Synthesis, Characterization, and DFT Analysis of Bis-Terpyridyl-Based Molecular Cobalt Complexes

Article abstract of DOI:10.1021/acs.inorgchem.7b00595

Terpyridine ligands are widely used in chemistry and material sciences owing to their ability to form stable molecular complexes with a large variety of metal ions. In that context, variations of the substituents on the terpyridine ligand allow modulation of the material properties. Applying the Stille cross-coupling reaction, we prepared with good yields a new series of terpyridine ligands possessing quinoline-type moieties in ortho, meta, and para positions and dimethylamino substituents at central or distal positions. The corresponding cobalt(II) complexes were synthesized and fully characterized by elemental analysis, single-crystal X-ray crystallography, mass spectrometry, and UV-vis, ¹H NMR, and Fourier transform infrared (FT-IR) spectroscopy as well as by cyclic voltammetry (CV). Density functional theory (DFT) calculations were performed to investigate the electronic structure of all the Co(II) bis-terpyridyl molecular complexes. In this work, we show that terpyridine ligand functionalization allows tuning the redox potentials of the Co(III)/Co(II), Co(II)/Co(I), and Co(I)/Co(I) (tpy)^?- couples over a 1 V range.

Full text of DOI:10.1021/acs.inorgchem.7b00595

Article
pubs.acs.org/IC
Synthesis, Characterization, and DFT Analysis of Bis-Terpyridyl-Based
Molecular Cobalt Complexes
Safwan Aroua,† Tanya K. Todorova,† Paul Hommes,‡ Lise-Marie Chamoreau,§ Hans-Ulrich Reissig,‡
Victor Mougel,† and Marc Fontecave*,†
†Laboratoire de Chimie des Processus Biologiques, UMR 8229 CNRS, Colleg
Berthelot, 75231 Paris Cedex 05, France
̀ ́
e de France, Universite Paris 6, 11 Place Marcelin
‡Institut fur Chemie und Biochemie, Freie Universitat Berlin, Takustrasse 3, 14195 Berlin, Germany
̈
̈
§Sorbonne Universites
Paris Cedex 5, France
́ ́ ́
, UPMC Universite Paris 6, Institut Parisien de Chimie Moleculaire, UMR 8232 CNRS, 4 Place Jussieu, 75252
S
* Supporting Information
ABSTRACT: Terpyridine ligands are widely used in chemistry and material
sciences owing to their ability to form stable molecular complexes with a large
variety of metal ions. In that context, variations of the substituents on the
terpyridine ligand allow modulation of the material properties. Applying the
Stille cross-coupling reaction, we prepared with good yields a new series of
terpyridine ligands possessing quinoline-type moieties in ortho, meta, and para
positions and dimethylamino substituents at central or distal positions. The
corresponding cobalt(II) complexes were synthesized and fully characterized
by elemental analysis, single-crystal X-ray crystallography, mass spectrometry,
1
and UV−vis, H NMR, and Fourier transform infrared (FT-IR) spectroscopy
as well as by cyclic voltammetry (CV). Density functional theory (DFT)
calculations were performed to investigate the electronic structure of all the
Co(II) bis-terpyridyl molecular complexes. In this work, we show that
terpyridine ligand functionalization allows tuning the redox potentials of the Co(III)/Co(II), Co(II)/Co(I), and Co(I)/Co(I)
(tpy)•− couples over a 1 V range.
presence of ammonium acetate,13 have been proposed. Such
methods and other similar ones allowed the introduction of
aliphatic functional groups.
INTRODUCTION
■
Molecular complexes containing a metal center and one or two
terpyridine ligands have been extensively studied not only as
photosensitizers,1 redox shuttles for dye-sensitized solar cells
(DSSC),2 anolytes for redox flow batteries,3 ion sensors,4
nanowire transistors,5 and supramolecular polymers6 but also
for nonlinear optics7 and catalysis for proton8 and CO2
reduction.8a,9 The properties of such complexes have been
shown to be strongly influenced by the electronic nature of the
terpyridine ligands. The potential applications of terpyridyl
complexes as photosensitizers have been previously investigated
by Damrauer and co-workers, using an iron complex, and by
Berlinguette and co-workers, using ruthenium complexes.1
Limitations arise from the lack of efficient procedures to
synthesize new terpyridine ligands, in particular those bearing
electronically different substituents, which is still a current
research topic.10
Following the initial synthesis of terpyridine in 1932 by
oxidative coupling of pyridine using anhydrous FeCl3 salt,11
several alternative synthetic routes have been employed;
however, these have met with limited efficiency in terms of
yield and selectivity. Syntheses based on Pd(0)-catalyzed
pyridine coupling12 or oxidation of diacetylpyridine, followed
by a condensation with an α,β-unsaturated aldehyde in the
Syntheses using Hiyama, Suzuki, and Stille cross-coupling
reactions have recently been reported.14 In particular, the
application of Stille cross-coupling for the preparation of
terpyridines achieved in 1996 by Sauvage and co-workers14c
allowed the isolation of a variety of substituted terpyridines.
Surprisingly little has been done to study the electronic effects
of these ligands on the physicochemical properties of their
metal complexes.8a
In this work, we have studied the new series of substituted
terpyridine ligands L2−L8, shown in Figure 1, with the parent
terpyridine ligand L1 as a reference. The introduced functional
groups were chosen to allow investigating the influence of a
broad range of electronic environments. Ligand L2 holds two
highly antagonistic electronic groups: one electron-donating
substituent (dimethylamino) on the central ring and two
electron-withdrawing substituents (trifluoromethyl) on the
distal rings. L2 was chosen to investigate the influence of
functionalization at the 4-distal position relative to the 4-central
Received: March 6, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 1. New terpyridine-based ligands prepared in this study (L1 is used as a reference).
Figure 2. Cobalt bis-terpyridyl-based molecular complexes (C1−C8) investigated in this work.
position, with special interest in compensating effects. Ligands
L3−L5 holding quinolines in three different orientations (L3,
ortho; L4, meta; L5, para) were designed as compounds with
the potential ability to delocalize a negative charge. Ligands
L6−L8 form a class of strongly electron enriched dimethyla-
mino-based terpyridines. The number of 4-dimethylamino
groups varied from one on the central ring in L6 to two on
distal rings in L7 or one on each ring in L8. In addition, ligands
L6−L8 possess a methyl group at the 2-position acting as a
weak electron donor.
The corresponding cobalt(II) bis-terpyridyl molecular
complexes C1−C8 (Figure 2) were prepared and characterized
by single-crystal X-ray diffraction, 1H NMR, FT-IR and UV−vis
spectroscopy, mass spectrometry, and cyclic voltammetry, as
well as by DFT computations. This provides a means to
determine the influence of the electronic effects of the
terpyridine substituents on the physicochemical properties of
the complexes.
RESULTS AND DISCUSSION
■
Synthesis of Substituted Terpyridine Ligands. Syn-
thesis of the ligands using Hiyama, Negishi, and Suzuki cross-
coupling reactions proved unsuccessful, affording only traces of
products. The various reaction conditions investigated are
summarized in Scheme S3 in the Supporting Information.
Alternatively, we investigated a synthetic route based on the
Stille cross-coupling reaction, as described in Figure 3. Two
different strategies were followed: (i) for the preparation of
quinoline derivatives (L3−L5), 2-halopyridine 1 was coupled
to 2,6-bis(organostannyl)pyridine 2; (ii) for terpyridines having
a dimethylamino group on the central ring (L2, L6), 2-
(organostannyl)pyridine 3 was coupled to 2,6-diiodopyridine 4
(Figure 3). Both strategies provided sufficient amounts of
product, with isolated yields of 37−94%. Ligands L7 and L8
having distal dimethylamino groups were prepared according to
previously reported procedures involving the coupling of 2,6-
pyridinedicarboxylic acid derivatives 5 with β-ketoenamine 6
and a subsequent cyclocondensation reaction to 7 followed by
activation and aromatic nucleophilic substitution.15
B
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 3. Synthetic strategies applied to obtain ligands L2−L8.
Figure 4. Crystal structures of molecular complexes C1 and C3−C8. Hexafluorophosphate counterions, hydrogen atoms, and solvent molecules
were omitted for clarity (ellipsoid representations at the 50% probability level).
Synthesis of [Co(terpyridyl)2](PF6)2 Complexes. The
corresponding bis-substituted cobalt(II) complexes C1−C8
were obtained by reacting stoichiometric ratios of the ligand
and cobalt dichloride in methanol under reflux followed by
precipitation of the complex through anion exchange with
ammonium hexafluorophosphate (NH4PF6).16 All complexes
are paramagnetic, as confirmed by their 1H NMR spectra
showing peaks at up to ca. 120 ppm (Figures S12−S19 in the
Supporting Information). For all complexes the number of
peaks in proton NMR at 300 K is consistent with a C2v
symmetry. The bands assigned to the ligands in the FT-IR
spectra of complexes C1−C8 are characterized by significant
shifts reflecting complexation with cobalt(II) (Figure S49 in the
Supporting Information). The electronic effects of the ligands
can be probed by UV−vis spectroscopy: (i) the π−π*
transition bands (allowed transitions) in the UV−vis spectra
are red-shifted upon complexation with Co(II) (e.g., λmax(L1)
277 nm and λmax(C1) 317 nm, Figures S50−S57 in the
Supporting Information) and (ii) light absorption by complexes
C1−C8 starts at much lower energy (ca. 500−600 nm), in
comparison to the free ligands (below 400 nm) (Figures S50−
S57).
Crystal Structures. Crystals were obtained by slow
evaporation of acetone/toluene (2/1 v/v) solutions of the
complexes. The crystal structures were determined for C1 and
C3−C8 (single crystals of C2 of sufficient quality could not be
obtained) and are shown in Figure 4 (crystallographic
parameters are given in Tables S1 and S2 in the Supporting
Information). Distorted-octahedral geometries (ML6) were
found for all complexes, with two ligands per Co center and
C
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. (a) Deviation from orthogonality descriptors: acute angle α and obtuse angle β (averaged Ndist−Co−Ndist angles). (b) Deviation from
planarity descriptors d1, d2, and d3 (the three N atoms of the ligand define a plane). Arbitrarily, the central ring is always defined above the plane.
Deviations above or below the plane are denoted with + and −, respectively.
Table 1. Descriptors for the Deviation from Orthogonality (α and β in deg) and for the Deviation from Planarity (d1, d2, and d3
a
in Å)
C1
C2
C3
91.3
C4
90.3
C5
C6
91.0
C7
92.1
C8
92.4
exptl (standard deviation)
α (av)
β (av)
d1 (av)
d2 (av)
d3 (av)
α (av)
β (av)
d1 (av)
d2 (av)
d3 (av)
α (av)
β (av)
d1 (av)
d2 (av)
d3 (av)
93.15
93.59
+0.00(6)
−0.07(6)
−0.1(2)
91.7
89.3
97.3
96.0
92.6
95.8
94.5
94.5
+1.14(8)
−0.7(4)
−0.62(1)
91.9
+0.2(1)
−0.16(2)
+0.3(5)
91.6
+0.0(0)
+0.13(0)
−0.75(0)
+0.18(0)
89.8
+0.3(2)
−0.5(3)
+0.0(4)
90.6
+0.2(2)
−0.0(1)
−0.04(3)
90.5
+0.39(4)
−0.39(4)
DFT
doublet
quartet
91.7
92
81.1 (89.7)
103.5 (94.5)
0.0 (+0.02)
−1.23 (+0.51)
+1.23 (−0.70)
85.2 (91.3)
104 (97.4)
91.7
92
91.6
94.6
93.5
93.9
0.0
+0.01
+0.01
+0.01
90.3
+0.88
−0.41
−0.37
92.7
+0.01
+0.01
−0.01
93.8
0.0
0.0
0.0
0.0
+0.62
−0.62
90.8
+0.58
−0.60
91.9
+0.63
−0.65
91.9
0.0
94
94
97
95
93.9
97.2
96.2
96.3
+0.02
+0.01
+0.01
+0.02
−0.08
+0.09
+1.30
−0.36
−0.40
+0.01
+0.04
+0.04
+0.46 (+0.21)
−1.01 (+0.58)
+1.13 (−0.61)
+0.31
+0.33
−0.80
+0.29
+0.30
−0.64
+0.38
+0.24
−0.73
a
Experimental and corresponding computed values (M06-L/6-311+G(d,p) level of theory, 0 K) for both low-spin doublet and high-spin quartet spin
states of complexes C1−C8 are reported. For C5, the computed values in parentheses were obtained from an optimized structure with four toluene
solvent molecules (see Figure S75 in the Supporting Information).
Table 2. Average Co−Central Nitrogen or Co−Distal Nitrogen Bond Lengths (Å) in the Crystal Structures and the
Corresponding Computed Bond Lengths (M06-L/6-311+G(d,p) Level of Theory, 0 K) for both Low-Spin Doublet and High-
a
Spin Quartet Spin States of Complexes C1−C8
DFT
exptl (standard deviation)
doublet
quartet
Co−Ncent (av)
Co−Ndist (av)
Co−Ncent (av)
Co−Ndist (av)
Co−Ncent (av)
Co−Ndist (av)
C1 (100 K)
C2
2.05(3)
2.17(2)
1.898
1.901
1.925
1.898
1.899
1.905
1.904
1.906
2.109
2.111
2.091
2.105
2.187
2.199
2.182
2.185
2.089
2.068
2.101
2.090
2.081
2.058
2.090
2.073
2.185
2.187
2.167
2.179
2.262
2.280
2.248
2.257
C3 (100 K)
C4 (100 K)
C5 (200 K)
C6 (200 K)
C7 (200 K)
C8 (200 K)
2.06(1)
1.88(3)
2.04(1)
2.02(0)
2.046(5)
2.043(3)
2.15(1)
2.06(6)
2.272(1)
2.27(8)
2.22(2)
2.23(3)
a
Values in boldface are the computed Co−central nitrogen distances in agreement with the measured values.
three nitrogen atoms of each ligand coordinating the cobalt(II)
ion. For each complex, two hexafluorophosphate counterions
are present in the unit cell to balance the charges.
We first examined the structural changes induced by the
different substituents on the ligands. We defined the
orthogonality between the ligands in each complex by means
of two parameters: the acute α and the obtuse β Ndist−Co−Ndist
angles as shown in Figure 5a. A difference between these two
values reflects a deviation from orthogonality. The results are
summarized in Table 1. While in complex C1, as well as in C4,
C7, and C8, the angles α and β have comparable values with
differences <3°, significant deviations are observed in the case
of complexes C3, C5, and C6 (Table 1), with α and β values of
89 and 97°, respectively (C5), and 91 and 96°, respectively (C3
and C6). The largest deviation was observed for complex C5,
likely due to intermolecular π−π interactions with neighboring
D
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Cyclic voltammetry experiments of C1−C8 (1.0 mM) in acetonitrile as solvent with tetrabutylammonium perchlorate supporting
electrolyte (0.1 M). The scan rate was 50 mV/s (the first two reductions were measured). The molecular orbitals on which the electrons reside for
each of the two one-electron-reduction events are depicted for each complex.
complexes and cocrystallized toluene solvent molecules (see
below).
than the distances between the cobalt ion and the distal
nitrogen (Co−Ndist). Surprisingly, the Co−Ncent distance in
complex C4 is significantly shorter (1.88 Å) than the
corresponding Co−Ncent distances in the other complexes
that vary in the range of 2.02−2.06 Å. The Co−Ndist bond
lengths in that complex (2.06 Å) are also significantly shorter
than the Co−Ndist distances in the other complexes (2.15−2.27
Å). Since the redox state of these systems is the same, it is likely
that the uniqueness of C4 is due to adoption of a different spin
state (see below).
The steric bulk of the ligands was measured using the VBur
(% buried volume) parameter defined as the fraction of the
total volume of a sphere, centered on the metal, occupied by
the ligand, using the crystallographic data (see Figure S58, in
the Supporting Information for details).17 For the calculations,
the sphere radius was chosen as the distance between the cobalt
center and the centroid of the two carbon atoms adjacent to the
central N atom. That distance in C1 is 2.7 Å and was used as a
reference. The radius R and the mesh spacing were fixed to 3.5
and 0.05 Å, respectively. Consequently, the ligands L1−L8
could be divided into two classes in terms of steric hindrance
(Table S3 in the Supporting Information): L1−L4 define the
first class, occupying a low buried volume between 15.5 and
Additionally, deviation from planarity of the ligands in these
complexes was quantified using the three descriptors d1, d2, and
d3 (see Figure 5b): they account for the deviation of each
pyridyl ring with respect to a plane defined by the three
nitrogen atoms of one ligand. The descriptors measure the
corresponding distance between the plane and the 4-pyridyl
position in each pyridyl ring (see Table 1).
The ligands in C1 are fully planar (<0.07 Å deviation), while
those in C8 deviate slightly from ideal planarity (d1 is only +0.2
Å). Ligands in C3 show the largest deviation, which likely
originates from the steric repulsion of hydrogens from central
and distal 3-pyridyl positions. As a result, the central pyridyl
ring is displaced upward (d1 = +1.14 Å) and the distal rings are
displaced downward (d2 = −0.70 Å, d3 = −0.62 Å). In contrast,
ligands in C4−C7 have a distorted shape, with a d2 value as
high as 0.75 Å in C6. It seems that planarity is obtained when
the three pyridyl moieties of the ligand are identical, while
significant distortions occur when they differ in their
substituents.
Table 2 reveals that the bond lengths between the cobalt ion
and the central nitrogen (Co−Ncent) are 0.09−0.25 Å shorter
E
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Table 3. Measured Reduction Potential (V/Fc0/+) of C1−C8 in 0.1 M TBAClO4 Electrolyte Solution in CH3CN and Oxidation
a
Potential in 0.1 M LiTFSI Electrolyte Solution in CH3CN
C1
C2
C3
C4
C5
C6
C7
C8
Eox1
Ered1
Ered2
−0.07
−1.17
−2.04
−0.04
−1.24
−1.91
−0.09
−0.82
−1.61
−0.21
−1.32
−2.13
+0.84
−0.89
−1.68
+0.53
−1.54
−2.26
+0.25
−1.52
−2.30
+0.03
−1.92
−2.49
a
See the Supporting Information for details. Eox1 is the first oxidation of [Co(bis-terpyridyl)]2+ and Ered1 and Ered2 are the first and second one-
electron reductions of [Co(bis-terpyridyl)]2+, respectively.
19.0%, while L5−L8 define a second class, with a larger buried
volume of 24.5−26.5%. Such significant differences are
attributed to the presence of a methyl- or aryl-type carbon on
the 2-position of the distal pyridyl ring.
combined with its very high solubility in acetonitrile, makes it a
promising compound for redox flow battery applications.
Initial attempts to measure the oxidation potential of C1
were performed under the conditions used for the reduction
(glassy-carbon electrode, 0.1 M TBAClO4 electrolyte in
acetonitrile, Figure S62 in the Supporting Information).
However, the strong increase of current observed at ∼−0.1 V
indicated formation of a solid deposit on the electrode.
Interestingly, a fully reversible wave was observed when DMF
was used as a solvent (Figure S63 in the Supporting
Information). This could be overcome in acetonitrile by
changing the electrolyte. As shown in Figure S64 in the
Supporting Information, lithium bis(trifluoromethane sulfoni-
midate) (LiTFSI) proved to be the best electrolyte and was
further used to study C2−C8 (Figure S64). The choice of
electrode material was critical, and thus, we utilized both glassy-
carbon and gold electrodes (Figure S65 in the Supporting
Information). Specifically in the case of C2, no oxidation wave
could be detected on a glassy-carbon electrode, while a fully
reversible oxidation was observed on a gold electrode. C1−C4
display a fully reversible oxidation with a peak-to-peak
separation below 100 mV (Figure S65). C5−C8 have less
resolved redox waves, with large peak-to-peak separations and/
or irreversibility. Such an electrochemical behavior has been
previously observed and is likely due to a slow electron self-
exchange rate between the CoII and CoIII centers at the
electrode surface, in comparison to the electron transfer rate
between the electrode and the cobalt ions.18 The Co(III/II)
redox potentials are given in Table 3, and a comparison with
other previously reported cobalt bis-terpyridyl systems is given
in Figure S66 and Table S4 in the Supporting Information.18,19
It is shown that this class of systems can span a broad range of
potentials, more than 1 V. As expected, in the C6−C8 series
the Co(III/II) redox potential is shifting to more cathodic
values when more amino substituents are added on the tpy
ligand. Interestingly, the potential of C6 (+0.53 V) was much
more positive than that of R1 (−0.39 V), a previously reported
cobalt bis-terpyridyl complex with a N-pyrrolidine at the 4-
central position (Figure S66 and Table S4). This is likely
explained by the strong effect of the methyl groups in C6. It
appears that the factors governing Co(III/II) redox potentials
are rather complex and not straightforward, as strong variations
can result from rather minor structural and electronic
differences.
Cyclic Voltammetry. In order to investigate the influence
of the terpyridine substituents on the redox properties of the
complexes, cyclic voltammograms (CVs) of complexes C1−C8
were recorded in acetonitrile solution. The data are shown in
Figure 6 and in Table 3. All potentials are given with respect to
ferrocene/ferrocenium redox potential. The first reduction
wave for all complexes is reversible, while the reversibility of the
second reduction is dependent on the ligand used. As the most
cathodic wave was irreversible in the case of C2 and C4, the
scan rate was varied from 50 to 1000 mV/s. While this had no
effect on C2, C4 showed a reversible signal at high scan rates
(above 100 mV/s). On the other hand, the CV of C2 in DMF
at a scan rate of 1000 mV/s displays reversible features (Figure
S60 in the Supporting Information). The observed irrever-
sibility might result from a structural or chemical trans-
formation of the system, possibly a decoordination of the ligand
upon reduction.
As expected, the measured reduction potential values
strongly depend on the ligand structure, spanning over a
broad range of ca. 1.0 V (compare C3 and C8). The parent
complex C1 was used as a reference, with Ered1 = −1.17 V and
Ered2 = −2.04 V. The CV of C2 (Figure 6) shows that two
electron-withdrawing CF3 groups in the distal position
compensate for the effect of one electron-donating N(CH3)2
group on the central position (C2, Ered1 = −1.24 V and Ered2
=
−1.91 V). Interestingly, the quinoline derivatives L3 and L5
(ortho and para isomers) act as electron-deficient ligands,
leading to anodically shifted reduction waves (C3, Ered1 = −0.82
V, Ered2 = −1.61 V; C5, Ered1 = −0.89 V, Ered2 = −1.68 V). In
the case of complex C4, the measured redox potential values
are found in the range of those of C1 (C4, Ered1 = −1.32 V,
Ered2 = −2.13 V).
As far as the ligands with dimethylamino groups were
concerned, two main conclusions could be drawn. First, as
revealed by the similar redox potential values of C6 and C7
(C6, Ered1 = −1.54 V; Ered2 = −2.26 V; C7, Ered1 = −1.52 V;
E
red2 = −2.30 V), the effect of a single substituent on the central
pyridyl moiety (C6) is equivalent to that of one on each of the
distal positions (C7). This is consistent with the Co−Ncent
bond length being ca. 0.2 Å shorter than the Co−Ndist lengths,
suggesting a stronger interaction of the cobalt with the central
pyridyl ring. Second, introducing a dimethylamino group on
each of the three rings generates the most electron-rich ligand,
The Co(III/II) redox potentials cover a broad range of
potentials, varying from −0.21 V to +0.84 V vs Fc+/Fc. In the
context of dye sensitized solar cell applications,2 the very high
oxidation potentials observed with complexes C5−C7 (+0.84,
+ 0.53, and +0.25 V vs Fc+/Fc, respectively) could be utilized
with various organic photosensitizers.
Quantum Chemical Calculations. Density functional
theory (DFT) calculations were performed to investigate the
electronic structure of complexes C1−C8 and to better
L8, resulting in very negative potential values for C8 (Ered1
=
−1.92 V and Ered2 = −2.49 V). On the basis of these
considerations, the electron-donating power of the ligands
could be classified as follows: L8 > L7 ≈ L6 > L4 > L1 > L2 >
L5 > L3. Interestingly, the very cathodic potential of −2.30 V
with full reversibility observed in the case of complex C7,
F
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
understand the effect of the substituents on the structure and
physicochemical properties of the complexes. The electronic
and structural properties of [Co(tpy)2]2+, referred to as C1 in
this work, have previously been studied by Wieghardt et al.20
and Hauser et al.21
The DFT-computed descriptors also indicate that the ligands
in all complexes deviate significantly from planarity, except for
both spin states of C2 and C4. In C5 the p-quinoline units are
1.0−1.2 Å out of the plane (experimental d2,3 values are 0.4
Å), which is corrected to the largest deviation of ∼0.7 Å when
four solvent molecules are explicitly included in the geometry
optimization of the complex. Even greater deviation from
planarity of the ligands is calculated for the o-quinoline-based
complex C3 (d1 = 1.30 Å vs the corresponding experimental
value of 1.14 Å), owing to the large steric repulsion of
hydrogens from the central and distal pyridyl rings within each
ligand L3. Introducing two electron-withdrawing CF3 groups
on the distal pyridyl units along with a single electron-donating
N(CH3)2 group on the central ring in C2 has virtually no effect
on the geometry, and the structure remains essentially identical
with that of C1. For the terpyridines containing dimethylamino
groups, C6−C8, our calculations consistently find distortion of
planarity for their doublet spin states by ∼0.6 Å in terms of d2
and d3 descriptors, while d1 = 0. The high-spin-state structures
are more planar (displacements of ∼0.4 Å maximum), with the
largest deviations found for the d3 values in the range of 0.6−
0.8 Å.
All systems contained Co(II) ions in a d7 electronic
configuration, thus potentially allowing the existence of two
2
4
different spin states: a E low-spin doublet and a T high-spin
quartet state. Our calculations show that, for all complexes, the
quartet states are energetically more favorable (within 6 kcal
mol−1 for C1−C5 and as high as 9.4, 9.5, and 10.7 kcal mol−1
for the three dimethylamino-containing systems C6−C8,
respectively) (Table S5 in the Supporting Information). The
optimized geometries are in very good agreement with the
experimental X-ray data (Tables 1 and 2), with the Co−N bond
lengths within 0.04 Å of the corresponding values in the crystal
structure and the angles within 2°. For both doublet and
quartet spin states the Co−Ncent bond lengths are systemati-
cally shorter by ∼0.2 Å in comparison to the Co−Ndist lengths,
which is in line with our experimental findings. Moreover, the
Co−Ncent bond lengths vary in the narrow range of 1.90−1.93
Å for the low-spin states, while for the high-spin states these
distances are systematically longer by 0.15−0.19 Å, varying in
the range of 2.06−2.10 Å. The Co−Ndist bond lengths vary
slightly in the range of 2.09−2.20 and 2.17−2.28 Å for the low-
and high-spin states, respectively. Hence, the Co−Ncent bond
length can be used as a probe for the spin state of the complex,
which is in line with having six-coordinate high- and low-spin
CoII (t2g6eg1 and t2g5eg2) in an octahedral ligand field. In the
low-spin complexes, the Jahn−Teller distortion will cause
significant tetragonal compression, manifested in elongation of
the equatorial metal−ligand (Co−Ndist) bonds, which coupled
with the rigid structure of the tpy ligands will cause the
shortening of the Co−Ncent bond lengths. Our findings are in
agreement with previous results on cobalt polypyridyl based
complexes ([Co(tpy)2]2+, [Co(tbpy)3]2+, and [Co(phen)3]2+,20
as well as [Co(tpy)2]2+ and [Co(bpy)3]2+ 21). A closer
inspection of Table 2 thus reveals that best agreement with
our crystal structures is obtained when complexes C1−C3 and
C5−C8 are in their high-spin quartet state, while only C4 is
best described as a low-spin doublet. One should note that, as
demonstrated for C1, the spin state is strongly influenced by
the environment of the complex, depending on parameters
such as the nature of the counterion, the nature and number of
solvent molecules in crystalline media or solution, and the
temperature.22 Table 1 shows that the computed acute α and
obtuse β Ndist−Co−Ndist angles are also in excellent agreement
with the experimental values, except for complex C5. The two
L5 ligands in complex C5 deviate significantly from
orthogonality, due to favorable π−π stacking between the p-
quinoline moieties (see Figure S75 in the Supporting
Information). Test calculations with different DFT functionals,
such as B3LYP-D3, B97D, and M06, predicted virtually the
same geometry. A closer inspection of the C5 crystal structure
revealed that such dispersive interactions are prevented by the
existence of four aromatic rings: two toluene solvent molecules
and two central pyridyl units from the neighboring C5
complexes in the unit cell. We have thus performed an
optimization of C5 with four toluene molecules between the
quinoline ligands, and indeed the perpendicular arrangement of
the L5 ligands is maintained, in good agreement with the
experimental structure (see Table 1).
The spin density plots of complexes C1−C8 (Figures S67−
S74 in the Supporting Information) reveal the dominant
contribution of Co(II), with ρ = 0.97−0.99 in the low-spin
doublet state and ρ = 2.73−2.87 in the high-spin quartet state
and very little density on the pyridyl N atoms. Figures S67−S74
show the frontier molecular orbitals of complexes C1−C8. The
low-spin doublet states are characterized by a HOMO that is a
pure Co 3d orbital, except in C2 and C6, where the N(CH3)2
groups at the central rings introduce an axial symmetry and the
HOMO is composed of Co 3d and C and N 2p orbitals in the
same plane. The singly occupied MO (SOMO) is the σ*
2
2
antibonding Co−Ndist orbital composed of Co 3dx −y and Ndist
2p, while the LUMO orbital is the σ* antibonding Co−Ncent
2
orbital composed of Co 3dz and Ncent 2p. As far as the high-
spin quartets are concerned, the HOMO orbital is either a
ligand-based orbital (C1, C3, C5, C7, and C8) or an
2
2
antibonding Co−Ndist orbital composed of Co 3dx −y and N
2p as in C2, C4, and C6. The three unpaired electrons reside
on two ligand-based orbitals and the Co−Ncent σ* antibonding
2
orbitals of Co 3dz , while the LUMO is always a ligand-based
orbital. Thus, the main difference in the spin states in all
2
complexes C1−C8 is the population of the Co 3dz orbital in
the quartet states, decreasing the back-bonding of Co to the
central pyridyl ring, thus providing an explanation of the change
in the Co−Ncent distance, which is elongated by ∼0.2 Å when
passing from a low-spin to a high-spin state.
Finally, our calculations show that the first electron reduction
of complexes C1−C8 is always metal-based and corresponds to
the reduction of CoII to CoI, with an electron occupying the
2
antibonding σ* Co−Ncent orbital (having mainly 3dz
character). The second reduction, on the other hand,
corresponds to a ligand-based event (see Figure 6). While in
C1 the electron is delocalized over the entire ligands, in C2 it is
localized on the distal rings due to the electron-withdrawing
CF3 groups and in the quinoline-based C3−C5 complexes it is
localized primarily on the central ring, involving parts of the
distal quinoline rings, but not the terminal phenyl rings (C5).
Similarly, in the dimethylamino-substituted complexes C7 and
C8 the electron resides on the central ring, whereas in C6 it is
delocalized on the entire L6 ligand.
G
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
using the Gaussian 09 software package.27 A quasi-relativistic
Stuttgart/Dresden effective core potential was used on Co.28 The
integral evaluation made use of the grid defined as “ultrafine” in G09.
Frequency calculations on optimized geometries ensured that
structures were minima (zero imaginary frequency) on the potential
energy surface.
CONCLUSIONS
■
We have synthesized a series of new terpyridine ligands bearing
a variety of substituting moieties using Stille cross-coupling
reactions. The corresponding bis-substituted Co(II) complexes
were isolated and fully characterized. The crystal structures
revealed distortions in some of these systems, especially when
the terpyridine ligands were functionalized with different
substituents. The DFT calculations provided insight into
understanding the differences of Co−N bond lengths observed
in the crystal structures, assigning them to differences in terms
of spin states. In particular, the Co−Ncent bond length can be
used as a probe for differentiating the low-spin doublet from
the high-spin quartet state. In this work, we have shown that
the redox potential of the Co(III)/Co(II), Co(II)/Co(I), and
Co(I)/Co(I) (tpy)•− couples can be easily modulated over 1 V
by tuning the substituents on the tpy ligands, allowing a broad
range of applications. Study of the catalytic properties of this
class of complexes is currently ongoing.
Synthesis of L2. In a Schlenk flask, a toluene solution (7.0 mL) of
S3 (181 mg, 0.485 mmol) and S9 (0.635 g, 1.45 mmol) was degassed
under argon for 15 min. The catalyst [Pd(PPh3)4] (47.0 mg, 0.042
mmol) was added, and the solution was degassed again under argon
for 15 min. The solution was heated at 110 °C for 96 h. The solvent
was removed under vacuum, and the residue was purified by column
chromatography (cyclohexane/ethyl acetate gradient) to yield L2 as a
white powder (103 mg, 0.25 mmol, 52%). IR (ATR): νmax (cm−1)
2926 (w), 2854 (w), 2816 (w), 1589 (w), 1541 (w), 1506 (w), 1471
(w), 1431 (w), 1413 (w), 1379 (w), 1328 (w), 1276 (w), 1253 (w),
1228 (w), 1165 (w), 1124 (m), 1070 (w), 1004 (w), 991 (w), 975
(w), 920 (w), 894 (w), 840 (m), 788 (w), 700 (w), 663 (m), 607 (w).
1H NMR (300 MHz, CDCl3): δ 3.34 (s, 3H), 7.59 (d, J = 5.0 Hz, 2H),
7.91 (s, 2H), 8.8−9.0 (m, 2H). 13C NMR (75 MHz, CDCl3): δ 40.28,
104.93, 117.67, 119.72, 124.68, 149.86 (2C), 156.00−157.5 (3C). HR-
+
MS (ESI+): m/z calcd for C19H15F6N4 413.1195, found 413.1201
([M + H]+).
EXPERIMENTAL SECTION
■
Synthesis of L3. In a Schlenk flask, a toluene solution (18 mL) of
S5 (0.53 g, 1.3 mmol) and S10 (1-chloroisoquinoline) (1.0 g, 3.9
mmol, Fluorochem) was degassed under argon for 15 min. The
catalyst [Pd(PPh3)4] (0.153 g, 0.13 mmol) was added, and the
solution was degassed again under argon for 15 min. The solution was
heated at 110 °C for 96 h. The solvent was removed under vacuum,
and the residue was purified by column chromatography (cyclo-
hexane/ethyl acetate gradient). The isolated product was recrystallized
in acetonitrile and dried under vacuum to yield L3 as a white powder
(191 mg, 0.57 mmol, 44%). IR (ATR): νmax (cm−1) 3043 (w), 3007
(w), 2924 (w), 2852 (w), 1568 (w), 1435 (w), 1373 (w), 1346 (w),
1319 (w), 1249 (w), 1238 (w), 1215 (w), 1126 (w), 1089 (w), 1022
(w), 952 (w), 916 (w), 869 (w), 835 (w), 819 (w), 798 (w), 789 (w),
734 (w), 704 (w), 669 (w). 1H NMR (300 MHz, CDCl3): δ 7.51 (t, J
= 7.5 Hz, 2H), 7.68 (t, J = 7.5 Hz, 2H), 7.76 (d, J = 5.5 Hz, 2H), 7.88
(d, J = 7.5 Hz, 2H), 8.16 (s, 3H), 8.68 (d, J = 5.5 Hz, 2H), 8.75 (d, J =
7.5 Hz, 2H). 13C NMR (75 MHz, CDCl3): δ 121.40, 124.81, 126.84,
126.91, 127.62, 128.13, 130.13, 137.19, 137.89, 141.81, 157.32, 157.48.
General Considerations. NMR spectra were recorded on a
Bruker AVANCE III 300 spectrometer (Bruker BioSpin GmbH,
Rheinstetten, Germany). HR-ESI-MS spectra were recorded in
positive mode on a Waters LCT Premier XE mass spectrometer
equipped with an electrospray ionization source. The ionization was
carried out in positive mode in the m/z 80−1500 range. LR-ESI-MS
spectra of the molecular complexes were recorded in positive mode on
a PE Sciex API 3000 mass spectrometer equipped with an electrospray
ionization source. UV−vis spectra were recorded on a Cary100 UV−
vis spectrophotometer (Agilent Technologies, Santa Clara, CA, USA).
Suitable crystals of the complexes were mounted in Paratone oil and
transferred into a cold nitrogen gas stream. Intensity data were
collected with Bruker Kappa-APEX2 systems using microsource Cu
Kα or fine-focus sealed-tube Mo Kα radiation or using an Oxford
Diffraction XCallibur S κ geometry diffractometer (Mo Kα radiation,
graphite monochromator, λ = 0.71073 Å). Data collection was carried
out with the Bruker APEX2 suite of programs (APEX2 diffractometer)
and by CrysAlisPro Oxford Diffraction software (XCallibur diffrac-
tometer). Unit-cell parameter determination, integration, and data
reduction were performed with SAINT (APEX II). Alternatively,
unique intensities detected on all frames using the Oxford Diffraction
Red program were used to refine the values of the cell parameters
(XCallibur). SADABS23 (APEX II) and the ABSPACK Oxford
Diffraction programs were used for scaling and multiscan absorption
corrections. The structures were solved with WinGX and Olex2 1.2
using the SHELXT-201424 package and refined by full-matrix least-
squares methods with SHELXL-2014. All non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were placed at calculated
positions and refined with a riding model. Electrochemical measure-
ments were performed using a Bio-Logic SP300 potentiostat model
(Bio-Logic, Claix, France). CVs were recorded in a glovebox to
prevent oxygen. An electrochemical cell with a glassy-carbon electrode
as a working electrode (i.d. 1.0 mm), a platinum wire as a counter
electrode (i.d. 1 mm), and a silver wire in silver nitrate isolated from
the solution using a Vycor frit (10 mM in 0.1 M TBAClO4 solution in
acetonitrile) as a reference electrode in a 3.0 mL solution was used to
measure the CVs. A gold working electrode (i.d. 3.0 mm) was also
used when stated. FT-IR spectra were recorded on a IR Prestige-21
spectrometer with Universal ATR Sampling Accessory (Shimadzu,
Kyoto, Japan). Purifications with flash liquid chromatography were
carried out using a Reveleris X2 system (Grace, Columbia, USA)
equipped with disposable silica columns. For synthesis, the solutions
were degassed (when mentioned) by bubbling argon in the solution
for at least 20 min. Dry solvents (when mentioned) were obtained
using a MBraun Solvent Purification System.
+
HR-MS (ESI+): m/z calcd for C23H16N3 334.1339, found 334.1344
([M + H]+).
Synthesis of L4. In a Schlenk flask, a toluene solution (18 mL) of
S5 (0.53 g, 1.3 mmol) and S11 (3-chloroisoquinoline) (1.0 g, 3.9
mmol, Fluorochem) was degassed under argon for 15 min. The
catalyst [Pd(PPh3)4] (0.153 g, 0.13 mmol) was added, and the
solution was degassed again under argon for 15 min. The solution was
heated at 110 °C for 96 h. The solvent was removed in vacuo, and the
residue was purified by column chromatography (cyclohexane/ethyl
acetate gradient). The isolated product was recrystallized in
acetonitrile and dried under vacuum to yield L4 as a white powder
(170 mg, 0.51 mmol, 39%). IR (ATR): νmax (cm−1) 3057 (w), 3024
(w), 2980 (w), 2964 (w), 2926 (w), 2854 (w), 1625 (w), 1581 (w),
1562 (w), 1492 (w), 1463 (w), 1438 (w), 1427 (w), 1386 (w), 1342
(w), 1267 (w), 1255 (w), 1236 (w), 1192 (w), 1136 (w), 945 (w), 900
1
(w), 885 (w), 850 (w), 817 (m), 734 (m), 675 (w). H NMR (300
MHz, CDCl3): δ 7.64 (t, J = 7.5 Hz, 2H), 7.76 (t, J = 7.5 Hz, 2H), 8.06
(m, 5H), 8.56 (d, J = 7.5 Hz, 2H), 9.06 (s, 2H), 9.37 (s, 2H). 13C
NMR (75 MHz, CDCl3): δ 117.78, 120.93, 127.57, 127.65, 127.75,
128.81, 130.55, 136.69, 137.98, 150.19, 152.08, 155.95. HR-MS (ESI
+
+): m/z calcd for C23H16N3 334.1339, found 334.1344 ([M + H]+).
Synthesis of L5. In a Schlenk flask, a toluene solution (12 mL) of
S5 (0.36 g, 0.89 mmol) and S7 (2-iodoquinoline) (0.60 g, 2.67 mmol,
Fluorochem) was degassed under argon for 15 min. The catalyst
[Pd(PPh3)4] (0.104 g, 0.09 mmol) was added, and the solution was
degassed again under argon for 15 min. The solution was heated at
110 °C for 96 h. The solvent was removed under vacuum, and the
residue was purified by column chromatography (cyclohexane/ethyl
acetate gradient) to yield L5 as a yellowish powder (282 mg, 0.84
Computational Methods. Molecular geometries were optimized
at the M06-L25/6-311+G(d,p)26 level of density functional theory
H
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
mmol, 94%). IR (ATR): νmax (cm−1) 3051 (w), 2995 (w), 2895 (w),
2798 (w), 2763 (w), 2742 (w), 2655 (w), 1616 (w), 1595 (m), 1556
(m), 1502 (m), 1433 (m), 1423 (m), 1321 (w), 1305 (w), 1247 (w),
1236 (w), 1122 (m), 1084 (m), 1060 (w), 1014 (w), 993 (w), 956
(w), 1552 (w), 1471 (w), 1456 (w), 1386 (w), 1354 (w), 1311 (w),
1271 (w), 1222 (w), 1186 (w), 1020 (w), 875 (w), 833 (s), 821 (s),
740 (w), 667 (w). Anal. Calcd for C53H38CoF12N6P2 (C3·PhMe): C,
57.46; H, 3.46; N, 7.59. Found: C, 57.36; H, 3.35; N, 7.25. LRMS
(ESI+): m/z 870.6 ([M − PF6]+), 362.7 ([M − 2PF6]2+).
1
(w), 941 (w), 831 (m), 815 (m), 785 (m), 733 (m), 771 (w). H
1
C4: isolated yield 71%. H NMR (300 MHz, CD3CN): δ 111.46,
NMR (300 MHz, CDCl3): δ 7.85 (t, J = 7.5 Hz, 2H), 7.77 (t, J = 7.5
Hz, 2H), 7.89 (d, J = 8.5 Hz, 2H), 8.08 (t, J = 7.5 Hz, 1H), 7.22 (d, J =
8.5 Hz, 2H), 8.35 (d, J = 8.5 Hz, 2H), 7.78 (d, J = 7.5 Hz, 2H), 8.85
55.40, 46.74, 21.27, 10.15, 9.27, 7.96, 7.89, 6.08. IR (ATR): νmax
(cm−1) 3093 (w), 2972 (w), 2885 (w), 1625 (w), 1602 (w), 1568 (w),
1505 (w), 1489 (w), 1446 (w), 1390 (w), 1298 (w), 1269 (w), 1174
(w), 968 (w), 831 (s), 810 (s), 758 (w), 736 (w), 678 (w). Anal. Calcd
for C53H38CoF12N6P2 (C4·PhMe): C, 57.46; H, 3.46; N, 7.59. Found:
57.24; H, 3.36; N, 7.29. LRMS (ESI+): m/z 870.7 ([M − PF6]+), 362.7
([M − 2PF6]2+).
+
(d, J = 8.5 Hz, 2H). HR-MS (ESI+): m/z calcd for C23H16N3
334.1339, found 334.1344 ([M + H]+).
Synthesis of L6. In a Schlenk flask, a toluene solution (2.5 mL) of
S3 (65.0 mg, 0.175 mmol) and S12 (2-methyl-6-(tributylstannyl)-
pyridine) (0.20 g, 0.523 mmol, Sigma-Aldrich) was degassed under
argon for 15 min. The catalyst [Pd(PPh3)4] (17.0 mg, 0.015 mmol)
was added and the solution was degassed again under argon for 15
min. The solution was heated at 110 °C for 96 h. The solvent was
removed in vacuo, and the residue was purified by column
chromatography (cyclohexane/ethyl acetate gradient) to yield L6 as
a white powder (20 mg, 0.065 mmol, 37%). IR (ATR): νmax (cm−1)
2985 (w), 2922 (w), 2885 (w), 2802 (w), 1608 (w), 1570 (w), 1541
(w), 1506 (w), 1423 (w), 1411 (w), 1226 (w), 1141 (w), 1082 (w),
1
C5: isolated yield 90%. H NMR (300 MHz, CD3CN): δ 98.64,
72.23, 27.88, 24.14, 14.77, 4.95, 3.75, 2.33. IR (ATR): νmax (cm−1)
3093 (w), 2972 (w), 2893 (w), 1593 (w), 1516 (w), 1485 (w), 1435
(w), 1379 (w), 1342 (w), 1271 (w), 1220 (w), 1195 (w), 1148 (w),
1101 (w), 1083 (w), 1029 (w), 877 (w), 825 (s), 806 (s), 779 (m),
761 (m), 738 (m), 686 (w). Anal. Calcd for C53H38CoF12N6P2 (C5·
PhMe): C, 57.46; H, 3.46; N, 7.59. Found: C, 57.18; H, 3.43; N, 7.27.
LRMS (ESI+): m/z 870.7 ([M − PF6]+), 362.8 ([M − 2PF6]2+).
1
1
C6: isolated yield 86%. H NMR (300 MHz, CD3CN): δ 113.79,
983 (w), 852 (w), 800 (w), 744 (w), 698 (w), 661 (w), 638 (w). H
77.10, 35.85, 20.17, −0.95, −11.67. IR (ATR): νmax (cm−1) 2972 (w),
2900 (w), 1618 (w), 1604 (w), 1573 (w), 1533 (w), 1456 (w), 1438
(w), 1379 (w), 1307 (w), 1251 (w), 1166 (w), 1114 (w), 1070 (w),
1047 (w), 997 (w), 906 (w), 873 (w), 831 (s), 800 (m), 742 (w), 651
(w). Anal. Calcd for C41H46CoF12N8OP2 (C6·CH3COCH3): C, 48.48;
H, 4.56; N, 11.03. Found: C, 48.32; H, 4.50; N, 10.93. LRMS (ESI+):
m/z 812.3 ([M − PF6]+), 333.7 ([M − 2PF6]2+).
NMR (300 MHz, CDCl3): δ 7.14 (d, J = 7.5 Hz, 2H), 7.71 (t, J = 7.5
Hz, 2H), 7.76 (s, 2H), 8.39 (d, J = 7.5 Hz, 2H). 13C NMR (75 MHz,
CDCl3): δ 24.70, 39.64, 103.72, 118.49, 122.95, 136.68, 155.78,
+
156.42, 157.52 (2C). HR-MS (ESI+): m/z calcd for C19H21N4
305.1761, found 305.1766 ([M + H]+).
General Procedure for the Synthesis of Molecular Com-
plexes (except C5). To a solution of anhydrous CoCl2 (1.0 equiv,
Sigma-Aldrich) in degassed methanol (1.0 mL/15 μmol CoCl2) was
added a powder of the terpyridyl ligand (2.0 equiv). The solution was
refluxed for 3 h under argon and cooled to room temperature. The
complex was precipitated after addition of a concentrated aqueous
solution of ammonium hexafluorophosphate (8.0 equiv, Sigma-
Aldrich), and the product was collected by filtration. The product
was thoroughly washed with 20 mL portions of water (Milli-Q),
toluene, and Et2O. The obtained powder was dissolved in a minimum
amount of acetone, and toluene was added dropwise until the complex
precipitated. The precipitate was collected and dried under vacuum to
provide C1−C8.
1
C7: isolated yield 88%. H NMR (300 MHz, CD3CN): δ 61.01,
58.65, 45.21, 8.87, 7.07, −7.92. IR (ATR): νmax (cm−1) 2933 (w), 2895
(w), 2818 (w), 1610 (w), 1573 (w), 1521 (m), 1506 (w), 1489 (w),
1436 (w), 1417 (w), 1384 (w), 1369 (w), 1311 (w), 1271 (w), 1232
(w), 1186 (w), 1166 (w), 1143 (w), 1126 (w), 1105 (w), 1066 (w),
1047 (w), 985 (w), 829 (s), 813 (s), 738 (w), 694 (w). Anal. Calcd for
C42H50CoF12N10P2: C, 48.33; H, 4.83; N, 13.42. Found: C, 47.89; H,
4.89; N, 13.30. LRMS (ESI+): m/z 898.8 ([M − PF6]+), 376.3 ([M −
2PF6]2+).
1
C8: isolated yield 74%. H NMR (300 MHz, CD3CN): δ 87.28,
67.31, 29.45, 19.16, 1.76, −9.82. IR (ATR): νmax (cm−1) 2972 (w),
2931 (w), 2893 (w), 2818 (w), 1608 (m), 1537 (w), 1506 (m), 1427
(w), 1367 (w), 1232 (w), 1178 (w), 1155 (w), 1045 (w), 989 (w), 821
(s), 810 (s), 707 (w). Anal. Calcd for C46H60CoF12N12P2: C, 48.90; H,
5.35; N, 14.88. Found: C, 46.76; H, 5.15; N, 14.29. LRMS (ESI+): m/z
984.9 ([M − PF6]+), 420.1 ([M − 2PF6]2+).
Synthesis of C5. To a solution of anhydrous CoCl2 (1.0 equiv,
Sigma-Aldrich) in degassed methanol (1.0 mL/7.5 μmol CoCl2) was
added a powder of the terpyridyl ligand (2.0 equiv). The solution was
refluxed for 3 h under argon and cooled to room temperature. A
concentrated solution of ammonium hexafluorophosphate (8.0 equiv,
Sigma-Aldrich) in methanol was added, and the solvent was removed
with a rotary evaporator. The product was thoroughly washed with 20
mL portions of water (Milli-Q), toluene, and Et2O. The obtained
powder was dissolved in a minimum amount of acetone, and toluene
was added dropwise until the complex precipitated. The precipitate
was collected and dried under vacuum to provide C5.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.7b00595.
1
C1: isolated yield 93%. H NMR (300 MHz, CD3CN): δ 98.88,
56.77, 48.24, 33.97, 21.25, 8.39. IR (ATR): νmax (cm−1) 3130 (w),
3095 (w), 2972 (w), 2900 (w), 2885 (w), 1600 (w), 1577 (w), 1473
(w), 1452 (w), 1436 (w), 1323 (w), 1244 (w), 1193 (w), 1051 (w),
1029 (w), 1014 (w), 902 (w), 879 (w), 837 (s), 819 (s), 767 (s), 740
(w), 729 (w), 690 (w). Anal. Calcd for C30H22CoF12N6P2: C, 44.19;
H, 2.72; N, 10.31. Found: C, 44.33; H, 2.66; N, 10.28. LRMS (ESI+):
m/z 670.3 ([M − PF6]+), 262.6 ([M − 2PF6]2+).
Synthetic procedures of L2−L6 and C1−C8, NMR
spectra of L2−L6 and C1−C8, MS spectra of L2−L6
and C1−C8, UV−vis spectra of L1−L8 and C1−C8,
buried volumes of C1−C8, cyclic voltammetry experi-
ments of C1−C8 for Co(III/II) couple, and computa-
tional details for C1−C8 (PDF)
C2: isolated yield 82%. 1H NMR (300 MHz, CD3CN): 99.66,
83.20, 68.40, 25.24, 23.75; IR (ATR): νmax (cm−1) 2987 (w), 2972
(w), 2891 (w), 1616 (w), 1533 (w), 1506 (w), 1417 (w), 1330 (m),
1288 (w), 1265 (w), 1230 (w), 1178 (w), 1165 (w), 1138 (w), 1089
(w), 1029 (w), 900 (w), 823 (s), 734 (w), 680 (m), 665 (w); Anal.
Calcd for C41H34CoF24N8OP2 (C2•CH3COCH3): C, 39.98; H, 2.78;
N, 9.10. Found: C, 40.31; H, 2.63; N, 8.95. LRMS (ESI+), m/z: 1028.4
([M − PF6]+), 441.9 ([M − 2PF6]2+).
Crystallographic data for C1 and C3−C8 (ZIP)
AUTHOR INFORMATION
Corresponding Author
*E-mail for M.F.: marc.fontecave@college-de-france.fr.
ORCID
■
Marc Fontecave: 0000-0002-8016-4747
1
C3: isolated yield 77%; H NMR (300 MHz, CD3CN): δ 105.71,
67.45, 40.06, 15.79, 13.88, 9.61, 6.80, 2.09. IR (ATR): νmax (cm−1)
Notes
3082 (w), 2972 (w), 2895 (w), 1705 (w), 1622 (w), 1585 (w), 1573
The authors declare no competing financial interest.
I
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(8) (a) Elgrishi, N.; Chambers, M. B.; Fontecave, M. Turning it off!
Disfavouring hydrogen evolution to enhance selectivity for CO
production during homogeneous CO2 reduction by cobalt-terpyridine
complexes. Chem. Sci. 2015, 6, 2522−2531. (b) Elgrishi, N.; Griveau,
S.; Chambers, M. B.; Bedioui, F.; Fontecave, M. Versatile
functionalization of carbon electrodes with a polypyridine ligand:
metallation and electrocatalytic H+ and CO2 reduction. Chem.
Commun. 2015, 51, 2995−2998.
(9) Elgrishi, N.; Chambers, M. B.; Artero, V.; Fontecave, M.
Terpyridine complexes of first row transition metals and electro-
chemical reduction of CO2 to CO. Phys. Chem. Chem. Phys. 2014, 16,
13635−13644.
(10) Colombe, J. R.; Bernhardt, S.; Stathakis, C.; Buchwald, S. L.;
Knochel, P. Synthesis of Solid 2-Pyridylzinc Reagents and Their
Application in Negishi Reactions. Org. Lett. 2013, 15, 5754−5757.
(11) Morgan, G. T.; Burstall, F. H. 3. Dehydrogenation of Pyridine of
Anhydrous Ferric Chloride. J. Chem. Soc. 1932, 20−30.
ACKNOWLEDGMENTS
■
We acknowledge support from Foundation de l’Orangerie for
individual Philanthropy and its donors. This work was
supported by the French National Research Agency (ANR,
Carbiored ANR-12-BS07-0024-03). This work has benefited
from the facilities and expertise of the Small Molecule Mass
Spectrometry platform of ICSN (Centre de Recherche de Gif -
www.icsn.cnrs-gif.fr). P.H. and H.-U.R. acknowledge the
Deutsche Forschungsgemeinschaft for support. The calcula-
tions were performed using the HPC resources of GENCI
(TGCC) through Grant 2017-810082.
REFERENCES
■
(1) (a) Shepard, S. G.; Fatur, S. M.; Rappe, A. K.; Damrauer, N. H.
Highly Strained Iron(II) Polypyridines: Exploiting the Quintet
Manifold To Extend the Lifetime of MLCT Excited States. J. Am.
Chem. Soc. 2016, 138, 2949−2952. (b) Brown, D. G.; Sanguantrakun,
N.; Schulze, B.; Schubert, U. S.; Berlinguette, C. P. Bis(tridentate)
Ruthenium−Terpyridine Complexes Featuring Microsecond Excited-
State Lifetimes. J. Am. Chem. Soc. 2012, 134, 12354−12357. (c) Lin,
H. W.; Wang, Y. S.; Huang, Z. Y.; Lin, Y. M.; Chen, C. W.; Yang, S. H.;
Wu, K. L.; Chi, Y.; Liu, S. H.; Chou, P. T. Origins of device
performance in dicarboxyterpyridine Ru(II) dye-sensitized solar cells.
Phys. Chem. Chem. Phys. 2012, 14, 14190−14195.
(2) (a) Yum, J. H.; Baranoff, E.; Kessler, F.; Moehl, T.; Ahmad, S.;
Bessho, T.; Marchioro, A.; Ghadiri, E.; Moser, J. E.; Yi, C. Y.;
Nazeeruddin, M. K.; Gratzel, M. A cobalt complex redox shuttle for
dye-sensitized solar cells with high open-circuit potentials. Nat.
Commun. 2012, 3, 631. (b) Sapp, S. A.; Elliott, C. M.; Contado, C.;
Caramori, S.; Bignozzi, C. A. Substituted polypyridine complexes of
cobalt(II/III) as efficient electron-transfer mediators in dye-sensitized
solar cells. J. Am. Chem. Soc. 2002, 124, 11215−11222. (c) Hamann, T.
W. The end of iodide? Cobalt complex redox shuttles in DSSCs.
Dalton. Trans. 2012, 41, 3111−3115. (d) Stergiopoulos, T.; Falaras, P.
Minimizing Energy Losses in Dye-Sensitized Solar Cells Using
Coordination Compounds as Alternative Redox Mediators Coupled
with Appropriate Organic Dyes. Adv. Energy Mater. 2012, 2, 616−627.
(3) Sevov, C. S.; Fisher, S. L.; Thompson, L. T.; Sanford, M. S.
Mechanism-Based Development of a Low-Potential, Soluble, and
Cyclable Multielectron Anolyte for Nonaqueous Redox Flow Batteries.
J. Am. Chem. Soc. 2016, 138, 15378−15384.
(4) Fermi, A.; Bergamini, G.; Roy, M.; Gingras, M.; Ceroni, P. Turn-
on Phosphorescence by Metal Coordination to a Multivalent
Terpyridine Ligand: A New Paradigm for Luminescent Sensors. J.
Am. Chem. Soc. 2014, 136, 6395−6400.
(5) Li, C.; Fan, W.; Straus, D. A.; Lei, B.; Asano, S.; Zhang, D. H.;
Han, J.; Meyyappan, M.; Zhou, C. W. Charge Storage Behavior of
Nanowire Transistors Functionalized with Bis(terpyridine)-Fe(II)
Molecules: Dependence on Molecular Structure. J. Am. Chem. Soc.
2004, 126, 7750−7751.
(6) (a) Zheng, Z. K.; Opilik, L.; Schiffmann, F.; Liu, W.; Bergamini,
G.; Ceroni, P.; Lee, L. T.; Schutz, A.; Sakamoto, J.; Zenobi, R.;
VandeVondele, J.; Schluter, A. D. Synthesis of Two-Dimensional
Analogues of Copolymers by Site-to-Site Transmetalation of Organo-
metallic Monolayer Sheets. J. Am. Chem. Soc. 2014, 136, 6103−6110.
(b) Groger, G.; Meyer-Zaika, W.; Bottcher, C.; Grohn, F.; Ruthard, C.;
Schmuck, C. Switchable Supramolecular Polymers from the Self-
Assembly of a Small Monomer with Two Orthogonal Binding
Interactions. J. Am. Chem. Soc. 2011, 133, 8961−8971. (c) Schubert,
U. S.; Eschbaumer, C. Macromolecules containing bipyridine and
terpyridine metal complexes: Towards metallosupramolecular poly-
mers. Angew. Chem., Int. Ed. 2002, 41, 2892−2926.
(12) (a) Rosevear, P. E.; Sasse, W. H. F. Synthesis of some 4,4′,4″-
trialkyl-2,2′6′,2″-terpyridyls. J. Heterocycl. Chem. 1971, 8, 483−485.
(b) Ben Hadda, T.; Le Bozec, H. Synthesis and characterization of
ruthenium(II) complexes containing the new tridentate ligand 4,4′,4″-
tri-tert-butyl-terpyridine. Inorg. Chim. Acta 1993, 204, 103−107.
(c) Robo, M. T.; Prinsell, M. R.; Weix, D. J. 4,4 ′,4 ″-Trimethy1−
2,2 ′:6 ′,2 ″-terpyridine by Oxidative Coupling of 4-Picoline. J. Org.
Chem. 2014, 79, 10624−10628.
(13) Sasaki, I.; Daran, J. C.; Balavoine, G. G. A. An Effective Route to
Polysubstituted Symmetric Terpyridines. Synthesis 1999, 1999, 815−
820.
(14) (a) Louerat, F.; Gros, P. C. Functional polypyridine ligands
from copper-mediated room temperature coupling of 4-chloro-2-
trimethylsilylpyridine. Tetrahedron Lett. 2010, 51, 3558−3560.
(b) Harzmann, G. D.; Neuburger, M.; Mayor, M. 4,4-Disubstituted
Terpyridines and Their Homoleptic FeII Complexes. Eur. J. Inorg.
Chem. 2013, 2013, 3334−3347. (c) Cardenas, D. J.; Sauvage, J. P.
Improved Synthesis of 2,6-Oligopyridines by Stille Cross-coupling
Reaction. Synlett 1996, 1996, 916−918. (d) Schubert, U. S.;
Eschbaumer, C. New Synthetic Strategy toward Pyridine-Based
Ligands for Supramolecular Chemistry Utilizing 2,6-Bis(trimethyltin)-
pyridine as the Central Building Block. Org. Lett. 1999, 1, 1027−1029.
(15) (a) Hommes, P.; Fischer, C.; Lindner, C.; Zipse, H.; Reissig, H.
U. Unprecedented Strong Lewis Bases−Synthesis and Methyl Cation
Affinities of Dimethylamino-Substituted Terpyridines. Angew. Chem.,
Int. Ed. 2014, 53, 7647−7651. (b) Hommes, P.; Reissig, H.-U.
Synthesis and Photophysical Properties of Substituted 2,2′:6′,2″-
Terpyridine Derivatives and Analogous Compounds with a Central
Thiophene Ring. Asian J. Org. Chem. 2016, 5, 1033−1040. (c) De
Rache, A.; Gueddouda, N. M.; Bourdoncle, A.; Hommes, P.; Reissig,
H. U.; Mergny, J. L. A Flexible Terpyridine Derivative Interacts
Specifically with G-Quadruplexes. Chem. - Eur. J. 2016, 22, 12651−
12654.
(16) Arana, C.; Yan, S.; Keshavarzk, M.; Potts, K. T.; Abruna, H. D.
Electrocatalytic Reduction of Carbon Dioxide with Iron, Cobalt, and
Nickel Complexes of Terdentate Ligands. Inorg. Chem. 1992, 31,
3680−3682.
(17) (a) Poater, A.; Ragone, F.; Giudice, S.; Costabile, C.; Dorta, R.;
Nolan, S. P.; Cavallo, L. Thermodynamics of N-Heterocyclic Carbene
Dimerization: The Balance of Sterics and Electronics. Organometallics
2008, 27, 2679−2681. (b) Poater, A.; Cosenza, B.; Correa, A.;
Giudice, S.; Ragone, F.; Scarano, V.; Cavallo, L. SambVca: A Web
Application for the Calculation of the Buried Volume of N-
Heterocyclic Carbene Ligands. Eur. J. Inorg. Chem. 2009, 2009,
1759−1766.
(18) Chambers, J.; Eaves, B.; Parker, D.; Claxton, R.; Ray, P. S.;
Slattery, S. J. Inductive influence of 4′-terpyridyl substituents on redox
and spin state properties of iron(II) and cobalt(II) bis-terpyridyl
complexes. Inorg. Chim. Acta 2006, 359, 2400−2406.
(19) Henderson, I. M.; Hayward, R. C. Substituent effects on the
stabilities of polymeric and small molecule bis-terpyridine complexes.
Polym. Chem. 2012, 3, 1221−1230.
(7) Tan, J. Y.; Li, R.; Li, D. D.; Zhang, Q.; Li, S. L.; Zhou, H. P.;
Yang, J. X.; Wu, J. Y.; Tian, Y. P. Thiophene-based terpyridine and its
zinc halide complexes: third-order nonlinear optical properties in the
near-infrared region. Dalton. Trans. 2015, 44, 1473−1482.
J
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(20) England, J.; Bill, E.; Weyhermuller, T.; Neese, F.; Atanasov, M.;
Wieghardt, K. Molecular and Electronic Structures of Homoleptic Six-
Coordinate Cobalt(I) Complexes of 2,2′:6′,2″-Terpyridine, 2,2′-
Bipyridine, and 1,10-Phenanthroline. An Experimental and Computa-
tional Study. Inorg. Chem. 2015, 54, 12002−12018.
(21) Vargas, A.; Krivokapic, I.; Hauser, A.; Daku, L. M. L. Towards
accurate estimates of the spin-state energetics of spin-crossover
complexes within density functional theory: a comparative case study
of cobalt(II) complexes. Phys. Chem. Chem. Phys. 2013, 15, 3752−
3763.
(22) (a) Kilner, C. A.; Halcrow, M. A. An unusual discontinuity in
the thermal spin transition in [Co(terpy)2][BF4]2. Dalton. Trans. 2010,
39, 9008−9012. (b) Enachescu, C.; Krivokapic, I.; Zerara, M.; Real, J.
A.; Amstutz, N.; Hauser, A. Optical investigation of spin-crossover in
cobalt(II) bis-terpy complexes. Inorg. Chim. Acta 2007, 360, 3945−
3950. (c) Kremer, S.; Henke, W.; Reinen, D. High-Spin−Low-Spin
Equilibria of Cobalt(2+) in the Terpyridine Complexes Co-
(terpy)2X2.nH2O. Inorg. Chem. 1982, 21, 3013−3022.
(23) APEX 2, SAINT and SADABS; Bruker AXS Inc., Madison, WI,
USA, 2009.
(24) Sheldrick, G. M. SHELXTL 2014/7; University of Gottingen,
̈
Gottingen, Germany, 2014.
̈
(25) Zhao, Y.; Truhlar, D. G. A new local density functional for main-
group thermochemistry, transition metal bonding, thermochemical
kinetics, and noncovalent interactions. J. Chem. Phys. 2006, 125,
194101−194118.
(26) (a) Mclean, A. D.; Chandler, G. S. Contracted Gaussian basis
sets for molecular calculations. I. Second row atoms, Z = 11−18. J.
Chem. Phys. 1980, 72, 5639−5648. (b) Krishnan, R.; Binkley, J. S.;
Seeger, R.; Pople, J. A. Self-consistent molecular orbital methods. XX.
A basis set for correlated wave functions. J. Chem. Phys. 1980, 72, 650−
654.
(27) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers, E. N.;
Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.;
Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.;
Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.;
Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.;
Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.;
Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A.
̈
D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J.
Gaussian 09; Gaussian, Inc., Wallingford, CT, USA, 2009.
(28) Dolg, M.; Wedig, U.; Stoll, H.; Preuss, H. Energy-adjusted ab
initio pseudopotentials for the first row transition elements. J. Chem.
Phys. 1987, 86, 866−872.
K
DOI: 10.1021/acs.inorgchem.7b00595
Inorg. Chem. XXXX, XXX, XXX−XXX