Stabilizing coordinated radicals via metal-ligand covalency: A structural, spectroscopic, and theoretical investigation of group 9 tris(dithiolene) complexes

Article abstract of DOI:10.1021/acs.inorgchem.5b00289

Proper assignment of redox loci in coordination complexes with redox-active ligands to either the metal or the ligand is essential for rationalization of their chemical reactivity. However, the high covalency endemic to complexes of late, third-row transition metals complicates such assignments. Herein, we systematically explore the redox behavior of a series of group 9 tris(dithiolene) complexes, [M(mnt)₃]^3- (M = Ir, Rh, Co; mnt = maleonitriledithiolate). The Ir species described comprise the first examples of homoleptic Ir dithiolene complexes. The enhanced metal-ligand covalency of the Ir-S interaction leads to remarkable reactivity of [Ir(mnt)₃]^3- and stabilization of mononuclear [Ir(mnt)₃]^2- complex ions as well as dimerized versions featuring weak, covalent, intermolecular S-S bonds. The dianionic Rh and Co analogues are, in contrast, highly unstable, resulting in the rapid formation of [Rh₂(mnt)₅]^4- and [Co(mnt)₂]22-, respectively. The synthesized complexes were studied by single-crystal X-ray diffraction, X-ray absorption spectroscopy, optical spectroscopy, magnetometry, density functional theory, and spectroscopy-oriented configuration interaction calculations. Spectroscopic and theoretical analyses suggest that the stability of [Ir(mnt)₃]^2- may be attributed to dilution of ligand radical character by a high degree of Ir 5d character in the singly occupied molecular orbital.

Full text of DOI:10.1021/acs.inorgchem.5b00289

Article
pubs.acs.org/IC
Stabilizing Coordinated Radicals via Metal−Ligand Covalency: A
Structural, Spectroscopic, and Theoretical Investigation of Group 9
Tris(dithiolene) Complexes
†
,‡
†
†
‡
Thorbjørn J. Morsing, Samantha N. MacMillan, Jacob W. H. Uebler, Theis Brock-Nannestad,
‡
,†
Jesper Bendix, and Kyle M. Lancaster*
†Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853, United States
‡Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen, Denmark
*
S Supporting Information
ABSTRACT: Proper assignment of redox loci in coordination
complexes with redox-active ligands to either the metal or the
ligand is essential for rationalization of their chemical
reactivity. However, the high covalency endemic to complexes
of late, third-row transition metals complicates such assign-
ments. Herein, we systematically explore the redox behavior o−f
a series of group 9 tris(dithiolene) complexes, [M(mnt)3]3
(
M = Ir, Rh, Co; mnt = maleonitriledithiolate). The Ir species
described comprise the first examples of homoleptic Ir
dithiolene complexes. The enhanced metal−ligand covalency of the Ir−S interaction leads to remarkable reactivity of
3−
2−
[
Ir(mnt) ] and stabilization of mononuclear [Ir(mnt) ] complex ions as well as dimerized versions featuring weak, covalent,
3
3
intermolecular S−S bonds. The dianionic Rh and Co analogues are, in contrast, highly unstable, resulting in the rapid formation
4−
2−
of [Rh (mnt) ] and [Co(mnt) ] , respectively. The synthesized complexes were studied by single-crystal X-ray diffraction, X-
2
5
2 2
ray absorption spectroscopy, optical spectroscopy, magnetometry, density functional theory, and spectroscopy-oriented
configuration interaction calculations. Spectroscopic and theoretical analyses suggest that the stability of [Ir(mnt)3]2 may be
attributed to dilution of ligand radical character by a high degree of Ir 5d character in the singly occupied molecular orbital.
−
INTRODUCTION
is better described as “molecular redox,” where the radical is
equally distributed over the entire molecule. Conversely,
radicals in the Rh and Co congeners completely lack metal
character. This behavior was attributed to enhanced metal−
ligand orbital overlap due to the radial expansion of the 5d
orbitals in Ir arising as a consequence of relativistic core orbital
■
Since the 1960s, the chemistry of metal dithiolene complexes
has fascinated chemists, sparking heated debates about their
electronic structures and spurring the development and
application of new bonding models and spectroscopic
1
−3
methods.
Indeed, these early examples of “redox non-
18−21
contraction.
innocence” have birthed an entire field of study related to
complexes of redox noninnocent ligands and their use in energy
conversion, materials applications, and catalysis.
tal electronic structure studies, particularly those employing
high-resolution X-ray crystallography, X-ray absorption spec-
troscopy (XAS), density functional theory (DFT), and ab initio
calculations have enabled systematic distinction between formal
and physical oxidation states in these types of complexes.
This bifurcation of redox loci has facilitated evaluation and
tuning of the relative roles of ligand-centered and metal-
To further explore molecular redox behavior and its potential
consequences on chemical reactivity, a series of group 9
tris(mnt) (mnt = maleonitriledithiolate) complexes was
4
−8
Fundamen-
3−
synthesized. Although [Co(mnt)3] is a stable species that
has been known for some time, the one-electron oxidized
2−
22
[
Co(mnt)3] has never been isolated. This instability was
presumed, based on electrochemical measurements, to be due
9
−12
to ligand dissociation. Given that the Ir congener will exhibit
2−
greater M−S covalency, [Ir(mnt) ] should be stable to
3
isolation and characterization.
13−16
centered redox on chemical reactivity.
Intrinsically larger metal−ligand covalencies complicate
systematic understanding of the redox behavior of late, 5d
complexes with redox-active ligands. For example, a recent
comparison of complexes of the redox-active tris-
EXPERIMENTAL SECTION
■
Syntheses. General Considerations. All reactions were performed
under an atmosphere of N using fully dried and degassed solvents
2
unless otherwise noted. Acetonitrile (BDH Chemicals) was dried by
(pentafluorophenyl)corrole (tpfc) ligand with group 9 metals
demonstrated that “ligand-centered” and “metal-centered”
17
labels for redox events are inappropriate for the Ir congener.
Rather, one-electron oxidation of (tpfc)Ir(py) (py = pyridine)
Received: February 5, 2015
2
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
the method of Grubbs using a JC Meyer Solvent System. N,N-
Dimethylformamide (DMF, Mallinckrodt) was dried over 4 Å
molecular sieves. Ethyl acetate (Fisher) and methanol (Macron) for
and dried in vacuo. Yield: 232 mg (28%) 13C NMR (125.7 MHz,
−1
−1
CD CN): δ: 118.90, 126.88. IR: See Table S1. UV−vis, cm (ε, M
3
−1
cm ): 15 400 (1200), 22 500 (5450), 34 400 (40 000), 43 500
(24 800). ESI-MS: See Table S2. Anal. Calcd for C H N S Co: C,
chromatography were used as received. Hydrated IrCl was purchased
3
36 48
9 6
from Pressure Chemical and used as received. (ASN)Br (ASN = 5-
50.39; H, 5.64; N, 14.70. Found: C, 50.09; H, 5.50; N, 14.86.
23
24
azonia-spiro[4,4]nonane) and Na mnt were synthesized according
{(ASN) [Ir(mnt) ]} ·CH Cl ({4[ASN] } ). 93 mg (0.094 mmol) of
2
2
3
2
2
2
2 2
to published procedures and dried under high (≤5 mTorr) vacuum
before use.
1[ASN] and 26 mg (0.079 mmol) of FcPF were suspended in 100
3
6
mL of dichloromethane and stirred vigorously for 5 min, during which
time the solution turned dark green. The solution was filtered to
(
ASN) [Ir(mnt) ] (1[ASN] ). 300 mg (0.85 mmol) of IrCl ·3H O
3
3
3
3
2
was refluxed with stirring under N for 24 h in 50 mL of dry CH CN
2
3
remove excess 1[ASN] , and diethyl ether was vapor diffused into it
3
in a 125 mL flask to afford a clear yellow solution of IrCl (NCCH ) .
3
3
3
over 48 h. The product was removed by filtration, washed with diethyl
+
This solution was taken to dryness on a rotary evaporator. The flask
ether, and dried. Yield: 72 mg (quantitative with respect to Fc ). IR:
was then charged with 570 mg (3.1 mmol) of Na mnt and 30 mL of
−1
−1
−1
2
See Table S1. UV−vis, cm (ε, M cm ): 15 600 (1500), 26 200
9200), 28 500 (8200), 34 500 (18 900), 38 000 (23 500). ESI-MS:
See Table S2. Anal. Calcd for C28.5H ClN S Ir: C, 37.74; H, 3.67; N,
dry DMF. This mixture was degassed via three freeze−pump−thaw
(
cycles and subsequently heated to reflux for 24 h under N . The
2
33
8 6
obtained dark red solution was evaporated to dryness on a high
vacuum line to yield a dark oily residue that was subsequently
dissolved in 6 mL of a 1:1 ethyl acetate/methanol mixture. This
solution was loaded onto a 5 cm alumina column equilibrated with
ethyl acetate. The column was washed with two volumes of ethyl
acetate to remove trace DMF, and the product was subsequently
eluted with neat methanol. The eluent solution containing the product
1
2.35. Found: C, 38.11; H, 3.35; N, 12.31.
PPh ) [Ir(mnt) ] (4[PPh ] ). 200 mg (0.122 mmol) of 1[PPh ]
4 3
(
4
2
3
4 2
and 41 mg (0.124 mmol) of FcPF were placed in a flask and dissolved
6
in 10 mL of dichloromethane. After standing for an hour, the solution
was filtered, and diethyl ether was vapor diffused into it. Precipitation
of colorless (Ph P)PF began almost immediately (identity confirmed
by single-crystal analysis), and when the green product started to
precipitate, the solution was filtered to remove (Ph P)PF , and the
ether diffusion continued until the solution was only slightly colored.
The product was removed by filtration and washed with diethyl ether.
Yield: 92 mg (58%). This initial product is not completely pure as
shown by elemental analysis: Anal. Calcd C: 55.82, H: 3.12, N: 6.51.
Found. 56.66, H: 2.90, N: 6.48. Small portions used for further studies
were purified by dissolving in dichloromethane and then dropwise
addition of ether, producing a fine needlelike precipitate: Found.
4
6
(
presumably Na [Ir(mnt) ]) was evaporated to dryness. This material
3 3
4
6
was dissolved in 4 mL of methanol and filtered. A solution of 1.1 g of
(
ASN)Br (5.3 mmol) in 1.5 mL of methanol was added quickly,
immediately yielding a turbid solution. All precipitate giving rise to this
turbidity subsequently dissolved within seconds and then reprecipi-
tated over the course of 5 min. The now-persistent precipitate was
pelleted by centrifugation, the supernatant was decanted, and the
precipitate was triturated with 3 × 5 mL of ethanol. The red product
was then transferred to a flask containing 50 mL of 2-propanol and 70
mL of acetone. Upon standing, orange-red crystals formed that were
subsequently collected on a fritted glass funnel, washed with ethanol,
and dried under vacuum. Yield: 288 mg (32% based on IrCl ·3H O)
5
5.93, H: 2.96, N: 6.52.
ASN) [Rh (mnt) ]·CH CH(OH)CH (5[ASN] ). 85 mg (0.094
(
4
2
5
3
3
4
mmol) of 2[ASN]3 and 26 mg (0.079 mmol) of FcPF6 were
suspended in 50 mL of 2-propanol. Acetone was added until all solid
dissolved, giving a maroon solution. This solution was filtered and
then concentrated in air giving a dark brown (almost black) precipitate
that was filtered and dried. Yield: 26 mg (45%). IR: See Table S1.
3
2
1
3
C NMR (125.7 MHz, CD CN): δ: 120.26, 125.69. IR: See
3
−1
−1
−1
Supporting Information, Table S1. UV−vis, cm (ε, M cm ):
2
3
1 900 (6780), 24 450 (6460), 29 400 (12 200), 33 100 (19 100),
8 200 (15 000). Electrospray ionization mass spectrometry (ESI-
−1
−1
−1
UV−vis, cm (ε, M cm ): 23 300 (8700), 32 200 (18 000), 37 500
33 500), 41 000 (42 600). ESI-MS: See Table S2. Anal. Calcd for
C H N OS Rh : C, 44.92; H, 4.93; N, 13.33. Found: C, 44.78; H,
MS): See Supporting Information, Table S2. Anal. Calcd for
C H N S Ir: C, 43.61; H, 4.88; N, 12.71. Found: C, 43.28; H,
(
3
6
48
9 6
55
72 14
10
2
4
.66; N, 12.56.
PPh ) [Ir(mnt) ] (1[PPh ] ). The procedure for the synthesis of
4
.57; N, 13.53.
ASN) [Co (mnt) ] ({6[ASN]} ). 232 mg (0.270 mmol) of
(
4
3
3
4 3
(
2
2
4
2
1
[ASN] was followed, starting from 0.333 g of IrCl ·3H O (1 mmol),
3 3 2
3
[ASN] and 75 mg (0.227 mmol) of FcPF were suspended in 50
3
6
until the addition of (ASN)Br. Instead 2.2 g (5.87 mmol) of (Ph P)Cl
4
mL of 2-propanol. 100 mL of acetone was added to fully dissolve both
compounds. The resulting dark brown solution was filtered and left to
evaporate for 18 h. Brown, needlelike crystals of product were filtered
off, washed with ethanol, and air-dried. Yield: 81 mg (77%). IR: See
was dissolved in 2 mL of methanol and added to the methanol
solution of Na [Ir(mnt) ]. A dark red oil immediately precipitated,
3
3
and the supernatant was decanted off. The oil was thoroughly washed
with methanol and dried under a stream of N . Diethyl ether was
2
−1
−1
−1
Table S1. UV−vis, cm (ε, M cm ): 12 500 (3200), 18 200
added, and the solid product precipitated. It was washed 4−5 times
with copious amounts of diethyl ether and dried in air. Yield: 0.82 g
(
(
2600), 21 800 (4700), 24 100 (5700), 27 100 (7400), 32 700
21 000), 37 500 (26 900). MS (ESI): See Table S2. Anal. Calcd for
(
6
53.3). Elemental Anal. Calcd: C: 61.88, H: 3.17, N: 5.15. Found: C:
1.42, H: 3.01, N: 5.36.
ASN) [Rh(mnt) ] (2[ASN] ). This molecule was synthesized
C H N S Co : C, 41.32; H, 3.47; N, 15.05. Found: C, 41.54; H,
32 32 10
8
2
3
.22; N, 14.99.
(
3
3
3
(
PPh ) (mnt ) , Tetraphenylphosphonium (Z)-2-[(Z)-1,2-di-
following a similar method as for 1[ASN] , using 325 mg (1.23
4 2 2 2
3
cyano-2-thioxoethenyldithio]-3-thioxo-2-butenedinitrile. The
addition of tetraphenylphosphonium to a solution of sodium
mmol) of RhCl ·3H O and 690 mg (3.71 mmol) of Na mnt. Yield:
3
2
2
13
4
75 mg (43%) of a deep red crystalline solid. C NMR (125.7 MHz,
2
−
maleonitriledithiolene effects the oxidation of mnt by air to give
CD CN): δ: 119.69/119.70 (J =1.9 Hz), 126.48. IR: See Table S1.
3
Rh
−1
(mnt2)2 as the Ph P salt: 1 g (5.3 mmol) of Na mnt was dissolved in
−
−1
−1
UV−vis, cm (ε, M cm ): 23 100 (13 000), 32 200 (22 200),
5 800 (44 700), 40 000 (113 400), 45 700 (81 600). ESI-MS: See
Table S2. Anal. Calcd for C H N S Rh: C, 47.93; H, 5.36; N, 13.97.
4
2
1
0 mL of water, and 0.330 g (0.88 mmol) of (PPh )Cl was dissolved in
3
4
2
5 mL of water. Both solutions were filtered and then poured together.
3
6
48
9 6
An orange precipitate formed immediately. The solid was removed by
filtration, washed with copious amounts of water, and air-dried. The
yield is quantitative with regard to (PPh )Cl, but the compound is not
Found: C, 47.75; H, 5.24; N, 13.83.
ASN) [Co(mnt) ] (3[ASN] ). Synthesized by adaptation of a
(
3
3
3
2
2
literature procedure: 0.75 g (4.03 mmol) of Na mnt and 0.35 g (1.19
4
2
mmol) of Na [Co(CO ) ]·3H O were refluxed in 110 mL of ethanol
completely pure as gauged by the mass of the product, which slightly
exceeds the expected 0.43 g. Small amounts of the crude product (50
mg portions) were recrystallized from acetonitrile (with 1% water
3
3
3
2
for 2 h. The dark green solution was filtered hot into an ice-cold
solution of 1.1 g (5.3 mmol) of (ASN)Br in 50 mL of ethanol and left
for half an hour to precipitate. The crude solid was filtered, dissolved
in acetone, and filtered again. 2-Propanol was added to the filtrate. The
resulting solution was allowed to evaporate for 36 h, affording dark
green needles that were collected by filtration, washed with ethanol,
t
added) by diffusion of BuOMe into this solution. This method
−1
afforded crystals suitable for X-ray crystallography. UV−vis, cm (ε,
−
1
−1
M
cm ): 22 700 (16 700), 26 500 (17 500), 43 300 (14 800). ESI-
−
MS, m/z: 139.9505 (mnt ).
B
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
PHYSICAL METHODS
Except for solvent signals, no impurities were observed in any
of the recorded spectra. IR spectra were recorded on an Agilent
Cary 630 FTIR spectrophotometer. Measurements were taken
of pure, powdered samples. Magnetic susceptibility measure-
ments were conducted on an MPMS-XL Quantum-Design
SQUID magnetometer on a finely ground sample. The
susceptibilities were corrected for diamagnetic contributions
from the sample holder and from the sample by means of
Pascal’s constants, in the temperature range of 2−300 K. The
sample was mounted in a diamagnetic capsule, and the
measurements were done at 1000 and 3000 Oe.
■
X-ray Crystallography and Refinement of Structures.
Single-crystal X-ray diffraction studies were performed using a
Siemens P4 SMART CCD area detector with a graphite-
monochromated Mo Kα (λ = 0.710 73 Å) radiation source for
the structure of {6[ASN]} . For the rest of the structures, the
2
data were collected on a Bruker D8 Venture diffractometer
monochromated with a doubly curved silicon crystal (also Mo
Kα). The data sets were collected and processed using Bruker
25
26
Apex 2 with the SAINT and SADABS programs and refined
2
7
using Olex2. All hydrogen atoms were added using a riding
Standard UV−vis spectra (200−1100 nm) were collected
using a Varian Cary 60 scanning spectrophotometer. UV−vis−
NIR spectra (NIR = near-infrared; 200−2780 nm) were
collected using a Cary 5 scanning spectrophotometer. Spectra
for solid samples were measured in transmission mode on finely
ground samples pressed between sapphire plates. All
voltammetric measurements were performed using a Pine
Instruments WaveNow USB Potentiostat/Galvanostat. Com-
pounds were dissolved at ∼1 mM concentration in MeCN
model and refined with an isotropic displacement factor of
.2Ueq of the parent atom.
X-ray Absorption Spectroscopy. Ir L - and L -edge XAS
1
3
1
spectra were measured at the Stanford Synchrotron Radiation
Lightsource (SSRL) beamline 9−3 under ring conditions of 3
GeV and 500 mA. A Si(220) double-crystal monochromator
was used for energy selection. A Rh-coated mirror (set to an
energy cutoff of 15 keV) was used for harmonic rejection.
Internal energy calibration was performed for each individual
spectrum by assigning the first inflection points of the spectrum
of a finely dispersed K [IrCl ] on Kapton to 13 419 eV (Ir L ).
n
containing 0.1 M (N Bu )PF as the supporting electrolyte. For
4
6
cyclic and differential pulse voltammetry, a 3 mm Pt disk
working electrode, coiled Pt wire counter electrode, and a
single-junction Ag/AgCl reference electrode were employed.
Reduction potentials are reported relative to the ferrocenium/
ferrocene couple, which was measured as an internal standard
following completion of experiments with dithiolene complex
analytes. For spectroelectrochemical measurements, a gold
2
6
1
Data were collected in transmission mode on finely ground
samples of the dithiolene complexes diluted in BN pressed into
mm Al spacers sealed with 38 μm Kapton tape. Sample
temperatures were maintained at 10 K in an Oxford liquid He
1
flow cryostat. L -edge data were collected from 13 180 to
1
1
3 740 eV; L -edge data were collected from 10 985 to 12 115
3
“
honeycomb” spectroelectrochemical card, which contains a Au
eV to measure extended X-ray absorption fine structure
working electrode mesh and two Au counter electrode areas,
were used, along with a separate single-junction Ag/AgCl
reference electrode. Spectra were measured using a Cary 60
UV−vis spectrophotometer. Mass spectra were collected on a
Bruker Solarix ESI/MALDI FT-ICR mass spectrometer
equipped with a 7 T magnet. The ESI samples were prepared
in acetonitrile and measured using external calibration with
either L-histidine or sodium trifluoroacetate cluster ions. Data
collected were processed in Bruker Dataanalysis 4.0 SP5.
−1
(
EXAFS) to k = 15 Å . Two L and four L scans were
3 1
averaged and processed using the SIXPACK software pack-
2
8
age. Background subtraction was achieved by fitting a
polynomial to the pre-edge region and subtracting this
polynomial from the entire spectrum. Spectra were normalized
by fitting a flattened polynomial to the postedge (L -edge:
1
1
3 425; L -edge: 11 235) and normalizing the edge jump to 1.0.
3
EXAFS data were fit using the OPT package of EXAFSPAK
using paths calculated using FEFF7.
2
9,30
The S K-edge XAS spectra were measured at SSRL beamline
−3 under ring conditions of 3 GeV and 500 mA. Samples
CALCULATIONS
4
■
were prepared by grinding to a fine powder and were spread to
a vanishing thickness onto 38 μm low-S Mylar tape. All samples
were measured in a He atmosphere at room temperature in
fluorescence mode using a Lytle detector. The incident beam
energy was calibrated by setting the energy of the first peak in
the S K-edge spectrum of Na S O ·5H O to 2472.02 eV.
Methods. All calculations were performed using version
2.9.0 of the ORCA quantum chemistry suite. Geometry
optimizations were performed for [M(mnt)3] (n = 1, 2, 3)
3
1
n−
using density functional theory (DFT) with the BP86 exchange
3
2,33
correlation functional
along with the segmented all-electron
2
2
3
2
relativistically contracted (SARC) def2-TZVP(-f) basis set for
3
4,35
Intensity was normalized with respect to the incident beam
using a He-filled ion chamber upstream of the sample. Data
represent an average of four scans measured from 2420 to 2720
eV. Data were processed with SIXPACK in a manner similar to
those of the Ir XAS. Spectra were normalized by fitting a
polynomial flattened to energies below 2490 eV to the data and
normalizing the region below 2490 eV to unity.
all atoms.
regular approximation (ZORA)
solvation as modeled by the conductor-like screening model
Optimizations also included the zeroth order
3
6,37
for relativistic effects and
3
8
(COSMO) using an infinite dielectricum. Integration
accuracies were set to Grid7 for transition metal atoms and
to Grid4 for ligand atoms. Numerical frequency calculations
were performed at the converged geometries to confirm global
geometric minima.
Other Physical Methods. NMR experiments were
39
performed on a Bruker 500 MHz spectrometer equipped
The B3LYP hybrid exchange correlation functional, in
conjunction with the ZORA-optimized SARC def2-TZVP(-f)
basis set on all atoms, was used to generate single-point
unrestricted Kohn−Sham (UKS) electronic structure solutions
for use in calculation of spectroscopic observables. Electron
paramagnetic resonance (EPR) properties were predicted using
coupled perturbation Kohn−Sham theory for the g-matrix, and
the spin−orbit coupling (SOC) operator was treated by the
1
3
with a cryoprobe and operating at 125.7 MHz for
C
measurements. 13C NMR spectra are reported in ppm relative
to tetramethylsilane (δ = 0) and were referenced internally to
CH CN (δ = 118.26). Apart from the shifts given, all measured
3
samples showed the expected signals from the ASN counter-
13
ions: C NMR: singlet at 22.74, triplet at 63.85 (J = 3.1 Hz),
N
1
H NMR: multiplets at 2.20 (Int. 1.0) and 3.50 (Int. 1.0).
C
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
4
0,41
spin−orbit mean field approximation.
Electronic absorption
Cyclic Voltammetry and Redox Activity of 1[ASN]3.
spectra were calculated using time-dependent DFT (TDDFT).
Spectroscopy-oriented configuration interaction (SORCI)
The cyclic voltammetry of 1[ASN] shows two reversible
3
+/0
oxidation waves at −0.30 and 0.36 V versus Fc (Figure 2a).
An additional, irreversible oxidation occurs at 0.88 V. UV−vis
spectroelectrochemistry measured at potentials between −0.50
and 0.90 V shows two sets of isosbestic points, suggesting the
accessibility of two stable, higher-valent species, presumably
4
2
calculations on crystallographically determined {4[ASN]2}2
and mnt2 were performed to estimate the relative energy
−
2
separations between the ground and first triplet excited states.
SORCI was performed on a complete active space (CAS) for
4[ASN] } comprising 10 electrons and 6 orbitals (CAS-
10,6)). CAS(10,9) was chosen for mnt . Active spaces were
chosen to include orbitals with S−S bonding and antibonding
character. The def2-SV(P) basis set was used for all atoms. As
described elsewhere, individual selection was used to ease the
[Ir(mnt)3]2 and [Ir(mnt) ] (Figure 2b). Isosbestic points are
−
−
{
2 2
3
2
2
−
(
not encountered in spectral changes beyond 0.6 V, indicating
decomposition at these high potentials. Chemical oxidation of
1[ASN] with 1 equiv of FcPF in dichloromethane affords a
3
6
green solution. Diffusion of diethyl ether into this solution gave
dark red needles of (ASN) [Ir(mnt) ] ({4[ASN] } ) in
42
computational burden. The size of the first-order interacting
space was reduced with a threshold Tsel = 1 × 10− Eh. A
further approximation involves reduction of the reference space
through another selectionall initial references that contribute
4
3 2
2 2
6
quantitative yield (Scheme 2). X-ray structural analysis of
{4[ASN] } reveals a remarkable structure that features two
2
2
2−
closely packed [Ir(mnt)3] complex ions (Figure 1b). The
mutual orientation of the olefin backbones of proximal mnt
ligands precludes π-stacking as a driving force for this
intermolecular association. However, a significant interaction
between the two complexes can be inferred by the short
intermolecular S···S separations at 2.741(3) and 3.050(3) Å.
This prediction is borne out by SQUID magnetometry,
−5
less than a second threshold (T = 1 × 10 ) to the zeroth-
pre
order states are rejected from the reference space. The initial
orbitals for the first step of the SORCI procedure were taken
from restricted Kohn−Sham orbitals from a B3LYP/def2-
SV(P)/ZORA calculation. We note that the def2-SV(P) basis
set was chosen to attenuate the substantial computational
expense of ab initio calculation of the {4[ASN] } electronic
2
2
which shows strong antiferromagnetic coupling between these
1
structure, and we stress that the values for the bond energies
must be taken as qualitative estimates of the relative S−S bond
strengths.
formally S = / complex ions (Figure 3). The strength of this
2
coupling is too high to permit a precise estimation of J values
from the magnetometry data. The SQUID data for {4[ASN]2}2
show a large amount of temperature-independent para-
RESULTS AND DISCUSSION
−3
3
■
magnetism contribution (TIP, χTIP = 1.11 × 10 cm
Synthesis of [M(mnt)3]3 (M = Ir, Rh, Co). Prior to this
work, no published examples existed of any homoleptic Ir
dithiolene complexes despite more than half a century of rich
Treatment of IrCl (MeCN)3
with 3 equiv of Na mnt in refluxing DMF affords a red
−
mol ). This TIP behavior has been described previously for
−1
a series of Re(II) complexes for which χ was in the range of
TIP
−3
3
−1 46
(1.36−1.79) × 10 cm mol . Through experimental and
theoretical studies, Dunbar and co-workers established that the
large TIP contribution observed for these S = 1/2 complexes
resulted from strong spin−orbit coupling (λ = 2000−3000
1
−3,9,43,44
dithiolene chemistry.
3
2
solution, putatively of Na [Ir(mnt) ]. After removal of DMF,
the complex trianion (ASN) [Ir(mnt) ] (1[ASN] ) is
3
3
−1
47,48
cm ), by virtue of the presence of a 5d metal center.
Other
3
3
3
precipitated as an air-stable, red solid from methanol by the
addition of 5-azonia-spiro[4,4]nonane bromide ((ASN)Br,
Scheme 1, 32% yield). Starting from RhCl (MeCN) , the Rh
Re(II) complexes supported by redox-active tetrathiafulvalene
phosphine ligands exhibited even larger TIP contributions, with
−3
3
−1 49
χTIP values in the range of (9.34−10.7) × 10 cm mol .
3
3
+
4
In contrast to 1[ASN] , oxidation of the PPh analogue
3
Scheme 1. Synthesis of [M(mnt)3]3 (M = Ir, Rh, Co) Salts
−
1[PPh4]3 produces a green solution from which may be
crystallized green needles of monomeric 4[PPh ] (Scheme 2
4
2
and Supporting Information, Figure S6). SQUID magneto-
1
metry is consistent with an S = / ground state for this species,
2
although it too exhibits a large TIP (Figure 3, χ = 3.82 ×
TIP
−3
3
−1
1
0
cm mol ). Field-dependent SQUID data (Supporting
Information, Figure S1) were measured to exclude the presence
of ordering paramagnetic impurities as an origin for the
anomalously high χ T values. The solution vis−NIR spectrum
M
of {4[ASN] } shows remarkable similarity to the solid-state
2
2
congener, (ASN) [Rh(mnt) ] (2[ASN] ), is isolated in 43%
spectrum of 4[PPh
solid-state spectrum (Figure 4a). This spectroscopic evidence
of {4[ASN] dissociation in solution is further supported by
4 2
] while bearing no resemblance to its own
3
3
3
yield, following the same procedure. The synthesis of
−
[
Co(mnt)3]3 has been reported, and (ASN) [Co(mnt) ]
}
2 2
3
3
(
3[ASN] ) was prepared by an adaptation of the literature
time-dependent DFT (TDDFT) calculations, which reproduce
with good fidelity the dramatic differences between monomeric
and dimeric [Ir(mnt)3]2 in the near-infrared (NIR) region
(Figure 4b).
3
45
procedure (28% yield).
−
Crystals suitable for X-ray diffraction were grown by slow
evaporation of concentrated acetone/2-propanol solutions. The
solid-state structure of 1[ASN] reveals a slightly distorted
Chemical oxidation of 1[ASN] with 2 equiv of tris(4-
3
3
octahedral IrS inner coordination sphere with an average Ir−S
bromophenyl)aminium hexachloroantimonate (“magic blue”)
6
−
distance of 2.348(4) Å (Figure 1a). The structure of 2[ASN] is
presumably generates [Ir(mnt) ] , although analytically pure
3
3
similar to that of 1[ASN] , with an average Rh−S distance of
material was not isolated. Extended X-ray absorption fine
structure (EXAFS) analysis of the reaction mixture shows that
the orange oxidation product retains the tris(mnt) coordination
sphere, with average bond metrics in reasonable agreement
3
2
.342(7) Å (Supporting Information, Figure S3). This contrasts
with the structure of 3[ASN] , whose average Co−S distance is
3
2.259(4) Å (Supporting Information, Figure S4).
D
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. (a) Thermal ellipsoid plot of the complex anion in crystals of 1[ASN] . The complex is very close to ideal octahedral geometry, and the
3
average Ir−S distance is 2.348(4) Å. ASN counterions were omitted for clarity. Thermal ellipsoids are plotted at 50% probability. (b) Thermal
ellipsoid plot of the dimeric complex anion in crystals of {4[ASN] } at 50% probability. The structure reveals close S···S distances between the two
2
2
complex ions of 2.741(3) and 3.050(3) Å.
+
/0
Figure 2. (a) CV of 1[ASN] measured from −0.70 to 1.20 V (vs Fc ) in MeCN with 0.1 M [(nBu) N]PF (100 mV/s). The first and second
3
4
6
2−
−
oxidations to [Ir(mnt) ] and [Ir(mnt) ] are reversible; isolated voltammograms are superimposed in red. (b) UV−vis spectra measured at
3
3
+
/0
discrete reduction potentials from −0.50 to 0.90 V vs Fc show isosbestic points consistent with clean formation of two new species. The UV−vis
−
spectrum of 1[ASN] is black, and the spectrum of {4[ASN] } is red. The blue trace is assigned to the putative [Ir(mnt) ] species.
3
2
2
3
Scheme 2. Chemical Oxidation of 1[cat] (cat = ASN, PPh )
3
4
Results in the Formation of Dimeric {4[ASN] } and
2
2
Monomeric 4[PPh4]2
Figure 3. Plot of χT vs T from SQUID magnetometry of powdered
{
4[ASN] } (red) and 4[PPh ] (blue). The data are open circles; fits
2 2 4 2
to the data are solid lines. Dashed lines are shown to indicate idealized
g = 2.0 spin-only values for the triplet expected for a hypothetically
uncoupled complex ion dimer in {4[ASN] } and for the doublet state
with a DFT-calculated structure of [Ir(mnt)3]− (Supporting
Information, Figure S19 and Table S13). Further character-
ization was not pursued.
2
2
of the monomeric complex ion in 4[PPh
]
. Fitting parameters:
4
2
−3
3
−1
{
4[ASN] } : χ = 1.11 × 10 cm mol ; 4[PPh ] : g = 1.572(4),
2
2
TIP
4 2
−3
3
−1
χTIP = 3.82 × 10 cm mol .
Probing the Nature of the S···S Interaction in
{
4[ASN] } . The close proximity of the two complexes in
2 2
E
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 4. (a) vis−NIR spectra and their first derivatives with respect to energy measured by transmission either through solution or through powder
pressed between sapphire plates for {4[ASN] } in MeCN solution (black), powdered {4[ASN] } (blue), and powdered 4[PPh ] (red). (b)
2
2
2
2
4 2
2
−
4−
TDDFT (B3LYP-def2-TZVP-ZORA) calculated vis−NIR spectra for monomeric [Ir(mnt)3] (red) and [Ir(mnt)3]2 (blue).
1
dimeric {4[ASN] } is puzzling owing to the charge of each
cm− higher energy relative to the ground state and implies a
2
2
Ir(mnt)3]2 unit. The 2.741(3) and 3.050(3) Å S···S distances
−
−1
[
bond energy of ca. 12 kcal mol . A similar SORCI calculation
of the [(mnt)2] dimer, whose structure is provided in the
•
−
2−
suggest that these Ir-bound mnt S atoms are coupled, either
via covalent bonding or diradical bonding. These possibilities
were explored through broken symmetry (BS ) DFT and ab
initio SORCI calculations. No BS solution emerges,
suggesting that the {4[ASN] } dimer associates via a bona
Supporting Information (Figure S9), gives an S−S bond energy
57
−1
of 36.0 kcal mol . The smaller S−S bond energy of the
4
2
{
4[ASN] } dimer accords with Ir 5d dilution of the S 3p
2 2
2
2
character in the interacting orbitals. A similar rationale may be
used for the smaller bond energy of mnt relative to that of
Me S (62.0 kcal mol ), which reflects the dilution of S 3p
character in the frontier orbitals by admixture with the C
orbitals of the conjugated mnt backbone.
Cyclic Voltammetry and Redox Activity of 2[ASN]3
and 3[ASN] . The cyclic voltammogram of 3[ASN] is devoid
fide covalent S−S bond rather than through a diradical
2−
2
1
0,58
bond.
Our SORCI results support this interpretation of
−1 62
2
2
the S−S interaction: the dominant (82%) ground-state
electronic configuration of the {4[ASN] } dimer features a
2
2
doubly occupied S−S σ-bonding orbital (Figure 5a). The S···S
3
3
of reversible oxidations up to scan rates of 1 V/s at 298 K,
suggesting that [Co(mnt)3]2 decomposes on the electro-
chemical time scale (Supporting Information, Figure S12a).
−
However, UV−vis spectroelectrochemistry measured between
0.50 and 1.00 V versus Fc+ indicates conversion to at least
/0
−
two new species (Supporting Information, Figures S14 and
S15). Chemical oxidation with stoichiometric FcPF yields the
6
dimeric complex (ASN) [Co(mnt) ] ({6[ASN]} ) in 77%
2
2 2
2
yield (Scheme 3, Figure 6a). This structural motif is well-
5
0,51
Figure 5. Averaged atomic natural orbital plots of the (a) HOMO and
b) LUMO of dimeric {4[ASN] } . The HOMO is predominantly of
known.
High-resolution ESI-MS of the reaction mixture
(
•−
2
2
suggests that “mnt ” is formed as a byproduct (Supporting
S−S σ-bonding character, while the LUMO is dominated by the
Information, Figure S11). We cannot rule out formation of
corresponding S−S σ-antibonding interaction. Orbitals are plotted at
−3
an isolevel of 0.3 e Å .
Scheme 3. Chemical Oxidation of 2[ASN] and 3[ASN]
3
3
Generates Homobimetallic 5[ASN] and Dimeric {6[ASN]}
2
distances in {4[ASN] } suggest that the S−S covalent bond is
4
2
2
substantially weaker than in typical disulfides, which have S···S
59,60
distances on the order of 2.02−2.04 Å.
Excited-state calculations were used to estimate the strength
of the shorter S−S bond in {4[ASN] } . In the four-state
2
2
paradigm for a covalent bond proposed by Coulson and
6
1
Fischer, bond energy is taken as the energetic separation
between the singlet state arising from full occupation of the
bonding molecular orbital (MO) and the triplet state arising
from single occupation of the bonding and antibonding MOs
by electrons with parallel spins. Our SORCI calculations
indicate that the dominant (79%) electronic configuration of
the first triplet excited state in {4[ASN] } has an electronic
2
2
configuration in which the S−S σ and σ* orbitals are singly
occupied (Figure 5b). This state is calculated to be to 4100
F
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. (a) Thermal ellipsoid plot of the complex anion in {6[ASN]} at 50% probability level. (b) Thermal ellipsoid plot of the complex anion in
2
−
3−
5
[ASN] at 50% probability level. The complex has an unusual geometry best described as a [Rh(mnt) ] moiety coordinating to a [Rh(mnt) ]
4
2
3
moiety with the average nonbridging Rh−S distance being 2.336(9) Å. The Rh−S distances involving bridging S atoms are Rh1−S5: 2.431(1) Å,
Rh1−S6: 2.432(1) Å, Rh2−S5: 2.355(1) Å, and Rh2−S6: 2.363(1) Å. Counterions were omitted for clarity in both (a) and (b).
2
−
mnt , which we found to fragment under our conditions for
wholly S-centered and is isotropically distributed over the six S
2
ESI-MS. Spectroelectrochemical data collected via oxidation of
atoms in the molecule in an orbital of A symmetry (D ) that is
2
3
3
[ASN] show features characteristic of {6[ASN]} ; however,
unable to mix with the E and A symmetry d-orbitals. In
3
2
1
these data also indicate that additional, unidentified species are
formed electrochemically.
contrast, the Rh and Ir congeners have E symmetry SOMOs
with 17% and 26% metal character, respectively (Figure 7). The
The cyclic voltammetry of 2[ASN] exhibits quasi-reversible
3
+/0
oxidations at −0.04 and 0.39 V versus Fc
Information, Figure S12b). However, treatment of 2[ASN]3
(Supporting
with 1 equiv of FcPF does not result in the formation of
6
III
(
ASN) [Rh(mnt) ]. Rather, the homobimetallic Rh com-
2
3
pound (ASN) [Rh (mnt) ] (5[ASN] ) is isolated in 45% yield
4
2
5
4
(Scheme 3, Figure 6b). A similar yield is achieved using 0.5
equiv of FcPF6.
The dinuclear complex ion in 5[ASN] is not a dimer but
4
rather forms by filling the octahedral coordination sphere of a
[
Rh(mnt)2]− unit with bridging thiolates donated from an
3−
intact [Rh(mnt) ] . This molecule presumably forms via
dissociation of mnt (Supporting Information, Figure S11)
from one Rh center, after which the coordination sphere is
3
Figure 7. QRO plots of the SOMOs of [M(mnt)
]2− (M = Co, Rh,
3
•−
Ir). QROs were calculated at the B3LYP/def2-TZVP-ZORA level.
−3
Orbitals are plotted at an isosurface level of 0.3 e Å .
3−
filled via bridging S-ligation from an intact [Rh(mnt) ] . We
3
note that this edge-sharing bioctahedral motif is rare−related
remaining SOMO character in the Rh and Ir complexes is of S
species include Re and Mo tris(dithiolenes) as well as Co and
3
p parentage, although in both cases the S 3p orbital
52−56
Ru tris(dithiocarbamates).
In these cases, the coordination
coefficients of one mnt ligand are twofold larger than in the
other two mnt ligands (Supporting Information, Table S14).
The observation of nonequivalent S−S distances (2.741(3) and
3.050(3) Å) in {4[ASN] } accords with this anisotropic
spheres are distorted from octahedral geometry, and the
metrical parameters show evidence of metal−metal inter-
actions. In 5[ASN] however, the octahedra are unperturbed,
4
2
2
and the Rh···Rh distance of 3.5759(4) Å is inconsistent with a
distribution of S character calculated in the radical SOMO of
metal−metal interaction. Formation of 5[ASN] is rapid, as
2−
4
each [Ir(mnt)3] unit. Moreover, the structure of 4[PPh4]2
+/0
evidenced by the appearance of a feature at 0.28 V versus Fc
in the cyclic voltammogram of 3[ASN] , corresponding to the
shows evidence of a static Jahn−Teller distortion consistent
2
2−
64
3
with the E ground state calculated for [Ir(mnt) ] . The Ir−
3
2−/3−
[
Rh (mnt) ]
couple (Supporting Information, Figures
2
5
S distances for two mnt ligands are 2.340 Å, while the Ir−S
distances for the disordered third mnt are 2.277(4) and
2.184(7) Å. Moreover, modest Jahn−Teller splitting of the
SOMO is consistent with the large TIP exhibited by both
S12b and S13). Similar to {6[ASN]} , electrochemical
2
generation of 5[ASN] does not proceed cleanly as shown by
4
comparison of spectroelectrochemical data of the UV−vis
spectrum of the pure compound (Supporting Information,
Figures S16 and S17).
{
4[ASN] } and 4[PPh ] , as well as the inability to observe
2 2 4 2
their EPR spectra at temperatures greater than or equal to 10
K.
65,66
Electronic Structures of Dianionic Group 9 Tris-
2−
(
dithiolenes). To rationalize the one-electron reactivity of
group 9 tris(mnt) complexes (summarized in Schemes 2 and
), we used DFT to calculate ground-state electronic structures
for the hypothetical dianionic tris(mnt) monomers. Quasi-
For [Ir(mnt) ] , the radical hole has greater metal character
3
•
−
than in the Co and Rh congeners. This suggests that the mnt
3
units retain sufficient donor strength to disfavor dissociation.
Because {4[ASN] } gave no EPR signal, the relative Ir and S
2
2
63
restricted orbital (QRO) transformations of calculations at
contributions to the SOMO of this species was examined using
3
4,39
the B3LYP/def2-TZVP-ZORA level
show that the singly
XAS. Ir L -edge XAS probes core-to-valence excitations from
1
occupied molecular orbital (SOMO) of the Co congener is
the 2s subshell. The first inflection points of ns-to-valence XAS
G
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 8. (a) Experimental, (b) zoomed first-derivative, and (c) TDDFT-calculated (B3LYP-def2-TZVP-ZORA) Ir L -edge XAS spectra for
1
1
[ASN] (black) and {4[ASN] } (red). In the calculated spectra, the energy axis has not been adjusted to correct for inaccuracy in modeling the Ir
3
2 2
2
s binding energy.
Figure 9. (a) S K-edge XAS of solid 1[ASN] (black) and {4[ASN] } (red). The pre-edge feature in the spectrum of {4[ASN] } is diagnostic of an
3
2
2
2 2
orbital vacancy with significant S-character. This assignment is verified by the TDDFT-calculated (B3LYP-def2-TZVP-ZORA) S K-edge XAS spectra
shown in (b). The energy axis of the calculated spectra has not been corrected for inaccuracy in modeling the S 1s binding energy.
spectra are typically used as measurements of physical oxidation
(Figure 9a). However, it must be remembered that while the
“ligand radical” S−S σ-bonding MO is filled, two holes with
equal S-parentage remain in the S−S σ* MO, giving rise to the
6
7,68
state.
For {4[ASN] } , this occurs 0.7 eV lower in energy
2 2
than for 1[ASN] (Figure 8a). While this shift in the L -edge
3
1
•
−
accords with metal-centered redox, it also implies that
4[ASN] } has a lower oxidation state than that of 1[ASN] .
observed 1s → mnt transition in the S K-edge spectrum. This
assignment is supported by TDDFT (Figure 9b). Estimation of
S parentage per radical hole by the method of Sproules gives a
{
2
2
3
However, the presence of an additional metal-centered orbital
72
vacancy in {4[ASN] } will give rise to an Ir 2s → {α(Ir 5d) +
value of 50%, in excellent agreement with the DFT-derived
value of 52%.
2
2
2
[
√(1 − α )](S 3p)} pre-edge excitation. This prediction is
69
supported by TDDFT calculations (Figure 8b). Because of
the excessive core-hole lifetime broadening endemic to XAS of
CONCLUSIONS
7
0,71
■
1
heavier elements,
this excitation will be unresolved from
Although its lighter congeners decompose following oxidation,
the rising edge. Consequently, its presence gives the appearance
of an edge shift to lower energy, despite the fact that the metal
has been partially oxidized.
[ASN] and 1[PPh ] may be cleanly oxidized by a single
3 4 3
electron to yield dimeric {4[ASN]2} and monomeric
2
4
[PPh ] , respectively. Spectroelectrochemistry and EXAFS
The S K-edge XAS of 1[ASN] and {4[ASN] } were
4 2
3
2 2
2
−
suggest that the complex dianion [Ir(mnt) ] may be further
oxidized to [Ir(mnt) ] . Experimentally calibrated DFT
recorded to probe the S parentage in the SOMO. Pre-edge
3
−
features observed in S K-edge XAS may be assigned as S 1s →
3
2
{
α(S 3d) + [√(1-α )](M nd)} transitions. The intensities of
calculations imply that the Co congener is unstable following
oxidation because an electron is removed from a purely ligand-
these transitions thus afford a metric of the S character of MOs
proximal to the ligand field. The S K-edge XAS measurements
at first glance appear at odds with the covalent bond formalism
for the interaction between complex ions in the {4[ASN]2}2
dimer. Pre-edge features near 2470 eV are diagnostic of ligand-
centered, A symmetry MO, which weakens donor strength
2
sufficient to favor dissociation. The HOMOs of the formally
Rh(III) and Ir(III) complexes are of E symmetry, permitting
admixture of M 4d/5d character. The greater M−S covalency
of the Ir species relative to the Rh congener disfavors
dissociation.
9
centered radical character in tris(dithiolene) complexes. Such a
feature is visible in the S K-edge XAS spectrum of {4[ASN]2}2
H
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
While the electronic structure of metallodithiolenes is
certainly well-trodden ground, our present comparison of
REFERENCES
■
1
(
(
(
1) Sproules, S. Prog. Inorg. Chem. 2014, 58, 1−144.
the reactivity (or lack thereof) following oxidation of group 9
tris(mnt) complexes has implications that should be considered
in the design of new transition metal catalysts. For example,
multiple catalysts have been reported that exploit metal−ligand
cooperativity such that generation of bound ligand radicals
permits bond formation with substrate molecules. This type of
cooperativity has been used for entantioselective trans-
formations. Zhang and co-workers have reported highly
asymmetric cyclopropanation and amination reactions facili-
tated by cooperativity between Co and bound radicals through
2) Eisenberg, R. Coord. Chem. Rev. 2011, 255, 825−836.
3) Eisenberg, R.; Gray, H. B. Inorg. Chem. 2011, 50, 9741−9751.
(4) Inorg. Chem. 2011, 50, 9737−9914.
(5) Eur, J. Inorg. Chem. 2012, 3, 340−580.
(6) Luca, O. R.; Crabtree, R. H. Chem. Soc. Rev. 2013, 42, 1440−
459.
1
(
(
7) Lyaskovskyy, V.; de Bruin, B. ACS Catal. 2012, 2, 270−279.
8) Suarez, A. I. O.; Lyaskovskyy, V.; Reek, J. N.; van der Vlugt, J. I.;
de Bruin, B. Angew. Chem., Int. Ed. 2013, 52, 12510−12529.
9) Sproules, S.; Wieghardt, K. Coord. Chem. Rev. 2011, 255, 837−
60.
10) Tomson, N. C.; Crimmin, M. R.; Petrenko, T.; Rosebrugh, L.
(
8
(
7
3,74
the use of chemo- or regioselective supporting ligands.
Success with this strategy requires bond formation to occur
while the radical ligand remains bound to the metal and,
consequently, under the influence of the supporting ligand.
Work from de Bruin and co-workers has shown that lability of
the radical ligand on Co can interfere with such synthetic
schemes, with bond-formation occurring after the ligand radical
E.; Sproules, S.; Boyd, W. C.; Bergman, R. G.; DeBeer, S.; Toste, F. D.;
Wieghardt, K. J. Am. Chem. Soc. 2011, 133, 18785−18801.
(11) Tenderholt, A. L.; Szilagyi, R. K.; Holm, R. H.; Hodgson, K. O.;
Hedman, B.; Solomon, E. I. J. Inorg. Biochem. 2007, 101, 1594−1600.
(12) Ray, K.; DeBeer George, S.; Solomon, E. I.; Wieghardt, K.;
Neese, F. Chem.Eur. J. 2007, 13, 2783−2797.
7
5,76
(13) Tsujimoto, Y.; Tassel, C.; Hayashi, N.; Watanabe, T.; Kageyama,
H.; Yoshimura, K.; Takano, M.; Ceretti, M.; Ritter, C.; Paulus, W.
Nature 2007, 450, 1062−1065.
has dissociated from the metal.
In such cases, enantioin-
duction fails. The substitution of a late third-row metal in such
systems should facilitate retention of the bound radical. Such an
approach has precedent in other work from de Bruin, notably in
(14) Bart, S. C.; Chlopek, K.; Bill, E.; Bouwkamp, M. W.; Lobkovsky,
E.; Neese, F.; Wieghardt, K.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128,
13901−13912.
(15) Knijnenburg, Q.; Gambarotta, S.; Budzelaar, P. H. Dalton Trans.
77
the report of stable Ir-carbene and Ir-nitrene radical species.
This principle will guide subsequent work from our laboratory
exploiting motifs of modern coordination chemistry, such as
second-sphere interactions and redox-noninnocence, for the
application of molecular redox processes in catalysis.
2
(
5
(
006, 5442−5448.
16) Heyduk, A. F.; Zarkesh, R. A.; Nguyen, A. I. Inorg. Chem. 2011,
0, 9849−9863.
17) Palmer, J. H.; Lancaster, K. M. Inorg. Chem. 2012, 51, 12473−
2482.
(18) Almlo
19) Pyykko
1
ASSOCIATED CONTENT
̈
fl, J.; Gropen, O. Rev. Comput. Chem. 2009, 8, 203.
■
(
(
̈
, P. Chem. Rev. 1988, 88, 563−594.
*
S
Supporting Information
20) Autschbach, J. J. Chem. Phys. 2012, 136, 150902.
Details of physical measurements, including SQUID magneto-
metry, infrared data, mass spectrometry data, calculations, and
crystallographic data. CCDC: 1031447, 1009417−1009423.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(
21) Pyykko
̈
, P. Annu. Rev. Phys. Chem. 2012, 63, 45−64.
(22) Werden, B. G.; Billig, E.; Gray, H. B. Inorg. Chem. 1966, 5, 78−
81.
(23) Aitken, R. A.; Philp, E. F.; Riddell, F. G.; Smith, M. H.
ARKIVOC 2002, 3, 63−70.
(24) Davison, A.; Holm, R.; Benson, R.; Mahler, W. Inorg. Synth.
2
007, 10, 8−26.
AUTHOR INFORMATION
■
(25) Bruker; Bruker AXS, Inc. SAINT, Version 7.68A; Bruker AXS:
Madison, WI, 2009.
Corresponding Author
(
26) Sheldrick, G.; SADABS, Version 2008/2; University of
*E-mail: kml236@cornell.edu.
Gottingen: Germany, 2003.
̈
Notes
(27) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.;
Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339−341.
The authors declare no competing financial interest.
(
(
(
28) Webb, S. Phys. Scr. 2005, 2005, 1011.
29) Rehr, J. J.; Albers, R. Rev. Mod. Phys. 2000, 72, 621.
30) Rehr, J. J.; Kas, J. J.; Prange, M. P.; Sorini, A. P.; Takimoto, Y.;
ACKNOWLEDGMENTS
■
Vila, F. C. R. Phys. 2009, 10, 548−559.
31) Neese, F. ORCA, an ab initio, Density Functional and
Semiempirical program package, Version 2.9.0; Max-Planck-Insitut
fur Bioanorganische Chemie: Mulheim and der Ruhr, 2004.
32) Becke, A. D. Phys. Rev. A 1988, 38, 3098−3100.
K.M.L. gratefully acknowledges the Cornell University College
of Arts and Sciences for startup funding. T.J.M. thanks the
Oticon Foundation and Knud Højgaards Foundation for travel
grants in support of this work. J.B. acknowledges support from
the Danish Research Councils for Independent Research (12-
(
̈
̈
(
(
33) Perdew, J. P. Phys. Rev. B: Condens. Matter 1986, 33, 8822−
1
1
25226). J.W.H.U. is an NSF Graduate Research Fellow (DGE-
144153). This work is based in part upon research conducted
8824.
(34) Pantazis, D. A.; Chen, X.-Y.; Landis, C. R.; Neese, F. J. Chem.
Theory Comput. 2008, 4, 908−919.
at the Stanford Synchrotron Radiation Lightsource. Use of the
Stanford Synchrotron Radiation Lightsource, SLAC National
Accelerator Laboratory, is supported by the U.S. Department of
Energy (DOE), Office of Science, Office of Basic Energy
Sciences under Contract No. DE-AC02-76SF00515. The SSRL
Structural Molecular Biology Program is supported by the DOE
Office of Biological and Environmental Research and by the
National Institutes of Health, National Institute of General
Medical Sciences (including P41GM103393).
(35) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7,
3
(
297−3305.
36) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem.
Phys. 1998, 108, 4783−4796.
(
(
37) van Wu
̈
llen, C. J. Chem. Phys. 1998, 109.
38) Klamt, A.; Schuurmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 2,
7
99−805.
(39) Stephens, P.; Devlin, F.; Chabalowski, C.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623−11627.
I
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(
(
(
(
(
40) Neese, F. J. Chem. Phys. 2001, 115, 11080−11096.
41) Neese, F. J. Chem. Phys. 2005, 122, 034107.
42) Neese, F. J. Chem. Phys. 2003, 119, 9428−9443.
43) McCleverty, J. A. Prog. Inorg. Chem. 1968, 10, 49−221.
44) Burns, R.; McAuliffe, C. Adv. Inorg. Chem. Radiochem. 1979, 22,
3
03−348.
45) McCleverty, J.; Locke, J.; Wharton, E.; Gerloch, M. J. Chem. Soc.
A 1968, 816−823.
46) Schelter, E. J.; Jitendra, K. B.; Galan
R. Inorg. Chem. 2003, 42, 4256−4258.
47) Dunbar, K. R.; Schelter, E. J.; Tsukerblat, B. S.; Ostrovsky, S. M.;
Mirovitsky, V. Y.; Palii, A. V. Polyhedron 2003, 22, 2545−2556.
48) Dunbar, K. R.; Schelter, E. J.; Palii, A. V.; Ostrovsky, S. M.;
(
(
́ ́
-Mascaros, J. R.; Dunbar, K.
(
(
Mirovitsky, V. Y.; Hudson, J. M.; Omary, M. A.; Klokishner, S. I.;
Tsukerblat, B. S. J. Phys. Chem. A 2003, 107, 11102−11111.
(
́
49) Uzelmeier, C. E.; Bartley, S. L.; Fourmigue, M.; Rogers, R.;
Grandinetti, G.; Dunbar, K. R. Inorg. Chem. 1998, 37, 6706−6713.
50) Balch, A. L.; Dance, I. G.; Holm, R. H. J. Am. Chem. Soc. 1968,
0, 1139−1145.
51) McNamara, W. R.; Han, Z.; Yin, C.-J. M.; Brennessel, W. W.;
Holland, P. L.; Eisenberg, R. Proc. Natl. Acad. Sci. U. S. A. 2012, 109,
5594−15599.
52) Hendrickson, A.; Martin, R.; Taylor, D. Aust. J. Chem. 1976, 29,
69−273.
53) Raston, C. L.; White, A. H. J. Chem. Soc., Dalton Trans. 1975,
(
9
(
1
(
2
(
2
(
(
410−2418.
54) Halbert, T. R.; Stiefel, E. I. Inorg. Chem. 1989, 28, 2501−2503.
55) Matsubayashi, G.; Maikawa, T.; Tamura, H.; Nakano, M.;
Arakawa, R. J. Chem. Soc., Dalton Trans. 1996, 1539−1544.
56) Yan, Y.; Mague, J. T.; Donahue, J. P.; Sproules, S. Chem.
Commun. 2015, DOI: 10.1039/c4cc09397f.
(
(57) Noodleman, L. J. Chem. Phys. 1981, 74, 5737−5743.
(58) Brown, C. A.; Pavlosky, M. A.; Westre, T. E.; Zhang, Y.;
Hedman, B.; Hodgson, K. O.; Solomon, E. I. J. Am. Chem. Soc. 1995,
17, 715−732.
59) Sutter, D.; Dreizler, H.; Rudolf, H. D. Z. Naturforsch., A: Phys.
Sci. 1965, 20, 1676.
1
(
(
(
(
(
(
60) Beagley, B.; McAloon, K. T. Trans. Faraday. Soc. 1971, 67, 1676.
61) Coulson, C. A.; Fischer, I. Philos. Mag. 1949, 40, 386−393.
62) Jursic, B. S. Int. J. Quantum Chem. 1997, 62, 291−296.
63) Neese, F. J. Am. Chem. Soc. 2006, 128, 10213−10222.
64) Jahn, H. A.; Teller, E. Proc. R. Soc. London, Ser. A 1937, 220−
2
35.
(
(
(
(
(
65) Bates, C. A. Phys. Rep. 1978, 35, 187−304.
66) Krebs, J.; Stauss, G. Phys. Rev. B: Condens. Matter 1977, 15, 17.
67) De Groot, F. Chem. Rev. 2001, 101, 1779−1808.
68) Glatzel, P.; Bergmann, U. Coord. Chem. Rev. 2005, 249, 65−95.
69) DeBeer George, S.; Petrenko, T.; Neese, F. J. Phys. Chem. A
2
008, 112, 12936−12943.
70) Krause, M. O.; Oliver, J. J. Phys. Chem. Ref. Data 1979, 8, 329−
38.
71) Ham
Lett. 1991, 67, 2850.
72) Sproules, S.; Weyhermu
Inorg. Chem. 2011, 50, 12623−12631.
73) Xu, X.; Lu, H.; Ruppel, J. V.; Cui, X.; Lopez de Mesa, S.; Wojtas,
L.; Zhang, X. P. J. Am. Chem. Soc. 2011, 133, 15292−15295.
74) Lu, H.; Hu, Y.; Jiang, H.; Wojtas, L.; Zhang, X. P. Org. Lett.
012, 14, 5158−5161.
75) Paul, N. D.; Chirila, A.; Lu, H.; Zhang, X. P.; de Bruin, B.
Chem.Eur. J. 2013, 19, 12953−12958.
76) Paul, N. D.; Mandal, S.; Otte, M.; Cui, X.; Zhang, X. P.; de
Bruin, B. J. Am. Chem. Soc. 2014, 136, 1090−1096.
77) De Bruin, B.; Hetterscheid, D. G.; Koekkoek, A. J.;
Gruetzmacher, H. Prog. Inorg. Chem. 2007, 247−354.
(
3
(
̈ ̈ ̈
alainen, K.; Siddons, D.; Hastings, J.; Berman, L. Phys. Rev.
(
̈
ller, T.; Goddard, R.; Wieghardt, K.
(
(
2
(
(
(
J
DOI: 10.1021/acs.inorgchem.5b00289
Inorg. Chem. XXXX, XXX, XXX−XXX