

15B): the pre-edge peak at 2472.9 eV in the spectrum of

2[NTf₂]₂is no longer present in the spectrum of 2[OTf]₂.

This feature is either pushed to higher energy, which may

account for the increased intensity of the second feature at

2474.0 eV in 2[OTf]₂, or the intensity is diminished by the

decreased S 3p character in the associated MO. Both are

plausible and can be rationalized by the axial coordination of

the triﬂate anions. Triﬂate binding causes the Ni−S bonds to

elongate by ca. 0.2 Å, which reﬂects decreased covalency in

Ni−S bonds. DFT calculations are consistent with this

assessment and reveal that orbital mixing between the SMe

groups and the Ni 3d_xydecreases in 2[OTf]₂. The α and β a₁

orbitals calculated for 2[OTf]₂, which are best described as

Ni−OTf σ*, also have negligible S 3p character. Collectively,

these solid-state XAS results corroborate the electronic

structure variations observed in the DFT calculations.

2

chemically engages the Ni 3d_zand raises the energy of the

associated antibonding MO to access more Ni(III) character.

Paramagnetic NMR resonances and magnetic susceptibility

data revealed the presence of unpaired spin in the doubly

oxidized complexes of 1 and 2, and DFT calculations showed

how these seemingly similar open-shell electronic conﬁgu-

rations diﬀer depending on the identity of the coordinating

anions. With bound triﬂate in six-coordinate 1[OTf]₂and

2[OTf]₂, the complexes are best described as containing a

closed-shell o-diiminoquinone ligand with high-spin Ni(II),

whereas open shell calculations on four-coordinate 2[NTf₂]₂

indicated that this complex is best described as Ni(II) with

diradical character localized on the ligand. Both of these

electronic conﬁgurations can be found within the various

resonant forms shown in Figure 16, but they diﬀer in that their

CONCLUSION

■

In summary, we have described the synthesis, electrochemistry,

and structures of Ni complexes with three tetradentate ligands

containing ﬂanking redox-innocent NMe₂and SMe donor

substituents. Ni complexes containing the triaryl o-phenyl-

enediamine-derived ligands L1 and L2 revealed two fully

reversible ligand-centered redox events. Metal-centered Ni²⁺/

Ni¹⁺redox couples were observed when scanning to more

negative potentials and were fully reversible in 1, whereas the

SMe substituents in 2 yielded irreversible reductions and the

appearance of new redox features in the CV trace. Replacing

the phenylene backbone in L2 with an ethylene linker in L3

attenuated the reversible ligand-centered redox activity in the

CV of 3, but quasi-reversible waves were still observed,

indicating that the ﬂanking aryl groups still support some

ligand-centered redox activity in the absence of the conjugated

phenylene backbone. Chemically oxidizing 1 and 2 allowed the

structures and electronic properties of the electrochemically

generated products to be characterized by single-crystal XRD,

UV-vis-NIR, NMR, EPR, and S K-edge XAS spectroscopies

with diﬀerent counteranions. The XRD data suggested that the

redox events were primarily ligand-centered and proceed via

formation of the o-diiminosemiquinone radical followed by the

fully oxidized o-diiminoquinone, as is often reported for o-

phenylenediamine and derivatives there-of.

Figure 16. Selected resonance structures of doubly oxidized 1 (X =

NMe₂) and 2 (X = SMe) with emphasis on structures that show how

radicals can delocalize onto the ﬂanking aryl groups.

relative weighting depends on the presence or absence of axial

coordination to Ni. From the simplest perspective, axial

coordination in these doubly oxidized complexes causes the

ligand-centered spin density to collapse onto Ni due to

reshuﬄing of the metal-based molecular orbitals, consistent

with predictions based on ligand ﬁeld theory.

Overall, the electronic structure studies on Ni complexes

containing our triaryl tetradentate ligands with redox-innocent

donor groups have revealed notable diﬀerences with respect to

other complexes that either (1) lack ﬂanking aryl groups in

conjugation with the o-phenylenediamine backbone or (2)

contain redox-noninnocent metal donor groups, such as those

described by Thomas and Wieghardt (Chart 2 and Figure

8).^17,20We anticipate that these results will provide the

foundation necessary to evaluate electronic structure variations

in complexes that also participate in metal−ligand cooperative

binding with diﬀerent substrates, as we have shown recently

with Ru complexes containing L2.³¹The redox chemistry,

ligand noninnocence, electronic structure, and reactivity of

these and related complexes will be described in several

forthcoming reports.

We demonstrated how ancillary ligands with diﬀerent

−

coordinating abilities (pyridine, BF₄, OTf⁻, and NTf₂) aﬀect

the electronic structure of oxidized salts of 1 and 2. As has

been observed for other Ni complexes with redox-active

ligands,⁵⁸⁻⁶²treating the singly oxidized complexes 1[BF₄]

and 2[BF₄] with excess pyridine resulted in coordination-

induced valence tautomerism with the electronic structure

M

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

1

13.1 g (43%). H NMR (400 MHz, CDCl₃): δ 2.49 (s, 6H, S

CH₃), 7.09 (dd, J = 1.21 Hz, J = 7.9 Hz, 2H, ArH), 7.18 (td, J = 1.5

Hz, J = 7.8 Hz, 2H, ArH), 7.27 (m, 4H, ArH), 8.43 (s, 2H, N

CH). N,N′-Bis[2-(methylthio)phenyl]ethanediylidene was dried over

P₂O₅for 12 h before proceeding to the next step.

EXPERIMENTAL SECTION

■

General Considerations. All reactions were performed under an

atmosphere of N₂or Ar using glovebox or standard Schlenk

techniques, unless stated otherwise. Pentane, toluene, Et₂O, THF,

MeCN, and CH₂Cl₂were dried and degassed using a Pure Process

Technologies Solvent Puriﬁcation System. Benzene was distilled from

sodium benzophenone and stored over 3 Å molecular sieves.

Deuterated solvents were deoxygenated by three freeze−pump−

thaw cycles and stored over 3 Å molecular sieves. N,N′-Bis[2-

(methylthio)phenyl]-1,2-diaminobenzene (L2) was prepared as

described previously.³¹All other chemicals were purchased from

commercial vendors and used as received.

N,N′-bis[2-(methylthio)phenyl]ethanediylidene (13.0 g, 43.3

mmol) was dissolved in THF (250 mL) under an atmosphere of

N₂and cooled to −10 °C. NaBH₄(3.27 g, 86.5 mmol) in THF (50

mL) was added dropwise over 30 min. The reaction mixture was

allowed to slowly warm to RT. The yellow/orange solution slowly

turned colorless, and a small amount of eﬀervescence was observed.

The mixture was stirred for 12 h and then evaporated to dryness

under a vacuum to reveal a white solid that immediately turned yellow

upon exposure to air. The yellow solid was dissolved in CH₂Cl₂and

washed with DI water. Recrystallization from CH₂Cl₂resulted in

¹H, ¹³C, and ¹¹B NMR data were recorded on a Bruker DRX-400

instrument operating at 400 MHz for H, 100 MHz for ¹³C, and 128

1

MHz for ¹¹B or on a Bruker Avance-500 instrument operating at 500

MHz for ¹H, 126 MHz for ¹³C, and 470 MHz for ¹⁹F. ¹⁹F NMR data

were collected on a Bruker Avance-300 instrument operating at 282

MHz. Chemical shifts are reported in δ units relative to residual

solvent peaks (¹H and ¹³C), BF₃·Et₂O (¹¹B; 0.0 ppm), or 0.05%

C₆H₅CF₃in C₆D₆(¹⁹F; −62.9 ppm). Microanalytical data (CHN)

were collected using an EAI CE-440 Elemental Analyzer in the

University of Iowa Department of Chemistry. IR spectra were

acquired on a Thermo Scientiﬁc Nicolet iS5 instrument using an

attenuated reﬂection accessory. Melting points were determined in

sealed capillaries using a REACH MP device. UV−vis-NIR absorption

data were collected on an Agilent Cary 5000 near-infrared

spectrophotometer. HR-EI mass spectra were recorded on a Waters

GCT Premier Instrument using TOF. Fragment ions (M, molecule;

L, ligand) were assigned based on comparison to calculated natural

abundance isotopic distributions.

1

yellow blocks of L3. Yield: 11.1 g (84%). Mp: 77−80 °C. H NMR

(400 MHz, CDCl₃): δ 2.30 (s, 6H, S−CH₃), 3.51 (s, 4H, N−CH₂),

5.37 (br s, 2H, N−H), 6.66−6.73 (m, 4H, Ar−H), 7.20 (td, J = 1.7

Hz, J = 7.9 Hz, 2H, Ar−H), 7.41 (dd, J = 1.5 Hz, J = 8.1 Hz, 2H, Ar−

H). ¹³C NMR (100 MHz, CDCl₃): δ 18.28 (S−CH₃), 43.13 (N−

CH₂), 110.38 (Ar−C), 117.67 (Ar−C), 120.49 (Ar−C), 129.61 (Ar−

C), 134.37 (Ar−C), 148.01 (Ar−C). IR (ATR, cm⁻¹): 3326 m, 3057

m, 3017 vw, 2988 vw, 2958 vw, 2918 m, 2845 m, 1916 w, 1876 w,

1837 vw, 1580 s, 1494 vs, 1482 s, 1449 s, 1429 s, 1309 s, 1282 m,

1240 s, 1214 m, 1172 w, 1135 m, 1088 w, 1037 s, 965 m, 956 m, 919

m, 839 w, 821 m, 747 vs, 737 vs, 717 s, 692 m.

Ni(^MeNNNN^Me)-DAB, 1. A 50 mL Schlenk ﬂask was charged with

NiBr₂(0.379 g, 1.73 mmol) and L1 (0.600 g, 1.73 mmol) in 25 mL of

THF. The reaction was heated to reﬂux overnight, and a mint green

solid slowly formed. After cooling to RT, the mixture was ﬁltered and

the solid was washed with Et₂O (3 × 20 mL). A solution of NaO^tBu

(0.333 g, 3.46 mmol) in THF (25 mL) was transferred via cannula

into the Schlenk ﬂask containing the solid. The color immediately

turned dark green/brown, and the mixture was stirred for 12 h. The

mixture was ﬁltered, and the remaining gray precipitate was washed

with THF (3 × 10 mL). The combined ﬁltrate and washings were

evaporated to dryness under a vacuum to yield a dark green solid.

Cooling concentrated THF solutions or layering with Et₂O yielded

single crystals. Yield: 0.42 g (60%). Anal. Calcd for C₂₂H₂₄N₄Ni: C,

65.54; H, 6.00; N, 13.90. Found: C, 65.66; H, 5.88; N, 13.68. Mp:

>250 °C. EI-HRMS (70 eV) m/z: [M] calcd for C₂₂H₂₄N₄Ni

N,N′-Bis[2-(N,N-dimethylamine)phenyl]-1,2-diaminobenzene,

H₂(^MeNNNN^Me)-DAB, L1. Pd₂(dba)₃(1.70 g, 1.85 mmol), rac-BINAP

(1.73 g, 2.80 mmol, 7.5 mol %), NaO^tBu (10.67 g, 111.0 mmol),

degassed water (1.0 mL), and o-phenylenediamine (4.0 g, 37.0 mmol)

were added to dry toluene (250 mL). The reaction mixture was

heated to reﬂux under N₂, and a solution of 2-bromo-N,N′-

dimethylaniline (14.8 g, 74.0 mmol, 10.7 mL) in toluene (50 mL)

was added dropwise over 6 h. The reaction was heated to reﬂux for 72

h under a N₂atmosphere. After cooling to RT, the mixture was

treated with a saturated aqueous solution of NH₄Cl immediately upon

being exposed to air. The mixture was ﬁltered through a pad of Celite

and washed with toluene until the ﬁltrate was colorless. The organic

phase was separated, concentrated under reduced pressure, and

puriﬁed by column chromatography on 70−230 mesh silica gel (2:98

Et₂O/hexane) to aﬀord a yellow/white solid. The solid was

recrystallized from MeCN to aﬀord colorless blocks. Yield: 7.94 g

(62%). Mp: 91−94 °C. Anal. Calcd for C₂₂H₂₆N₄: C, 76.26; H, 7.56;

N, 16.17. Found: C, 76.33 H, 7.46; N, 16.03. HR-EIMS (70 eV) m/z:

[M] calcd for C₂₂H₂₆N₄346.215; found 346.218. ¹H NMR (400

MHz, CDCl₃): δ 2.58 (s, 12H, N−CH₃), 6.65 (br s, 2H, N−H), 6.80

(td, J = 1.6 Hz, J = 7.6 Hz, 2H, Ar−H), 6.95 (t, J = 7.4, 2H, Ar−H),

7.00 (dd, J = 3.3 Hz, J = 6.0 Hz, 2H, Ar−H), 7.05 (d, J = 8.5 Hz, 2H,

Ar−H), 7.08 (dd, J = 1.1 Hz, J = 7.8 Hz, 2H, Ar−H), 7.38 (dd, J = 3.6

Hz, J = 5.8 Hz, 2H, Ar−H). ¹³C NMR (100 MHz, CDCl₃): δ 44.16

(N−CH₃), 114.14 (Ar−C). 119.50 (Ar−C), 119.78 (Ar−C), 120.50

(Ar−C), 122.80 (Ar−C), 124.38 (Ar−C), 135.19 (Ar−C), 139.11

(Ar−C), 142.39 (Ar−C). IR (KBr, cm⁻¹): 3318 m, 3267 m, 3037 m,

2981 m, 2940 m, 2863 w, 2824 m, 2784 m, 1588 s, 1526 s, 1503 s,

1491 m, 1451 m, 1433 m, 1410 m, 1333 m, 1289 s, 1275 s, 1211 w,

1182 m, 1153 s, 1111 m, 1092 m, 1043 s, 941 s, 900 vw, 850 vw, 737

s.

1

402.135; found 402.136. H NMR (400 MHz, THF-d₈): δ 3.05 (s,

12H, N−CH₃), 6.27 (t, J = 7.2 Hz, 2H, Ar−H), 6.34 (dd, J = 3.0 Hz, J

= 6.2 Hz, 2H, Ar−H), 6.85 (t, J = 7.5 Hz, 2H, Ar−H), 7.05 (d, J = 8.2

Hz, 2H, Ar−H), 7.18 (dd, J = 3.7 Hz, J = 6.0 Hz, 2H, Ar−H), 7.44 (d,

J = 8.1 Hz, 2H, Ar−H). ¹³C NMR (100 MHz, THF-d₈): δ 50.34 (N−

CH₃), 113.27 (Ar−C), 114.05 (Ar−C), 114.24 (Ar−C), 116.52 (Ar−

C), 119.47 (Ar−C), 128.60 (Ar−C), 147.24 (Ar−C), 147.78 (Ar−

C), 149.58 (Ar−C). IR (ATR, cm⁻¹): 3049 w, 3008 w, 2977 w, 2917

w, 1591 m, 1572 w, 1550 m, 1480 s, 1436 m, 1396 w, 1359 s, 1323 m,

1294 m, 1276 m, 1239 w, 1208 m, 1179 m, 1135 w, 1113 w, 1090 w,

1061 w, 1041 m, 991 m, 923 m, 902 w, 883 m, 867 w, 833 w, 784 vw,

749 m, 739 m, 718 s. UV−vis (CH₂Cl₂): 210 nm (ε = 63700 M⁻¹

cm⁻¹), 278 nm (ε = 17300 M⁻¹cm⁻¹), 355 nm (ε = 29400 M⁻¹

cm⁻¹), 557 nm (ε = 344 M⁻¹cm⁻¹), 702 nm (ε = 409 M⁻¹cm⁻¹).

Ni(^MeSNNS^Me)-DAB, 2. A 25 mL Schlenk ﬂask was charged with

NiBr₂(0.186 g, 0.85 mmol), L2 (0.300 g, 0.85 mmol), and THF (12

mL). The mixture was heated to reﬂux for 12 h, resulting in a mint

green precipitate. After cooling to RT, the mixture was ﬁltered and the

green solid was washed with Et₂O (3 × 20 mL). A solution of

NaO^tBu (0.164 g, 1.7 mmol) in THF (20 mL) was transferred via

cannula into the Schlenk ﬂask containing the solid. The color

immediately changed to dark green, and the mixture was stirred

overnight. The mixture was ﬁltered, and the remaining gray

precipitate was washed with THF (3 × 10 mL). The combined

ﬁltrate and washings were evaporated to dryness under a vacuum to

reveal a tacky, dark green solid. Single crystals of 2 were grown by

vapor diﬀusion with benzene/Et₂O. Yield: 0.19 g (54%). Anal. Calcd

for C₂₀H₁₈N₂NiS₂: C, 58.70; H, 4.43; N, 6.85. Found: C, 58.89; H,

4.40; N, 6.84. Mp: 171 °C. EI-HRMS (70 eV) m/z: [M] calcd for

N,N′-Bis[2-(methylthio)phenyl]-1,2-diaminoethane,

H₂(^MeSNNS^Me)-DAE, L3. 2-(Methylthio)aniline (14.0 g, 100 mmol,

12.6 mL) was added to methanol (50 mL) in air. The mixture was

heated to reﬂux, and 40% glyoxal in water (4.0 g, 69.0 mmol, 7.8 mL)

was added dropwise. The reaction mixture was heated to reﬂux for 2

h, which caused slow formation of an orange precipitate. The reaction

mixture was cooled to 0 °C, and the orange solid was collected by

ﬁltration. Recrystallization from boiling pyridine yielded orange

needles of N,N′-bis[2-(methylthio)phenyl]-ethanediylidene. Yield:

N

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

C₂₀H₁₈N₂NiS₂408.026; found 408.029. ¹H NMR (500 MHz, C₆D₆):

δ 1.71 (s, 6H, S−CH₃), 6.39 (td, J = 1.0 Hz, J = 7.2 Hz, 2H, Ar−H),

6.79 (dd, J = 1.4 Hz, J = 7.6 Hz, 2H, Ar−H), 6.88 (dd, J = 3.3 Hz, J =

6.0 Hz, 2H, Ar−H), 6.99 (td, J = 1.5 Hz, J = 7.8 Hz, 2H, Ar−H), 7.80

(d, J = 8.5 Hz, 2H, Ar−H), 7.84 (dd, J = 3.4 Hz, J = 5.9 Hz, 2H, Ar−

H). ¹³C NMR (126 MHz, C₆D₆): δ 24.16 (S−CH₃), 114.36 (Ar−C),

115.50 (Ar−C), 115.71 (Ar−C), 118.52 (Ar−C), 120.83 (Ar−C),

131.38 (Ar−C), 132.05 (Ar−C), 147.67 (Ar−C), 156.03 (Ar−C). IR

(ATR, cm⁻¹): 3045 w, 3012 w, 2914 vw, 1581 m, 1560 m, 1544 m,

1476 s, 1455 vs, 1432 s, 1410 m, 1345 m, 1330 s, 1300 s, 1239 m,

1206 m, 1172 w, 1153 w, 1124 w, 1045 w, 1033 m, 1026 m, 961 m,

918 m, 730 vs. UV−vis (CH₂Cl₂): 215 nm (ε = 108000 M⁻¹cm⁻¹),

230 nm (ε = 82900 M⁻¹cm⁻¹), 295 nm (ε = 46800 M⁻¹cm⁻¹), 349

nm (ε = 53600 M⁻¹cm⁻¹), 598 nm (ε = 475 M⁻¹cm⁻¹), 796 nm (ε

= 366 M⁻¹cm⁻¹).

640 s, 631 m, 590 m. UV−vis-NIR (CH₂Cl₂): 229 nm (ε = 21300

M⁻¹cm⁻¹), 255 nm (ε = 20400 M⁻¹cm⁻¹), 286 nm (sh, ε = 15700

M

⁻¹cm⁻¹), 398 nm (ε = 11000 M⁻¹cm⁻¹), 547 nm (ε = 2770 M⁻¹

cm⁻¹), 1150 nm (ε = 970 M⁻¹cm⁻¹), 1373 nm (ε = 1580 M⁻¹cm⁻¹),

1689 nm (ε = 1470 M⁻¹cm⁻¹).

[Ni(^MeNNNN^Me)-DAB][NTf₂], 1[NTf₂]. Prepared using the same

procedure described for 1[BF₄] with 1 (0.050 g, 0.120 mmol) and

AgNTf₂(0.047 g, 0.120 mmol). Yield: 0.056 g (67%). Anal. Calcd for

C₂₄H₂₄F₆N₅NiO₄S₂: C, 42.19; H, 3.54; N, 10.25. Found: C, 42.81; H,

3.25; N, 10.04. ¹⁹F NMR (282 MHz, CH₂Cl₂): δ −78.19. UV−vis-

NIR (CH₂Cl₂): 211 nm (ε = 42500 M⁻¹cm⁻¹), 238 nm (sh, ε =

22400 M⁻¹cm⁻¹), 280 nm (ε = 20500 M⁻¹cm⁻¹), 399 nm (ε =

19700 M⁻¹cm⁻¹), 496 nm (sh, ε = 2900 M⁻¹cm⁻¹), 583 nm (ε =

4000 M⁻¹cm⁻¹), 1124 nm (ε = 2000 M⁻¹cm⁻¹), 1340 nm (ε = 3580

M⁻¹cm⁻¹), 1663 nm (ε = 3900 M⁻¹cm⁻¹).

Ni(^MeSNNS^Me)-DAE, 3. A 25 mL Schlenk ﬂask was charged with

NiBr₂(0.215 g, 0.99 mmol), L3 (0.300 g, 0.99 mmol), and THF (12

mL). The reaction was heated to reﬂux overnight, resulting in a mint

green precipitate the next day. After cooling to RT, the mixture was

ﬁltered and the mint green solid was washed with Et₂O (3 × 20 mL).

A solution of NaO^tBu (0.189 g, 1.97 mmol) in THF (20 mL) was

transferred via cannula into the Schlenk ﬂask containing the solid. The

color immediately turned blue, and the mixture was stirred overnight.

The blue mixture was ﬁltered, and the remaining gray solid was

washed with THF (3 × 10 mL). The combined ﬁltrate and washings

were evaporated to dryness under a vacuum to yield a dark blue/green

solid. The solid was dissolved in benzene and layered with Et₂O to

yield single crystals. Yield: 0.15 g (42%). Anal. Calcd for

C₁₆H₁₈N₂NiS₂: C, 53.21; H, 5.02; N, 7.76. Found: C, 53.25; H,

4.90; N, 7.61. Mp: 180−182 °C. EI-HRMS (70 eV) m/z: [M] calcd

for C₁₆H₁₈N₂NiS₂360.026; found 360.024. ¹H NMR (400 MHz,

C₆D₆): δ 1.88 (s, 6H, S−CH₃), 3.33 (br s, 4H, N−CH₂), 6.40 (t, J =

7.1 Hz, 2H, Ar−H), 6.50 (d, J = 8.3 Hz, 2H, Ar−H), 6.86 (d, J = 7.3

Hz, 2H, Ar−H), 7.11 (t, J = 7.5 Hz, 2H, Ar−H). ¹³C (100 MHz,

C₆D₆): δ 24.56 (S−CH₃), 54.53 (N−CH₂), 111.12 (Ar−C), 112.74

(Ar−C), 116.88 (Ar−C), 131.70 (Ar−C), 132.12 (Ar−C), 162.00

(Ar−C). IR (ATR, cm⁻¹): 3197 m, 3123 s, 2964 s, 2933 s, 2877 m,

1577 w, 1480 m, 1460 w, 1445 s, 1407 s, 1342 m, 1283 m, 1200 w,

1178 m, 1137 m, 1067 s, 1048 w, 1033 w, 1018 s, 976 vs, 955 s, 888

w, 868 w, 801 m, 770 vs, 762 vs, 737 s, 708 w, 672 m. UV−vis

(CH₂Cl₂): 240 nm (ε = 40700 M⁻¹cm⁻¹), 284 nm (ε = 23900 M⁻¹

cm⁻¹), 330 nm (ε = 16300 M⁻¹cm⁻¹), 390 nm (ε = 6720 M⁻¹cm⁻¹),

667 nm (ε = 1170 M⁻¹cm⁻¹).

[Ni(^MeSNNS^Me)-DAB][NTf₂], 2[NTf₂]. Prepared using the same

procedure described for 1[BF₄] with 2 (0.050 g, 0.122 mmol) and

AgNTf₂(0.047 g, 0.122 mmol). Yield: 0.069 g (83%). Anal. Calcd for

C₂₂H₁₈F₆N₃NiO₄S₄: C, 38.33; H, 2.63; N, 6.10. Found: C, 38.32; H,

2.20; N, 5.99. ¹⁹F NMR (282 MHz, CH₂Cl₂): δ −78.79. UV−vis-NIR

(CH₂Cl₂): 210 nm (ε = 42000 M⁻¹cm⁻¹), 258 nm (ε = 19500 M⁻¹

cm⁻¹), 289 nm (ε = 19600 M⁻¹cm⁻¹), 399 nm (ε = 11900 M⁻¹

cm⁻¹), 547 nm (ε = 3100 M⁻¹cm⁻¹), 1150 nm (ε = 940 M⁻¹cm⁻¹),

1373 nm (ε = 1500 M⁻¹cm⁻¹), 1692 nm (ε = 1350 M⁻¹cm⁻¹).

[Ni(^MeNNNN^Me)-DAB][OTf]₂, 1[OTf]₂. AgOTf (0.064 g, 0.248

mmol) was added to a stirring solution of 1 (0.050 g, 0.120 mmol)

in CH₂Cl₂(10 mL). The color turned purple and then slowly dark

red. The reaction mixture was stirred for 30 min and then ﬁltered

through Celite. Layering the CH₂Cl₂ﬁltrate with Et₂O resulted in

gold blocks. Yield: 0.053 (60%). Anal. Calcd for C₂₄H₂₄F₆N₄NiO₆S₂:

C, 41.10; H, 3.45; N, 7.99. Found: C, 40.81; H, 3.31; N, 7.67. Mp:

>250 °C. HR-EIMS (70 eV) m/z: [M−2OTf]²⁺calcd for

C₂₂H₂₄N₄Ni 402.135; found 402.129. ¹H NMR (500 MHz,

CD₂Cl₂): δ 2.40, 3.01, 12.73, 14.94, 21.13, 39.44. ¹⁹F NMR (470

MHz, CD₂Cl₂): −38.5. IR (ATR, cm⁻¹): 3376 vw, 2965 w, 1585 w,

1571 w, 1536 w, 1477 m, 1455 w, 1432 w, 1324 w, 1296 vs, 1230 vs,

1211 vs, 1175 vs, 1147 s, 1100 m, 1036 s, 1020 vs, 917 s, 875 m, 858

w, 843 w, 823 m, 767 vs, 741 s, 678 s. UV−vis (CH₂Cl₂): 251 nm (ε

= 12000 M⁻¹cm⁻¹), 546 nm (ε = 9550 M⁻¹cm⁻¹), 581 nm (ε =

9590 M⁻¹cm⁻¹).

[Ni(^MeSNNS^Me)-DAB][OTf]₂, 2[OTf]₂. AgOTf (0.063 g, 0.244 mmol)

was added to a stirring solution of 2 (0.050 g, 0.122 mmol) in CH₂Cl₂

(10 mL). The color turned purple and then eventually to dark blue.

The mixture was stirred for 30 min and then ﬁltered through Celite.

Layering the ﬁltrate with Et₂O resulted in gold blocks. Yield: 0.048 g,

(56%). Anal. Calcd for C₂₂H₁₈F₆N₂NiO₆S₄·CH₂Cl₂: C, 34.87; H,

[Ni(^MeNNNN^Me)-DAB][BF₄], 1[BF₄]. A slurry of AgBF₄(0.024 g,

0.120 mmol) in CH₂Cl₂(5 mL) was added dropwise to a stirring

solution of 1 (0.050 g, 0.120 mmol) in CH₂Cl₂(10 mL). The color

immediately turned dark purple. The mixture was stirred for 30 min

and ﬁltered through Celite. Layering the concentrated ﬁltrate with

Et₂O resulted in dark green needles. Yield: 0.041 g (68%). Anal. Calcd

for C₂₂H₂₄BF₄N₄Ni: C, 53.93; H, 4.94; N, 11.44. Found: C, 53.78; H,

5.14; N, 11.07. Mp: >250 °C. HR-EIMS (70 eV) m/z: [M−BF₄]⁺

calcd for C₂₂H₂₄N₄Ni 402.135; found 402.142. ¹¹B NMR (128 MHz,

CD₂Cl₂): δ −1.36. ¹⁹F NMR (282 MHz, CD₂Cl₂): δ −151.63. IR

(ATR, cm⁻¹): 2964 m, 2878 w, 2687 vw, 1591 vw, 1579 vw, 1563 w,

1528 w, 1483 m, 1428 s, 1367 m, 1335 s, 1297 m, 1284 m, 1203 w,

1190 w, 1175 w, 1158 m, 1096 m, 1007 m, 1031 vs, 958 s, 926 m, 910

m, 875 s, 854 m, 841 m, 814 m, 739 s, 717 s, 691 s, 651 s. UV−vis-

NIR (CH₂Cl₂): 279 nm (ε = 13900 M⁻¹cm⁻¹), 395 nm (ε = 13200

M⁻¹cm⁻¹), 548 nm (ε = 2690 M⁻¹cm⁻¹), 584 nm (ε = 2580 M⁻¹

cm⁻¹), 1130 nm ((ε = 1400 M⁻¹cm⁻¹), 1335 (ε = 2300 M⁻¹cm⁻¹),

1666 nm (ε = 2400 M⁻¹cm⁻¹).

1

2.54; N, 3.54. Found: C, 34.61; H, 2.59; N, 3.76. Mp: >250 °C. H

NMR (400 MHz, CD₂Cl₂): δ 1.85, 3.69, 4.30, 8.13, 25.34, 26.37,

35.97. ¹⁹F NMR (282 MHz, CD₂Cl₂): δ −41.7. IR (ATR, cm⁻¹):

3092 vw, 3086 vw, 3029 vw, 1578 w, 1531 w, 1460 w, 1431 m, 1398

w, 1330 m, 1311 s, 1302 s, 1262 w, 1230 s, 1205 vs, 1159 vs, 1136 s,

1059 m, 1025 m, 1014 vs, 980 s, 974 s, 825 m, 799 m, 766 s, 760 vs,

724 m, 716 s, 663 m, 665 m, 633 vs. UV−vis (CH₂Cl₂): 230 nm (ε =

28400 M⁻¹cm⁻¹), 255 nm (ε = 25800 M⁻¹cm⁻¹), 414 nm (ε = 5500

M

⁻¹cm⁻¹), 581 nm (ε = 14200 M⁻¹cm⁻¹), 604 nm (ε = 14000 M⁻¹

cm⁻¹).

[Ni(^MeSNNS^Me)-DAB][NTf₂]₂, 2[NTf₂]₂. AgNTf₂(0.095 g, 0.244

mmol) was added to a stirring solution of 2 (0.050 g, 0.122 mmol) in

CH₂Cl₂(10 mL). The color turned purple and then slowly to dark

blue-red. The mixture was stirred for 30 min and ﬁltered through

Celite. Layering the ﬁltrate with Et₂O yielded green needles. Yield:

0.100 g (86%). Anal. Calcd for C₂₄H₁₈F₁₂N₄NiO₈S₆: C, 29.73; H,

1.87; N, 5.78. Found: C, 29.78; H, 1.60; N, 5.70. Mp: 140−142 °C.

¹H NMR (400 MHz, CD₂Cl₂): δ 5.66, 6.65, 8.04, 15.77. ¹⁹F NMR

(282 MHz, CD₂Cl₂): −72.18. IR (ATR, cm⁻¹): 3121 vw, 3083 vw,

3054 vw, 2965 w, 2878 vw, 1574 m 1529 w, 1510 vw, 1464 m, 1431

m, 1394 m, 1384 m, 1336 s, 1325 s, 1315 s, 1226 m, 1196 s, 1179 vs,

1126 vs, 1055 s, 980 m, 966 m, 957 m, 884 w, 854 m, 787 s, 775 s,

765 vs, 736 s, 719 m, 651 s. UV−vis (CH₂Cl₂): 258 nm (ε = 21200

[Ni(^MeSNNS^Me)-DAB][BF₄], 2[BF₄]. Prepared using the same

procedure described for 1[BF₄] with 2 (0.050 g, 0.122 mmol) and

AgBF₄(0.024 g, 0.122 mmol). Yield: 0.031 g (52%). Anal. Calcd for

C₂₀H₁₈BF₄N₂NiS₂: C, 48.43; H, 3.66; N, 5.65. Found: C, 48.07; H,

3.84; N, 5.36. Mp: 219−221 °C. ¹¹B NMR (128 MHz, CD₂Cl₂): δ

−1.12. ¹⁹F NMR (282 MHz, CD₂Cl₂): δ −149.48. IR (ATR, cm⁻¹):

3056 w, 2965 m, 2935 m, 2878 w, 1578 vw, 1558 w, 1531 m, 1462 m,

1420 s, 1357 m, 1312 s, 1272 m, 1219 w, 1196 m, 1173 m, 1157 m,

1029 vs, 974 s, 920 m, 875 m, 858 m, 740 vs, 716 s, 702 m, 663 m,

O

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

M⁻¹cm⁻¹), 405 nm (ε = 5580 M⁻¹cm⁻¹), 598 nm (ε = 8520 M⁻¹

cm⁻¹).

K were conducted in a similar fashion using a third S = 1/2 spin

system to account for spin localization on the ¹⁹F atom.

Single-Crystal X-ray Diﬀraction Studies. Single crystals

retrieved from acetonitrile [H₂(^MeNNNN^Me)-DAB, L1], THF/Et₂O

(1), benzene/Et₂O (2 and 3), and CH₂Cl₂/Et₂O (1[BF₄], 1[OTf]₂,

2[BF₄], 2[NTf₂]₂, and 2[OTf]₂) were mounted on a MiTeGen

micromount with ParatoneN oil. The data were collected on a Bruker

Nonius Kappa CCD, with an Apex II charge-coupled-device (CCD)

detector (L1, 1[OTf]₂, 2, 2[NTf₂]₂, and 2[OTf]₂), or on a Nonius

Kappa CCD difractometer (1, 1[BF₄], and 2[BF₄]). The samples

were cooled to 190(1) K by an Oxford Cryostreams 700 low-

temperature device. Both instruments were equipped with a graphite

monochromatized Mo Kα X-ray source (λ = 0.71073). A hemisphere

of data were collected using phi and omega scans. Data collection,

initial indexing, and cell reﬁnement were conducted using the Bruker

Apex II suite (Bruker Nonius Kappa CCD) or using COLLECT and

DENZO/SCALEPACK⁷⁹(Nonius Kappa CCD). The data were

corrected for absorption using redundant reﬂections and the SADABS

program.⁸⁰Structures were solved by direct methods and diﬀerence

Fourier techniques performed in Olex2.⁸¹Hydrogen atom positions

were idealized and allowed to ride on the attached carbon or nitrogen

atoms, with exception of 1, 1[BF₄], 2, and 2[BF₄] where close contact

hydrogen atoms of the aryl groups were allowed to reﬁne. A solvent

mask was applied to the reﬁnement of 2[NTf₂]₂to account for

disordered solvent in the crystal lattice that could not be modeled

satisfactorily.⁸²Disordered counteranions of 1[BF₄], 2[BF₄], and

2[OTf]₂were modeled upon locating the second orientation, and the

two sites were summed together with a site-occupancy factor (SOF)

set at 1.0. The displacement parameters of the disordered components

were constrained to produce satisfactory ellipsoids. HKL reﬂections

with error/esd values 10 were omitted from the models. The ﬁnal

reﬁnement included anisotropic temperature factors on all non-

hydrogen atoms. Structure solution and reﬁnement were performed in

Olex2⁸¹using SHELXT⁸³and SHELXL.⁸⁴Publication ﬁgures were

generated using Mercury CSD 3.10.⁸⁵Data collection and reﬁnement

details are listed in Tables S1 and S2.

Electrochemical Studies. Cyclic voltammetry (CV) experiments

were performed under a N₂atmosphere in a Genesis glovebox using a

CH Instruments CHI660D potentiostat. The electrochemical cell was

purged in the glovebox for 45 min with ultrahigh-purity (UHP) Ar

(99.999%) before each CV experiment was performed. The

electrolyte was 0.1 M (ⁿBu₄N)PF₆(Aldrich, ≥99.0%) in THF with

an analyte concentration of 1 mM. The electrochemical cell consisted

of a glassy carbon working electrode with a platinum wire used as the

counter and quasi reference electrodes. All potentials are reported

versus the Fc⁺/Fc redox couple, measured by adding a small amount

(∼0.005 M) of ferrocene to the electrochemical solutions at the end

of each series of measurements.

EPR Studies. X-Band EPR spectra were recorded on a Bruker

EMX EPR spectrometer equipped with a ﬂow-through cryostat in

conjunction with a Eurotherm B-VT-2000 controller. Samples were

measured in 4 mm O.D. quartz tubes featuring J-Young tops. For low-

temperature measurements, samples were slowly frozen by dipping

the EPR tube containing the sample in and out of liquid nitrogen until

frozen and then quickly transferred into the EPR cavity.

The doubly oxidized compounds were modeled as an S = 1 spin

system with added D and E zero ﬁeld splitting parameters. Initial

ﬁtting followed the same procedure as described above for the singly

oxidized compounds. Upon satisfactory agreement to the exper-

imental data with initial ﬁts, esﬁt was allowed to reﬁne to ﬁnd g values,

A hyperﬁne terms, and D and E zero ﬁeld splitting parameters with

the best ﬁt.

X-ray Absorption Spectroscopy Studies. All XAS data were

collected at the Stanford Synchrotron Radiation Lightsource (SSRL).

Solid samples were prepared air-sensitively as reported previ-

ously.⁸⁸⁻⁹³S K-edge XAS data were collected on Beamline (BL) 4-

3, which is equipped with a 20-pole, 2.0 T wiggler and a double-

crystal monochromator cooled with liquid nitrogen. The sample

chambers at each beamline were maintained under a He atmosphere

at room temperature, and ﬂuorescence was measured using a PIPS

detector. Data were collected in triplicate scans for each sample

during each run, and a calibration scan was collected before and after

each sample set on a Na₂S₂O₃standard with the ﬁrst pre-edge set to

2472.02 eV.⁷⁷Spectra were collected using energy step sizes of 1.0,

0.8, and 1.5 eV, over the pre-edge (2380−2465 eV), edge (2465−

2490 eV), and post-edge (2490−2690 eV) regions, respectively.

All calibrations, background subtractions, normalizations, and

averaging were performed using the Athena program in the IFFEFIT

XAS software package as previously reported.^94,95The background

was removed from each spectrum by ﬁtting a ﬁrst order polynomial to

the pre-edge region. The data were normalized by ﬁtting the post-

edge region with a ﬁrst or second order polynomial and setting the

step function to an intensity of 1.0 at a set point of 2490 eV.

Density Functional Theory Calculations. Unrestricted Kohn−

Sham density functional theory calculations were performed using the

QCHEM 5.1 package using a B3LYP/6-31G(d,p) functional/basis set

combination. Each single-molecule calculation was performed on

isolated molecules in the gas phase, neglecting any contribution from

the solvent or crystal lattice. Each complex was built starting from

XRD geometry data in order to more reasonably represent the

molecules seen in experiment; preliminary investigations indicated

several close-in-energy minima from subtle rotations of the benzene

rings.

Since geometry optimization resulted in a slight puckering of the

rings, we tested the sensitivity of the excess spin density localization to

the precise geometry involved. The localization was observed in a

fashion that was independent of several geometries (i.e., free drawn

optimization, XRD structure, and XRD structure plus geometry

−

optimization) except for 2[NTf₂]₂, which shows diﬀerences in NTf₂

binding to the metal (as described in the text). The calculation is not

sensitive to functional. PBE0 and M06 both show localization on the

nickel for the 2[OTf]₂, and PBE0 shows localization for 1[OTf]₂.

Spin density calculations showed no obvious dependence on the basis

set selected, and spin density diﬀerences were ultimately calculated

using the 6-31G(d,p) basis set. Finally, unrestricted Kohn−Sham

calculations were performed on the structures of square planar

(neutral, singlet) metal complexes 1 and 2, as a negative control. As

expected, these structures did not show any spin density diﬀerence.

All EPR simulations were performed in MATLAB (2018b) using

the Easyspin (v5.2.25) toolbox and utilizing the least-squares ﬁt

function (esﬁt).^86,87Spectra of the singly oxidized complexes that

contained a single species were initially ﬁt by manually ﬁnding g values

ASSOCIATED CONTENT

■

1

for an S = 1/2 spin system containing ¹⁴N and H nuclei. Upon

S

* Supporting Information

satisfactory agreement to the experimental spectra, esﬁt was used to

optimize the g values and A hyperﬁne terms. Spectra that contained

multiple species were ﬁt by separately modeling each species as an S =

1/2 spin system. Once esﬁt found a best ﬁt solution for each species, a

weighting term (Sys.weight) was added and manually adjusted to

match the weighting observed in the experimental spectrum. The esﬁt

function was again allowed to reﬁne to yield the ﬁnal g values, A

hyperﬁne terms, and weighting terms for the two species. Simulations

for the EPR spectrum of 2[BF₄] collected in 2-MeTHF/DCM at 109

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.inorg-

chem.9b01675.

Tabulated crystallographic data, molecular structures,

and spectroscopic data (NMR, UV−vis, EPR, and S K-

edge XAS). Bond distances and angles from DFT

calculations (PDF)

P

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

(12) Jacquet, J.; Desage-El Murr, M.; Fensterbank, L. Metal-

promoted coupling reactions implying ligand-based redox changes.

ChemCatChem 2016, 8, 3310−3316.

Accession Codes

CCDC 1919326−1919334 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.

(13) Allgeier, A. M.; Mirkin, C. A. Ligand design for electrochemi-

cally controlling stoichiometric and catalytic reactivity of transition

metals. Angew. Chem., Int. Ed. 1998, 37, 894−908.

(14) Broere, D. L. J.; Plessius, R.; van der Vlugt, J. I. New avenues

for ligand-mediated processes - expanding metal reactivity by the use

of redox-active catechol, o-aminophenol and o-phenylenediamine

ligands. Chem. Soc. Rev. 2015, 44, 6886−6915.

AUTHOR INFORMATION

■

(15) Balch, A. L.; Holm, R. H. Complete electron-transfer series of

the [M-N₄] type. J. Am. Chem. Soc. 1966, 88, 5201−5209.

(16) Kruger, H. J.; Peng, G.; Holm, R. H. Low-potential

nickel(III,II) complexes: new systems based on tetradentate

amidate-thiolate ligands and the influence of ligand structure on

potentials in relation to the nickel site in [NiFe]-hydrogenases. Inorg.

Chem. 1991, 30, 734−742.

Corresponding Authors

*E-mail: scott-daly@uiowa.edu.

*E-mail: scott-k-shaw@uiowa.edu.

*E-mail: james-shepherd@uiowa.edu.

ORCID

Suzanne C. Bart: 0000-0002-8918-9051

Scott K. Shaw: 0000-0003-3767-3236

James J. Shepherd: 0000-0002-6164-485X

Scott R. Daly: 0000-0001-6229-0822

(17) Chaudhuri, P.; Hess, M.; Mu

̈

ller, J.; Hildenbrand, K.; Bill, E.;

Weyhermuller, T.; Wieghardt, K. Aerobic oxidation of primary

̈

alcohols (including methanol) by copper(II)- and zinc(II)-phenoxyl

radical catalysts. J. Am. Chem. Soc. 1999, 121, 9599−9610.

(18) Holm, R. H. Nickel(II) complexes. I. Spectra of tricyclic Schiff

base complexes of nickel(II) and copper(II). J. Am. Chem. Soc. 1960,

82, 5632−5636.

Notes

The authors declare no competing ﬁnancial interest.

(19) Thomas, F. Ligand-centered oxidative chemistry in sterically

hindered salen complexes: an interesting case with nickel. Dalton

Trans 2016, 45, 10866−10877.

(20) Ciccione, J.; Leconte, N.; Luneau, D.; Philouze, C.; Thomas, F.

Geometric and electronic structures of nickel(II) complexes of redox

noninnocent tetradentate phenylenediamine ligands. Inorg. Chem.

2016, 55, 649−665.

(21) Mitra, K. N.; Choudhury, S.; Castineiras, A.; Goswami, S.

Chemistry of ruthenium(II) complexes of N-substituted 1,2-

benzoquinone diimines. Synthesis, structure and redox properties. J.

Chem. Soc., Dalton Trans. 1998, 2901−2906.

(22) Skara, G.; Pinter, B.; Geerlings, P.; De Proft, F. Revealing the

thermodynamic driving force for ligand-based reductions in quinoids;

conceptual rules for designing redox active and non-innocent ligands.

Chem. Sci. 2015, 6, 4109−4117.

ACKNOWLEDGMENTS

■

This work was generously supported by the National Science

Foundation (1650894 and 1651381). We thank Dale Swenson

for collecting the single-crystal XRD data. We also would like

to thank Bruker for collecting and solving the single-crystal

XRD data for 3. C.M.D. would like to thank the U.S.

Department of Education for a Graduate Assistance in Areas of

National Need (GAANN) fellowship. Portions of this research

were carried out at the Stanford Synchrotron Radiation

Laboratory (SSRL), which is a national user facility supported

by the US Department of Energy, Oﬃce of Science, Oﬃce of

Basic Energy Sciences under contract number DE-AC02-

76SF00515. We thank Erik Nelson and Matthew Latimer for

their assistance with the S K-edge XAS measurements at SSRL.

(23) Fry, N. L.; Rose, M. J.; Nyitray, C.; Mascharak, P. K. Facile

ligand oxidation and ring nitration in ruthenium complexes derived

from a ligand with dicarboxamide-N and phosphine-P donors. Inorg.

Chem. 2008, 47, 11604−11610.

REFERENCES

■

(1) Chirik, P. J. Preface: Forum on redox-active ligands. Inorg. Chem.

2011, 50, 9737−9740.

́

(24) Majumdar, P.; Falvello, L. R.; Tomas, M.; Goswami, S. A novel

dinuclear ruthenium complex bridged through a substituted

phenazine ligand formed by ruthenium-promoted oxidative assembly

of 1,3-diaminobenzene. Chem. - Eur. J. 2001, 7, 5222−5228.

(25) Metcalfe, R. A.; Lever, A. B. P. Tetraammineruthenium(II) and

-ruthenium(III) complexes of o-benzoquinone diimine and their

redox series. Inorg. Chem. 1997, 36, 4762−4771.

(2) Storr, T.; Mukherjee, R. Preface for the forum on applications of

metal complexes with ligand-centered radicals. Inorg. Chem. 2018, 57,

9577−9579.

(3) Luca, O. R.; Crabtree, R. H. Redox-active ligands in catalysis.

Chem. Soc. Rev. 2013, 42, 1440−1459.

(4) Kaim, W.; Schwederski, B. Non-innocent ligands in bioinorganic

chemistryAn overview. Coord. Chem. Rev. 2010, 254, 1580−1588.

(5) Chirik, P. J.; Wieghardt, K. Radical ligands confer nobility on

base-metal catalysts. Science 2010, 327, 794.

(26) Santana da Silva, R.; Gorelsky, S. I.; Dodsworth, E. S.; Tfouni,

E.; Lever, A. B. P. Synthesis, spectral and redox properties of

tetraammine dioxolene ruthenium complexes. J. Chem. Soc., Dalton

Trans. 2000, 4078−4088.

(27) Shaw, J. L.; Wolowska, J.; Collison, D.; Howard, J. A. K.;

McInnes, E. J. L.; McMaster, J.; Blake, A. J.; Wilson, C.; Schroeder, M.

Redox Non-innocence of Thioether Macrocycles: Elucidation of the

Electronic Structures of Mononuclear Complexes of Gold(II) and

Silver(II). J. Am. Chem. Soc. 2006, 128, 13827−13839.

(28) Stephen, E.; Huang, D.; Shaw, J. L.; Blake, A. J.; Collison, D.;

Davies, E. S.; Edge, R.; Howard, J. A. K.; McInnes, E. J. L.; Wilson, C.;

Wolowska, J.; McMaster, J.; Schroder, M. Redox non-innocence of

thioether crowns: spectroelectrochemistry and electronic structure of

formal nickel(III) complexes of aza-thioether macrocycles. Chem. -

Eur. J. 2011, 17, 10246−10258.

(29) Martinez-Alanis, P. R.; Sanchez Eguia, B. N.; Ugalde-Saldivar,

V. M.; Regla, I.; Demare, P.; Aullon, G.; Castillo, I. Copper versus

thioether-centered oxidation: mechanistic insights into the non-

(6) Praneeth, V. K. K.; Ringenberg, M. R.; Ward, T. R. Redox-active

ligands in catalysis. Angew. Chem., Int. Ed. 2012, 51, 10228−10234.

(7) van der Vlugt, J. I. Cooperative catalysis with first-row late

transition metals. Eur. J. Inorg. Chem. 2012, 2012, 363−375.

(8) Lyaskovskyy, V.; de Bruin, B. Redox non-innocent ligands:

Versatile new tools to control catalytic reactions. ACS Catal. 2012, 2,

270−279.

(9) Alig, L.; Fritz, M.; Schneider, S. First-row transition metal

(de)hydrogenation catalysis based on functional pincer ligands. Chem.

Rev. 2019, 119, 2681−2751.

(10) Berben, L. A.; de Bruin, B.; Heyduk, A. F. Non-innocent

ligands. Chem. Commun. 2015, 51, 1553−1554.

(11) van der Vlugt, J. I. Radical-type reactivity and catalysis by

single-electron transfer to or from redox-active ligands. Chem. - Eur. J.

2019, 25, 2651−2662.

Q

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

innocent redox behavior of tripodal benzimidazolylaminothioether

ligands. Chem. - Eur. J. 2013, 19, 6067−6079.

(30) Inoue, S.; Mitsuhashi, M.; Ono, T.; Yan, Y.-N.; Kataoka, Y.;

Handa, M.; Kawamoto, T. Photo- and electrocatalytic hydrogen

production using valence isomers of N₂S₂-type nickel complexes.

Inorg. Chem. 2017, 56, 12129.

(47) Girolami, G. S.; Rauchfuss, T. B.; Angelici, R. J. Synthesis and

technique in inorganic chemistry: A laboratory manual, 3rd ed.;

University Science Books: Sausalito, CA, 1999.

(48) West, A. R. Solid state chemistry and its applications; John Wiley

& Sons: 2014.

́

(49) Chmielewski, P. J.; Latos-Grazynski, L. EPR and 2H NMR

̇

studies on the oxidation of nickel(II) tetraphenylcarbaporphyrin to

form novel organometallic nickel(III) complexes. Inorg. Chem. 1997,

36, 840−845.

(50) Ottenwaelder, X.; Ruiz-García, R.; Blondin, G.; Carasco, R.;

Cano, J.; Lexa, D.; Journaux, Y.; Aukauloo, A. From metal to ligand

electroactivity in nickel(II) oxamato complexes. Chem. Commun.

2004, 504−505.

(51) Krueger, H. J.; Holm, R. H. Stabilization of nickel(III) in a

classical N₂S₂coordination environment containing anionic sulfur.

Inorg. Chem. 1987, 26, 3645−3647.

(52) De Castro, B.; Freire, C. EPR and electrochemical study of

nickel(III) complexes of bis(3,5-dichlorosalicylaldehyde) o-phenyl-

enediimine. Evidence for adduct formation with pyridines. Inorg.

Chem. 1990, 29, 5113−5119.

(53) Lyons, C. T.; Stack, T. D. P. Recent advances in phenoxyl

radical complexes of salen-type ligands as mixed-valent galactose

oxidase models. Coord. Chem. Rev. 2013, 257, 528−540.

(54) Evangelio, E.; Ruiz-Molina, D. Valence tautomerism: New

challenges for electroactive ligands. Eur. J. Inorg. Chem. 2005, 2005,

2957−2971.

(31) Durgaprasad, G.; Luna, J. A.; Spielvogel, K. D.; Haas, C.; Shaw,

S. K.; Daly, S. R. Ru(II) complexes with a chemical and redox-active

S₂N₂ligand: Structures, electrochemistry, and metal-ligand coopera-

tivity. Organometallics 2017, 36, 4020−4031.

(32) Bayer, E.; Breitmaier, E. Umlagerung und komplexbildung des

glyoxal-bis-[2-mercapto-anils] und anderer mercaptoanile. Chem. Ber.

1968, 101, 1579−1593.

(33) Chłopek, K.; Bothe, E.; Neese, F.; Weyhermuller, T.;

̈

Wieghardt, K. Molecular and electronic structures of tetrahedral

complexes of nickel and cobalt containing N,N’-disubstituted, bulky o-

diiminobenzosemiquinonate(1-) π-radical ligands. Inorg. Chem. 2006,

45, 6298−6307.

(34) Rosenkoetter, K. E.; Wojnar, M. K.; Charette, B. J.; Ziller, J. W.;

Heyduk, A. F. Hydrogen-atom noninnocence of a tridentate [SNS]

pincer ligand. Inorg. Chem. 2018, 57, 9728−9737.

(35) Singh, J.; Hundal, G.; Gupta, R. Studies on nickel(II)

complexes with amide-based ligands: Syntheses, structures, electro-

chemistry and oxidation chemistry. Eur. J. Inorg. Chem. 2008, 2008,

2052−2063.

(36) Stavropoulos, P.; Carrie, M.; Muetterties, M. C.; Holm, R. H.

Reaction sequence related to that of carbon monoxide dehydrogenase

(acetyl coenzyme A synthase): thioester formation mediated at

structurally defined nickel centers. J. Am. Chem. Soc. 1990, 112,

5385−5387.

(37) Hong, D.; Tsukakoshi, Y.; Kotani, H.; Ishizuka, T.; Ohkubo, K.;

Shiota, Y.; Yoshizawa, K.; Fukuzumi, S.; Kojima, T. Mechanistic

insights into homogeneous electrocatalytic and photocatalytic hydro-

gen evolution catalyzed by high-spin Ni(II) complexes with S₂N₂-type

tetradentate ligands. Inorg. Chem. 2018, 57, 7180−7190.

(38) Jain, R.; Mamun, A. A.; Buchanan, R. M.; Kozlowski, P. M.;

Grapperhaus, C. A. Ligand-assisted metal-centered electrocatalytic

hydrogen evolution upon reduction of a bis(thiosemicarbazonato)-

Ni(II) complex. Inorg. Chem. 2018, 57, 13486−13493.

(39) Smith, C. I.; Crayston, J. A.; Hay, R. W. Reduction of carbon

dioxide by nickel macrocyclic catalysts adsorbed on a mercury

electrode or a copper rotating disc electrode. J. Chem. Soc., Dalton

Trans. 1993, 3267−3269.

(40) Connelly, N. G.; Geiger, W. E. Chemical redox agents for

organometallic chemistry. Chem. Rev. 1996, 96, 877−910.

(41) Souverain, D.; Leborgne, A.; Sauvet, G.; Sigwalt, P. Polymer-

izations of ethylenic monomers initiated by superacidsII: complex-

ation of some trifluoromethanesulphonates by their conjugate acid.

Eur. Polym. J. 1980, 16, 861−865.

(42) Foropoulos, J., Jr; DesMarteau, D. D. Synthesis, properties, and

reactions of bis((trifluoromethyl)sulfonyl)imide, (CF₃SO₂)₂NH.

Inorg. Chem. 1984, 23, 3720−3723.

(55) Zima, A. M.; Lyakin, O. Y.; Ottenbacher, R. V.; Bryliakov, K. P.;

Talsi, E. P. Iron-catalyzed enantioselective epoxidations with various

oxidants: Evidence for different active species and epoxidation

mechanisms. ACS Catal. 2017, 7, 60−69.

(56) Himmel, H.-J. Valence tautomerism in copper coordination

chemistry. Inorg. Chim. Acta 2018, 481, 56−68.

̈

(57) Lee, H.; Borgel, J.; Ritter, T. Carbon-fluorine reductive

elimination from nickel(III) complexes. Angew. Chem., Int. Ed. 2017,

56, 6966−6969.

(58) Rotthaus, O.; Jarjayes, O.; Thomas, F.; Philouze, C.; Perez Del

Valle, C.; Saint-Aman, E.; Pierre, J. L. Fine tuning of the oxidation

locus, and electron transfer, in nickel complexes of pro-radical ligands.

Chem. - Eur. J. 2006, 12, 2293−2302.

(59) Rotthaus, O.; Jarjayes, O.; Perez Del Valle, C.; Philouze, C.;

Thomas, F. A versatile electronic hole in one-electron oxidized Ni^II

bis-salicylidene phenylenediamine complexes. Chem. Commun. 2007,

4462−4464.

(60) Storr, T.; Wasinger, E. C.; Pratt, R. C.; Stack, T. D. P. The

geometric and electronic structure of a one-electron-oxidized

nickel(II) bis(salicylidene)diamine complex. Angew. Chem., Int. Ed.

2007, 46, 5198−5201.

(61) Chiang, L.; Herasymchuk, K.; Thomas, F.; Storr, T. Influence

of electron-withdrawing substituents on the electronic structure of

oxidized Ni and Cu salen complexes. Inorg. Chem. 2015, 54, 5970−

5980.

(62) Kunert, R.; Philouze, C.; Jarjayes, O.; Thomas, F. Stable M(II)-

radicals and nickel(III) complexes of a bis(phenol) N-heterocyclic

carbene chelated to group 10 metal ions. Inorg. Chem. 2019, 58,

8030−8044.

(43) Vij, A.; Zheng, Y. Y.; Kirchmeier, R. L.; Shreeve, J. n. M.

Electrophilic addition and substitution reactions of bis-

((trifluoromethyl)sulfonyl)amide and its N-chloro derivative. Inorg.

Chem. 1994, 33, 3281−3288.

(44) Lecarme, L.; Chiang, L.; Philouze, C.; Jarjayes, O.; Storr, T.;

Thomas, F. Detailed geometric and electronic structures of a one-

electron-oxidized Ni salophen complex and its amido derivatives. Eur.

J. Inorg. Chem. 2014, 2014, 3479−3487.

(63) Ghosh, M.; Weyhermuller, T.; Wieghardt, K. Electronic

̈

structure of the members of the electron transfer series [NiL]z (z =

3+, 2+, 1+, 0) and [NiL(X)]_n(X = Cl, CO, P(OCH₃)₃) species

containing a tetradentate, redox-noninnocent, Schiff base macrocyclic

ligand L: an experimental and density functional theoretical study.

Dalton Trans 2010, 39, 1996−2007.

(45) Enamullah, M.; Makhloufi, G.; Ahmed, R.; Joy, B. A.; Islam, M.

A.; Padula, D.; Hunter, H.; Pescitelli, G.; Janiak, C. Synthesis, X-ray,

and Spectroscopic Study of Dissymmetric Tetrahedral Zinc(II)

Complexes from Chiral Schiff Base Naphthaldiminate Ligands with

Apparent Exception to the ECD Exciton Chirality. Inorg. Chem. 2016,

55, 6449−6464.

(64) Rotthaus, O.; Labet, V.; Philouze, C.; Jarjayes, O.; Thomas, F.

Pseudo-octahedral Schiff base nickel(II) complexes: does single

oxidation always lead to the nickel(III) valence tautomer? Eur. J.

Inorg. Chem. 2008, 2008, 4215−4224.

(65) Ferentinos, E.; Raptopoulou, C. P.; Psycharis, V.; Terzis, A.;

Krzystek, J.; Kyritsis, P. Magnetostructural correlations in S = 1 trans-

[Ni{(OPPh₂)(EPPh₂)N}₂(dmso)₂], E = S, Se, and related complexes.

Polyhedron 2018, 151, 177−184.

(46) Drago, R. S. Physical methods for chemists, 2nd ed.; Surfside

Scientiﬁc Publishers: Gainesville, FL, 1992.

R

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry

Article

(66) Schweinfurth, D.; Krzystek, J.; Schapiro, I.; Demeshko, S.;

Klein, J.; Telser, J.; Ozarowski, A.; Su, C.-Y.; Meyer, F.; Atanasov, M.;

Neese, F.; Sarkar, B. Electronic structures of octahedral Ni(II)

complexes with “click” derived triazole ligands: A combined

structural, magnetometric, spectroscopic, and theoretical study.

Inorg. Chem. 2013, 52, 6880−6892.

(67) Krzystek, J.; Park, J.-H.; Meisel, M. W.; Hitchman, M. A.;

Stratemeier, H.; Brunel, L.-C.; Telser, J. EPR spectra from “EPR-

silent” species: High-frequency and high-field EPR spectroscopy of

pseudotetrahedral complexes of nickel(II). Inorg. Chem. 2002, 41,

4478−4487.

(68) Arumugam, K.; Shaw, M. C.; Mague, J. T.; Bill, E.; Sproules, S.;

Donahue, J. P. Long-Range Spin Coupling: A Tetraphosphine-

Bridged Palladium Dimer. Inorg. Chem. 2011, 50, 2995−3002.

(69) Dunn, T. J.; Chiang, L.; Ramogida, C. F.; Hazin, K.; Webb, M.

I.; Katz, M. J.; Storr, T. Class III Delocalization and Exciton Coupling

in a Bimetallic Bis-ligand Radical Complex. Chem. - Eur. J. 2013, 19,

9606−9618.

(70) Herebian, D.; Wieghardt, K. E.; Neese, F. Analysis and

interpretation of metal-radical coupling in a series of square planar

nickel complexes: correlated ab initio and density functional

investigation of [Ni(LISQ)₂] (LISQ = 3,5-di-tert-butyl-o-diimino-

benzosemiquinonate (1-)). J. Am. Chem. Soc. 2003, 125, 10997−

11005.

(71) Mavros, M. G.; Tsuchimochi, T.; Kowalczyk, T.; McIsaac, A.;

Wang, L.-P.; Van Voorhis, T. What can density functional theory tell

us about artificial catalytic water splitting? Inorg. Chem. 2014, 53,

6386−6397.

(72) Cramer, C. J.; Truhlar, D. G. Density functional theory for

transition metals and transition metal chemistry. Phys. Chem. Chem.

Phys. 2009, 11, 10757−10816.

(73) Sarangi, R.; DeBeer George, S.; Rudd, D. J.; Szilagyi, R. K.;

Ribas, X.; Rovira, C.; Almeida, M.; Hodgson, K. O.; Hedman, B.;

Solomon, E. I. Sulfur K-Edge X-ray Absorption Spectroscopy as a

Probe of Ligand-Metal Bond Covalency: Metal vs Ligand Oxidation

in Copper and Nickel Dithiolene Complexes. J. Am. Chem. Soc. 2007,

129, 2316−2326.

(74) Glaser, T.; Hedman, B.; Hodgson, K. O.; Solomon, E. I. Ligand

K-Edge X-ray Absorption Spectroscopy: A Direct Probe of Ligand-

Metal Covalency. Acc. Chem. Res. 2000, 33, 859−868.

(85) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.;

McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de

Streek, J.; Wood, P. A. Mercury CSD 2.0 - new features for the

visualization and investigation of crystal structures. J. Appl. Crystallogr.

2008, 41, 466−470.

(86) Stoll, S.; Schweiger, A. EasySpin, a comprehensive software

package for spectral simulation and analysis in EPR. J. Magn. Reson.

2006, 178, 42−55.

(87) Misra, S. K. Multifrequency Electron Paramagnetic Resonance:

Data and Techniques; John Wiley & Sons: 2014.

(88) Donahue, C. M.; McCollom, S. P.; Forrest, C. M.; Blake, A. V.;

Bellott, B. J.; Keith, J. M.; Daly, S. R. Impact of coordination

geometry, bite angle, and trans influence on metal-ligand covalency in

phenyl-substituted phosphine complexes of Ni and Pd. Inorg. Chem.

2015, 54, 5646−5659.

(89) Donahue, C. M.; Lezama Pacheco, J. S.; Keith, J. M.; Daly, S. R.

Sulfur K-edge X-ray absorption spectroscopy and time-dependent

density functional theory of arsenic dithiocarbamates. Dalton Trans

2014, 43, 9189−9201.

(90) Blake, A. V.; Wei, H.; Lee, K.; Donahue, C. M.; Keith, J. M.;

Daly, S. R. Solution and solid-state ligand K-edge XAS studies of

PdCl₂diphosphine complexes with phenyl and cyclohexyl sub-

stituents. Eur. J. Inorg. Chem. 2018, 2018, 2267−2276.

(91) Blake, A. V.; Wei, H.; Donahue, C. M.; Lee, K.; Keith, J. M.;

Daly, S. R. Solid energy calibration standards for P K-edge XANES:

electronic structure analysis of PPh₄Br. J. Synchrotron Radiat. 2018,

25, 529−536.

(92) Lee, K.; Wei, H.; Blake, A. V.; Donahue, C. M.; Keith, J. M.;

Daly, S. R. Ligand K-edge XAS, DFT, and TDDFT analysis of pincer

linker variations in Rh(I) PNP complexes: reactivity insights from

electronic structure. Dalton Trans 2016, 45, 9774−9785.

(93) Lee, K.; Wei, H.; Blake, A. V.; Donahue, C. M.; Keith, J. M.;

Daly, S. R. Measurement of diphosphine σ-donor and π-acceptor

properties in d⁰titanium complexes using ligand K-edge XAS and

TDDFT. Inorg. Chem. 2018, 57, 10277−10286.

(94) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS:

data analysis for X-ray absorption spectroscopy using IFEFFIT. J.

Synchrotron Radiat. 2005, 12, 537−541.

(95) Hedman, B.; Frank, P.; Penner-Hahn, J. E.; Roe, A. L.;

Hodgson, K. O.; Carlson, R. M. K.; Brown, G.; Cerino, J.; Hettel, R.;

Troxel, T.; Winick, H.; Yang, J. Sulfur K-edge X-ray absorption studies

using the 54-pole wiggler at SSRL in undulator mode. Nucl. Instrum.

Methods Phys. Res., Sect. A 1986, 246, 797−800.

(75) Szilagyi, R. K.; Lim, B. S.; Glaser, T.; Holm, R. H.; Hedman, B.;

Hodgson, K. O.; Solomon, E. I. Description of the ground state wave

functions of Ni dithiolenes using sulfur K-edge X-ray absorption

spectroscopy. J. Am. Chem. Soc. 2003, 125, 9158−9169.

(76) Sproules, S.; Wieghardt, K. Dithiolene radicals: Sulfur K-edge

X-ray absorption spectroscopy and Harry’s intuition. Coord. Chem.

Rev. 2011, 255, 837−860.

(77) Solomon, E. I.; Hedman, B.; Hodgson, K. O.; Dey, A.; Szilagyi,

R. K. Ligand K-edge X-ray absorption spectroscopy: covalency of

ligand-metal bonds. Coord. Chem. Rev. 2005, 249, 97−129.

(78) Dey, S.; Wayland, B. B.; Zdilla, M. J. Solution and solid state

properties for low-spin cobalt(II) dibenzotetramethyltetraaza[14]-

annulene [(tmtaa)CoII] and the monopyridine complex. Inorg. Chem.

2019, 58, 1224−1233.

(79) Macromolecular Crystallography, Part A; Otwinowski, Z., Minor,

W., Eds.; Academic Press: New York, 1997; Vol. 276.

(80) Bruker SADABS v.2.03; Bruker AXS Inc.: Madison, WI, 2001.

(81) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.

K.; Puschmann, H. OLEX2: a complete structure solution, refinement

and analysis program. J. Appl. Crystallogr. 2009, 42, 339−341.

(82) Spek, A. L. PLATON SQUEEZE: a tool for the calculation of

the disordered solvent contribution to the calculated structure factors.

Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 9−18.

(83) Sheldrick, G. M. SHELXT-Integrated space-group and crystal-

structure determination. Acta Crystallogr., Sect. A: Found. Adv. 2015,

71, 3−8.

(84) Sheldrick, G. M. Crystal structure refinement with SHELXL.

Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.

S

DOI: 10.1021/acs.inorgchem.9b01675

Inorg. Chem. XXXX, XXX, XXX−XXX

Article Doi

The Influence of Redox-Innocent Donor Groups in Tetradentate Ligands Derived from o-Phenylenediamine: Electronic Structure Investigations with Nickel

DOI: 10.1021/acs.inorgchem.9b01675

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acs.inorgchem.9b01675

Full text of DOI:10.1021/acs.inorgchem.9b01675

Products guided by the article

R&D Labs maybe for 144919-18-6

Relevant to this article

Hot Product