Synthesis and reactivity of tripodal complexes containing pendant bases

Article abstract of DOI:10.1021/ic5013389

The synthesis of a new tripodal ligand family that contains tertiary amine groups in the second-coordination sphere is reported. The ligands are tris(amido)amine derivatives, with the pendant amines attached via a peptide coupling strategy. They were designed to function as new molecular catalysts for the oxygen reduction reaction (ORR), in which the pendant acid/base group could improve the catalyst performance. Two members of the ligand family were each metalated with cobalt(II) and zinc(II) to afford trigonal-monopyramidal complexes. The reaction of the cobalt complexes [Co(L)]^- with dioxygen reversibly generates a small amount of a cobalt(III) superoxo species, which was characterized by electron paramagnetic resonance (EPR) spectroscopy. Protonation of the zinc complex Zn[N{CH₂CH₂NC(O)CH ₂N(CH₂Ph)₂}₃)]^- ([Zn(TN^Bn)]^-) with 1 equiv of acid occurs at a primary-coordination-sphere amide moiety rather than at a pendant basic site. The addition of excess acid to any of the complexes [M(L)]^- results in complete proteolysis and formation of the ligands H₃L. These undesired reactions limit the use of these complexes as catalysts for the ORR. An alternative ligand with two pyridyl arms was also prepared but could not be metalated. These studies highlight the importance of the stability of the primary-coordination sphere of ORR electrocatalysts to both oxidative and acidic conditions.

Full text of DOI:10.1021/ic5013389

Article
pubs.acs.org/IC
Synthesis and Reactivity of Tripodal Complexes Containing Pendant
Bases
Johanna M. Blacquiere,*,†,‡ Michael L. Pegis,† Simone Raugei,§ Werner Kaminsky,† Amelie Forget,†
́
Sarah A. Cook,⊥ Taketo Taguchi,⊥ and James M. Mayer*,†
†Department of Chemistry, University of Washington, Box 351700, Seattle, Washington 98195-1700, United States
‡Department of Chemistry, University of Western Ontario, London, Ontario N6A 5B7, Canada
§Physical Sciences Division, Pacific Northwest National Laboratory, P.O. Box 999, K2-57, Richland, Washington 99352, United States
⊥Department of Chemistry, University of CaliforniaIrvine, 1102 Natural Sciences II, Irvine, California 92697, United States
S
* Supporting Information
ABSTRACT: The synthesis of a new tripodal ligand family that
contains tertiary amine groups in the second-coordination sphere
is reported. The ligands are tris(amido)amine derivatives, with
the pendant amines attached via a peptide coupling strategy.
They were designed to function as new molecular catalysts for
the oxygen reduction reaction (ORR), in which the pendant
acid/base group could improve the catalyst performance. Two
members of the ligand family were each metalated with
cobalt(II) and zinc(II) to afford trigonal-monopyramidal complexes. The reaction of the cobalt complexes [Co(L)]− with
dioxygen reversibly generates a small amount of a cobalt(III) superoxo species, which was characterized by electron paramagnetic
resonance (EPR) spectroscopy. Protonation of the zinc complex Zn[N{CH2CH2NC(O)CH2N(CH2Ph)2}3)]− ([Zn(TNBn)]−)
with 1 equiv of acid occurs at a primary-coordination-sphere amide moiety rather than at a pendant basic site. The addition of
excess acid to any of the complexes [M(L)]− results in complete proteolysis and formation of the ligands H3L. These undesired
reactions limit the use of these complexes as catalysts for the ORR. An alternative ligand with two pyridyl arms was also prepared
but could not be metalated. These studies highlight the importance of the stability of the primary-coordination sphere of ORR
electrocatalysts to both oxidative and acidic conditions.
and co-workers have extensively developed phosphine ligands
with pendant, noncoordinating amine groups that dramatically
increase the rates of electrocatalytic proton reduction and
dihydrogen oxidation relative to analogues without pendant
proton shuttles.4−7 Related ligands have recently been shown to
be valuable in nitrogen reduction chemistry.8 Unfortunately,
phosphine−amine ligands do not appear to be compatible with
the oxidizing nature of the ORR, as shown by Yang et al. with
nickel complexes9 and later by Tronic et al. with ruthenium
compounds.10,11 Nocera and co-workers earlier developed a
family of “hangman” porphyrins and corroles that poise a single
carboxylic acid group above the metal center.12,13 We later
extended this approach to simpler iron porphyrin systems with
multiple pendant carboxylic acid or pyridinium groups.14,15 In
both cases, the addition of a pendant acid/base group improved
the selectivity of the ORR toward 4e− versus 2e− reduction.
The porphyrin and corrole systems have some disadvantages,
however, including synthetic accessibility, overpotential for the
ORR, and often limited stability under oxidizing conditions.
These examples illustrate that there is a need for new ligand
types with pendant acid/base functionalities because they show
INTRODUCTION
■
Efficient catalysis of proton−electron-transfer reactions is of
increasing importance, especially for the interconversion of
chemical and electrical energies. Key examples include the
hydrogenase reaction, 2e− + 2H+ ⇄ H2, and the oxygen
reduction reaction (ORR), O2 + 4e− + 4H+ → 2H2O. The
ORR is the cathode reaction in most fuel cells.1,2 Platinum is
the current state-of-the-art catalyst for the ORR, but its
limitations in expense and performance are stimulating much
work in this area. While the eventual technological solutions
will likely be heterogeneous electrocatalysts, the study of
molecular catalysts holds promise to provide new insight into
these complex reactions.3 An ORR catalyst, for instance, has
nine substrate units per turnover (O2 + 4e− + 4H+), so catalytic
cycles have many intermediates, whose identities can be studied
using a molecular catalyst. Identification of the intermediates
and construction of the structure−activity/selectivity relation-
ships of molecular ORR catalysts will inform design criteria for
heterogeneous systems.
Studies of molecular hydrogenase catalysts have shown great
value in incorporating “proton relays” in the second-
coordination sphere of the metal. These facilitate proton
transfer between the active site and the bulk solution and can
have other roles. DuBois, Rakowski-DuBois, Bullock, Helm,
Received: June 9, 2014
Published: August 8, 2014
© 2014 American Chemical Society
9242
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
promise for improving both the rate and selectivity of multi-
H+/e−-transfer reactions. Such ligands should support com-
plexes of abundant and inexpensive first-row transition metals
that are coordinatively unsaturated in order to allow substrate
binding. The ligands and complexes should be stable to
oxidative and acidic conditions. The synthetic route should
facilitate systematic ligand variation for both catalyst
optimization and mechanistic studies. Herein we report a
new tripodal ligand class and several metal complexes of these
ligands that satisfy many but not all of these requirements.
While our focus has been on new electrocatalysts for the ORR,
the design principles and challenges uncovered in this study
should be relevant to the development of proton-relay catalysts
for other processes as well.
Scheme 1. Synthetic Procedure for the Synthesis of Tertiary
a
Amine-Containing Pro-ligand H3(TNBn)
a
Conditions: (a) NaHCO3, THF/DMSO, 60 °C, 19 h (86% yield);
Molecular catalysts for the ORR have focused on
porphyrins,12,16−20 corroles,13,21 or similar macrocycles.22,23
These have taken inspiration from biological oxygenase
enzymes such as cytochromes P450 and cytochrome c
oxidase.24,25 One recent study used tris(2-pyridylmethyl)amine
(TPA) ligands with pendant primary amine or amide groups.26
While several of the synthetic systems are successful at
electrochemical or chemical dioxygen reduction, no one system
demonstrates high turnover frequencies, high selectivity, low
overpotentials, and long-term stability. Tripodal ligands27−30
have gained attention in both electrochemical26,31 and
chemical32,33 studies of dioxygen reduction and dioxygen
binding. A number of tripodal systems have incorporated
pendant hydrogen-bonding and/or acid/base functional-
ities.29,34−43 In particular, such ligand systems have been
shown to stabilize reactive intermediates that could be derived
from dioxygen, such as metal−oxo, hydroxo, and hydroperoxo
complexes.33,40,44−56 An open site for dioxygen binding is
available with complexes that have sufficiently bulky ligands
that stabilize the coordinatively unsaturated trigonal-monopyr-
amidal geometry. Herein we report the design and synthesis of
a tris(2-aminoethyl)amine (tren)-based ligand manifold and
coordination to both cobalt and zinc ions (Figure 1). The
(b) (1) LiOH, MeOH/H2O, 60 °C, 18 h, (2) HCl (90% yield); (c)
(1) NEt3, NHS, EDAC, CH2Cl2, (2) tren, 24 °C, 6 days (84% yield).
methyl-protected glycine with benzyl bromide affords a tertiary
amine moiety in high yields (ca. 86%; Scheme 1a).
Deprotection59 of the methyl ester is followed by amide
bond formation, which is facilitated by activation with N-
hydroxysuccinimide (NHS) and 1-ethyl-3-[3-(dimethylamino)-
propyl]carbodiimide hydrochloride (EDAC) as the coupling
agent (Scheme 1b,c). This route affords the protonated ligand
(or pro-ligand) H3(TNBn) in an overall yield of 65%. The pro-
1
ligand was characterized by H and 13C{1H} NMR and IR
spectroscopies and by high-resolution mass spectrometry (HR-
MS). By 1H NMR spectroscopy, as expected, only four unique
signals are observed for the distinct methylene moieties. The
benzyl methylene signals (δ = 3.59 ppm, CDCl3) integrate in a
2:1 ratio to each of the tren backbone methylene resonances.
The symmetry and integration data both confirm that all three
tren arms are functionalized. The amide N−H resonance is
obscured by the aromatic signals of the benzyl groups, but its
shift of δ = 7.21 ppm (CDCl3) is indicated by correlation to the
adjacent methylene group (δ = 3.21 ppm, CDCl3) in the
1H−1H COSY NMR spectrum. The IR stretching frequencies
at 3370 and 1669 cm−1 are diagnostic for the respective NH
and CO moieties of the amide functionality.60
A “control ligand” without the pendant tertiary amine
functionality was readily accessible by reacting tren with
propionyl chloride. The resulting pro-ligand H3(TEt) was
isolated by flash chromatography in moderate yield (Scheme 2)
1
and characterized by H and 13C NMR and IR spectroscopies
and HR-MS.
Scheme 2. Synthesis of a tren-Based “Control” Ligand,
H3(TEt)
a
Figure 1. Generic structure of the tren-based metal complexes studied
herein [R = Et, CH2N(CH2Ph)2].
modular ligand framework was designed to facilitate systematic
tuning of the catalyst structure to achieve optimal properties for
proton shuttling and selectivity (i.e., through modification of
the basicity and bulk of the pendant group). Reactivity with
dioxygen is described, but these complexes demetallate under
acidic conditions. We discuss the implications of these results to
new ligand designs for catalysis of proton/electron-transfer
reactions and describe one second-generation ligand.
a
Conditions: (a) NEt3, CH2Cl2, 0−24 °C, 21 h (38% yield).
A tren-based ligand with an aryl-substituted pendant amine
was synthesized in a similar manner to H3(TNBn) (Scheme 3).
The tertiary amine compound, 2-[N-methyl-N-
(phenylamino)]benzoic acid, was easily accessed in two steps
following modified literature procedures.61 To achieve high
yields in the tren coupling step, it proved essential to isolate
and extensively dry the NHS-activated carboxylate intermedi-
RESULTS
■
Synthesis of the Ligands. New ligands have been
prepared by the addition of amine-containing groups to tren
via a peptide coupling strategy (Scheme 1).57 Alkylation58 of
9243
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
Scheme 3. Synthesis of an Aryl-Substituted Tertiary Amine-
Scheme 5. Synthesis of Pendant Amine-Containing Complex
a
a
Containing Pro-ligand, H3(TNPh)
NEt4[Co(TNBn)]
a
Conditions: (a) (1) KH, DMF, (2) CoCl2 (43% yield); (b) NEt4Cl,
CH3CN (39% yield).
except DMF (vide infra) but is converted to a more soluble
derivative, NEt4[Co(TNBn)], upon stirring with NEt4Cl in
MeCN (Scheme 5b). The complex NEt4[Co(TEt)] (Chart 1),
a
Conditions: (a) NHS, EDAC, CH2Cl2, 24 °C, 6 h (84% yield); (b)
tren, CH2Cl2, 40 °C, 3 h (90% yield).
ate. The target aryl-substituted pro-ligand H3(TNPh) was
synthesized in four steps with an overall yield of 48%. Amide
bond formation at all three ligand arms was confirmed by
Chart 1. Tripodal Cobalt and Zinc Complexes Explored in
This Report
1
integration of the H NMR spectrum.
One additional ligand with pyridyl donors in place of two
amide donors was prepared in an attempt to avoid the protic
instability of the amide-based complexes that will be discussed
below. H(py2NBn) was synthesized using a (2-aminoethyl)bis-
(2-pyridylmethyl)amine core rather than tren (Scheme 4). The
Scheme 4. Synthesis of the Tripodal Ligand with Two
a
Pyridyl Donors, H(py2NBn)
without a pendant amine functionality, was obtained in an
analogous manner. Formation of these salts, designated as
NEt4[Co(L)], was confirmed by structural determination via X-
1
ray crystallography and a combination of H NMR, UV−vis,
and electron paramagnetic resonance (EPR) spectroscopies and
mass spectrometry (MS).63 Unfortunately, we were unable to
obtain cobalt complexes of the ligands (py2NBn)− and
(TNPh)3−. Synthetic routes to the latter were not pursued in
detail because it was judged that the lower basicity of the
second-coordination-sphere arylamines would exacerbate the
problems identified below.
Diamagnetic derivatives K[Zn(L)] were synthesized in a
similar manner for L = (TNBn)3− and (TEt)3− (Chart 1). H
1
NMR analyses of K[Zn(TNBn)] and K[Zn(TEt)] revealed
small but significant shifts in the ligand resonances relative to
the pro-ligands, most notably in the methylene unit adjacent to
the amide N−H, which shifts 0.07 and 0.12 ppm downfield
from the pro-ligand in the respective complexes [δ, DMF-d7:
3.36/3.29 for K[Zn(TNBn)]/H3(TNBn); 3.32/3.20 for K[Zn-
a
Conditions: (a) hydroxybenzotriazole, DCC, CH2Cl2, NEt3, 24 °C,
72 h (34% yield).
tripodal starting material was synthesized according to literature
procedures62 and was coupled to N,N-dibenzylglycine hydro-
chloride using a peptide coupling strategy. Formation of the
1
1
target compound was confirmed by H and 13C NMR and IR
(TEt)]/H3(TEt)]. H−1H COSY NMR analysis showed no
spectroscopies and HR-MS.
correlation from this signal to an N−H resonance, confirming
that the ligand is in a trianionic form.
Preparation and Physical Properties of the Tripodal
Cobalt and Zinc Complexes. Treatment of the pro-ligand
H3(TNBn) with excess KH in N,N-dimethylformamide (DMF)
rapidly produced bubbles, which is indicative of dihydrogen
evolution and K3L salt formation (Scheme 5). Deprotonation
of the amide was supported by the absence of a correlation
from the adjacent methylene group in the 1H−1H COSY NMR
spectra of NMR-scale reactions. In preparative-scale reactions,
the deprotonated ligand was subsequently treated with
anhydrous CoCl2 to afford the desired product K[Co(TNBn)].
The KCl salt byproduct was removed via filtration, and the
product was isolated in moderate yield (43%) by precipitation
with Et2O. K[Co(TNBn)] has low solubility in all solvents
The pro-ligand H(py2NBn) was similarly treated with KH
1
and ZnCl2 in an attempt to synthesize [Zn(py2NBn)]Cl. H
NMR analysis of crude reaction mixtures confirmed complete
reaction of the pro-ligand, but the complexity of the spectrum
indicated that a mixture of products was generated. Attempts to
selectively crystallize the desired complex were not achieved
despite several attempts with different counterions (e.g., [B(3,5-
(CF3)2C6H3)4]−, NO3 , BF4 , OTf−, and PF6 ). Similar
isolation challenges were encountered upon using CoCl2 or
FeCl2 as the metal precursor. Alsfasser et al. have acknowledged
that for related dipyridylamido tripodal ligands the observed
coordination environment is highly dependent on the ancillary
−
−
−
9244
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
Table 1. Summary of the Experimental Data for the Tripodal Complexes
[Co(TEt)]−
[Co(TNBn)]−
[Zn(TEt)]−
[Zn(TNBn)]−
λmax, nm (ε, M−1 cm−1
)
404 (30), 584 (100), 609 (93)
412 (31), 588 (104), 614 (103)
a
[M]− obsd (expected), m/z
370.2 (370.1)
4.36
913.5 (913.4)
4.21
375.3 (375.1)
918.5 (918.4)
b
μeff, BM
c
d
c
d
g values
Epa, mV vs Fc0/+
3.78 (4.28, 2.00)
3.83 (4.30, 2.00)
109
38
a
b
c
d
MALDI MS data. Evans’ method performed at 298 K. EPR data collected at 77 K. EPR data collected at 10 K.
Figure 2. Thermal ellipsoid diagrams of (a) NEt4[Co(TNBn)], (b) NEt4[Zn(TNBn)], (c) K3[Co(TEt)]3·2DMF, in which one DMF molecule has
been removed for clarity, and (d) K(18-c-6)[Co(TEt)]. The NEt4 cation is not shown in parts a and b, and hydrogen atoms are not shown. Non-
hydrogen atoms are represented by Gaussian ellipsoids at the 30% probability level.
ligands used (i.e., chlorine in our case).64 Possibly, the use of a
metal precursor that contains weakly or noncoordinating anions
would lead to a cleaner product distribution.
coordinate cobalt(II) complexes with trigonal-bipyramidal
coordination geometries, in which a solvent molecule acts as
an additional axial ligand, are generally pink in color.27−29,57
[Co(TNBn)]− and [Co(TEt)]− retain the blue color in
solution, suggesting that the trigonal-monopyramidal geometry
persists even when the complexes are dissolved in coordinating
solvents such as acetonitrile. This solution behavior is in
contrast to the structurally similar amidate-ligated complex
reported by Jones and MacBeth, which forms a solvent adduct
in MeCN.27 Their ligand, (N(o-PhNHC(O)iPr)3), is analogous
to (TEt)3− except that the amide substituent is an isopropyl
group rather than ethyl and the ligand backbone is constructed
from tris(2-aminoethyl)phenyl(amine) rather than tren. The
difference in solution behavior likely results from electronic
effects rather than steric effects because the MacBeth ligand has
the less σ-donating aryl backbone and the more sterically
demanding isopropyl substituents.
1H NMR spectroscopy proved to be a reliable tool to confirm
formation of the paramagnetic cobalt complexes. The signals
are all paramagnetically shifted, with a diagnostic peak at ca. 80
ppm for both [Co(TNBn)]− and [Co(TEt)]−. For [Co-
(TNBn)]−, the broad signals at 6.85, 5.74, and 3.69 ppm (DMF-
d7) show that the benzyl signals are significantly less
paramagnetically shifted than the tren-backbone methylene
units. Evans’ method analysis65 of [Co(TNBn)]− and [Co-
(TEt)]− gave μeff = 4.21 and 4.36, respectively, consistent with
the assignment of both cobalt(II) species as high-spin (S = 3/2).
The blue cobalt(II) complexes display three prominent
optical features between 400 and 650 nm (Table 1). The
energies and intensities of these bands are similar to the
features previously observed for related cobalt(II) complexes
with trigonal-monopyramidal coordination geome-
The tripodal complexes [M(TEt)]− and [M(TNBn)]− were
also analyzed by matrix-assisted laser desorption ionization MS
(MALDI MS) in negative-ion mode with a polyaromatic
hydrocarbon matrix.67 For each of the cobalt and zinc
complexes, the molecular anion was observed with an excellent
tries27,29,50,57,66 and are assigned as A2 → A2, A2 → E, and
4
4
4
4
66
4A2 → E transitions. These results strongly indicate that in
solution the blue cobalt complexes are four-coordinate with an
open site cis to the three amidate moieties. By comparison, five-
4
9245
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
a
Table 2. Selected Bond Lengths and Angles for the Tripodal Cobalt and Zinc Complexes
K3[Co(TEt)]3·2DMF
K(18-c-6) [Co(TEt)]
NEt4[Co(TNBn)]
NEt4[Zn(TNBn)]
Bond Length (Å)
2.0877(18)
M−N1
M−N2
M−N3
M−N4
2.090(7)
1.955(7)
1.979(6)
1.928(7)
2.0878(12)
1.9573(13)
1.9588(12)
1.9594(13)
2.0915(11)
1.9672(12)
1.9718(12)
1.9665(11)
1.9579(13)
b
1.9579(13)
1.9593(18)
Angle (deg)
85.68(4)
85.68(4)
84.73(7)
118.46(8)
119.80(4)
119.80(4)
0.93
N1−M−N2
N1−M−N3
N1−M−N4
N2−M−N3
N3−M−N4
N2−M−N4
τ′
85.7(3)
85.5(3)
85.4(3)
117.6(3)
118.0(3)
122.5(3)
0.89
85.88(5)
85.22(5)
85.81(5)
119.60(6)
120.93(6)
117.74(6)
0.92
87.17(5)
86.94(5)
86.36(4)
118.21(5)
121.19(5)
119.68(5)
0.93
a
b
One of the three Co molecules in the unit cell. Generated by symmetry from Co−N2.
isotope match to the expected composition (Table 1 and
Figures S21−S24 in the Supporting Information, SI).
Structural Studies of the Tripodal Cobalt and Zinc
Complexes. Structural analyses of K[Co(TEt)], NEt4[Co-
(TNBn)], and NEt4[Zn(TNBn)] by X-ray diffraction (XRD)
methods show that each member of the series is trigonal
monopyramidal in the solid state with roughly C3v symmetry
(Figure 2 and Table 2). Kerton and Kozak have proposed τ′ as
a metric to assess the geometry and distortion in four-
coordinate metal complexes, similar to the use of τ for five-
coordinate structures.28 A trigonal-monopyramidal geometry
will give τ′ = 1, while a tetrahedron gives τ′ = 1.5. The τ′ values
of ca. 0.9 for the complexes reported herein (Table 2) are
similar to those of related trigonal-monopyramidal complexes28
and indicate minimal distortion from the ideal geometry.
In all cases, the equatorial metal−amide bonds are ca. 0.1 Å
shorter than the apical metal−amine bond. The structure of
K[Co(TEt)]− includes three cobalt molecules in the
asymmetric unit and two DMF solvent molecules. The
potassium ions bridge adjacent [Co(TEt)]− species through
interactions with the ligand carbonyl oxygen atoms to create an
extended structure (Figure 2a). This Lewis acid/base
interaction results in a distortion of the primary-coordination
sphere, where the potassium-associated amide has a slightly
longer Co−N bond by ca. 0.03 Å and, by compensation, a
second Co−N bond shortens by the same distance (Table 2). A
similar K+-bridged extended network was observed previously
in a cobalt complex with a tren-based cryptand ligand.57 These
observed solid-state interactions are likely the origin of the
limited complex solubility. In support of this view, crystal-
lization of K[Co(TEt)] in the presence of 18-crown-6 afforded
a salt that is soluble in a broader range of solvents and has a
structure with a sequestered potassium ion and a lattice that is
devoid of DMF molecules.
Figure 3. Perpendicular-mode X-band EPR spectra of (a) [Co-
(TNBn)]− and (b) [Co(TEt)]−; recorded at 10 K in a 1:1 THF/DMF
glass.
respectively (Table 1). The presence of these two features
indicates that the complexes have axial symmetry in solution
(E/D ∼ 0.0) and that they retain the C3v symmetry observed in
the solid state by XRD. The signal at g = 2.0 is split into an
eight-line pattern because of the expected hyperfine coupling to
7
the 57Co nucleus (I = /2). The magnitudes of the hyperfine
couplings in [Co(TNBn)]− (Az = 98 × 10−4 cm−1) and
[Co(TEt)]− (Az = 96 × 10−4 cm−1) are the same as those
found for other cobalt(II) trigonal-bi- or monopyramidal
complexes.50,68,70
EPR Spectral Properties of the Tripodal Cobalt
Complexes. The cobalt complexes were analyzed by X-band
EPR spectroscopy as frozen solutions at 10 and 77 K in
standard perpendicular (⊥) mode. At the higher temperature,
both [Co(TEt)]− and [Co(TNBn)]− displayed a derivative
signal at ca. g = 3.8 (Table 1) that is consistent with similar
high-spin (S = 3/2) trigonal-monopyramidal complexes.68
Broadening of the signals due to spin−lattice relaxation69 is
minimized at 10 K, and two features are observed for each
complex with g values at ca. 4.3 and 2.0 (160 and 344 mT,
respectively; Figure 3). These features are assigned to g⊥ and g∥,
Electrochemical Properties of the Tripodal Cobalt
Complexes. Cyclic voltammograms of [Co(TEt)]− and
[Co(TNBn)]− were recorded at a scan rate of 100 mV/s in
n
acetonitrile with a glassy carbon working electrode and Bu4N-
[PF6] as the supporting electrolyte (Figure 4 and Table 1).
Both complexes display irreversible oxidations (even at a higher
scan rate of 300 mV/s) similar to previously reported cobalt
9246
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
Figure 4. Cyclic voltammograms of (a) [Co(TNBn)]− and (b)
[Co(TEt)]− in CH3CN recorded at a scan rate of 100 mV/s with a
glassy carbon working electrode and Bu4N[PF6] as the electrolyte.
n
Figure 5. UV−vis spectra of 2.5 mM solutions of (a) [Co(TEt)]− and
(b) [Co(TNBn)]− upon the addition of 15 equiv of dioxygen
monitored over a 30 min period.
trigonal-monopyramidal complexes.50,57 This points to the
instability of the cobalt(III) complexes with low coordination
numbers. The more negative anodic peak potential (Epa) for
[Co(TNBn)]− relative to [Co(TEt)]− (Epa = 38 and 109 V vs
Fc+/0, respectively) could reflect the electron-donating nature of
the 3° amine moieties either inductively or by weak
coordination with the cobalt(III) product.
Reactivity of the Tripodal Complexes with Dioxygen.
The addition of excess dioxygen to solutions of [Co(TNBn)]−
or [Co(TEt)]− did not result in any visually observable color
change or in the formation of a dioxygen adduct, as judged by
1
MALDI MS. H NMR spectra in DMF-d7 of the oxygenated
solutions did not indicate the formation of a new species, but
quantification of the cobalt complexes relative to an internal
standard (1,3,5-trimethoxybenzene) revealed minor consump-
tion of [Co(TNBn)]− (17%) and [Co(TEt)]− (10%).
The dioxygen reactivity was probed optically for both
[Co(TEt)]− and [Co(TNBn)]− (Figure 5). The addition of
excess (15 equiv, 82 μmol) dioxygen to [Co(TEt)]− (2.5 mM,
5.6 μmol) resulted in increased absorption at wavelengths
below 520 nm over a period of 30 min, but very little change
was observed for the features around 600 nm (Figure 4a).
Similar behavior is observed for [Co(TNBn)]− but with a less
pronounced change in the absorbance (Figure 5b).
Figure 6. X-band EPR spectra, recorded at 77 K in a 1:1 THF/DMF
glass: (a) [Co(TEt)]− following dioxygen addition; (b) expansion of
the g = 2 region of part a; (c) [(Co(TNBn)]− following dioxygen
addition; (d) expansion of the g = 2 region of part c.
of [Co(TNBn)]−, some computational studies were performed
as described in the Discussion section.
The product of dioxygen reactivity was probed by EPR
spectroscopy by bubbling solutions of [Co(TEt)]− and
[Co(TNBn)]− with dry dioxygen for 2 min and then cooling
to a glass at 77 K (Figure 6). Minor formation of a new species
is evident with both complexes by the appearance of a new
signal at ca. g = 2 with eight-line hyperfine coupling to cobalt (I
Reactivity of the Tripodal Complexes with Acid. The
diamagnetic complex [Zn(TNBn)]− was treated with 1 equiv of
the acid [H-DMF]OTf in either DMF-d7 or CD3CN under
1
anaerobic conditions, and the reaction was analyzed by H
NMR spectroscopy (Figure 7a,b). Only minor amounts of the
starting material [Zn(TNBn)]− remain (ca. 5% relative to the
internal standard); signals that match the pro-ligand H3(TNBn)
are found in ca. 25% yield. The remaining 70% of material is
tentatively assigned as a protonated species. The site of
protonation was identified through 1H−1H COSY NMR
analysis, where a correlation between Hb and a new signal
was observed that strongly suggests that the site of protonation
is the adjacent amide moiety. Protonation of the tertiary amine
was not observed, which would have been indicated by
7
= /2): Az = 28 × 10−4 cm−1 for [Co(TEt)]− and Az = 25 ×
10−4 cm−1 for [Co(TNBn)]−. The location of the signal and
magnitude of the coupling are consistent with the formation of
a cobalt(III) superoxo adduct.71,72 The signal at g = 2
disappears upon sparging of the thawed oxygenated solution
with dinitrogen, indicating that the formation of the cobalt(III)
superoxo complexes is reversible. The superoxo signal was not
observed at 10 K because of obstruction by g∥ of the parent
species. To obtain additional insight into the dioxygen adduct
9247
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
Bn
Bn
●
▲
Figure 7. Protonation study with K[Zn(TN )] in DMF-d7: (a) mixture of K[Zn(TN )] ( ) and an internal standard ( ); (b) addition of 1 equiv
Bn
▽
of [H-DMF]OTf; (c) addition of 3 equiv of [H-DMF]OTf; (d) sample from part c spiked with 0.5 equiv of authentic H3(TN ) ( ). Internal
standard = 1,3,5-trimethoxybenzene. Signal intensities are normalized to a set height for the internal standard. DMF, from both residual proteo
■
.
solvent molecules and added [H-DMF]OTf, is identified by *, and unassigned signals are identified by
correlations from both methylene units Hc and Hd (see Figure
8). Expected correlations were observed between the
methylene units of the tren backbone (Ha and Hb; Figure 8).
reversible (Figures S25 and S26 in the SI). Optical monitoring
of [Co(TEt)]− likewise exhibited reversible protonation upon
the addition of 1 equiv of acid followed by base (Figures S25
and S26 in the SI).
The addition of a total of 3 equiv of [H-DMF]OTf to
[Zn(TNBn)]− resulted in the quantitative formation of
H3(TNBn) (by 1H NMR spectroscopy; Figure 7c). The
assignment of this species was confirmed upon spiking the
solution with 0.5 equiv of authentic pro-ligand (Figure 7d).
1
Analogous H NMR experiments with [Co(TNBn)]−, [Zn-
(TEt)]−, and [Co(TEt)]− afforded very similar results, in
which the addition of 3 equiv of acid also generates the pro-
ligand in quantitative yield (Figures S27−S32 in the SI).
DISCUSSION
■
Reported herein are new tetradentate nitrogen-donor ligands in
which a tren backbone has been coupled to a number of
moieties through an amide linkage. Most of our studies have
used the (TNBn)3− ligand in which the subunits are derived
from the amino acid glycine, with the amine moiety converted
to a tertiary center. In principle, this ligand is modular and
versatile because the use of different amino acids or substituents
on the amine would afford a range of derivatives. The versatility
in the design is demonstrated with the (TNPh)3− and
(py2NBn)− ligands. The former contains an aryl-substituted
pendant amine, while the latter retains the pendant N-benzyl
functionality of (TNBn)3− but substitutes two of the tripodal
arms with pyridyl donors. The control ligand (TEt)3−, without
basic groups in the second-coordination sphere, was designed
with ethyl groups adjacent to the amide to offer steric
protection of the open coordination site in the resulting
metal complexes. This protection mimics that of (TNBn)3−,
which was expected to form a monometallic cobalt(III)
superoxo rather than a μ-peroxodicobalt(III) species upon
oxygenation.51,75
Figure 8. 1H−1H COSY NMR spectrum of a solution of [Zn-
(TNBn)]− + 1 equiv of [H-DMF]OTf in CD3CN. The structure above
is shown for labeling purposes only.
The treatment of [Zn(TNBn)]− with weaker acids in CD3CN
with pKa values ranging from 14 to 27 (cf. pKa of [HDMF]OTf
= 6)73,74 also afforded the protonated ligand H3(TNBn). In the
presence of p-tert-butylphenol [pKa(MeCN) = 27], an
unassigned intermediate was observed at short reaction times,
which mostly converts (over ca. 1 h) to the pro-ligand.
The paramagnetic complex [Co(TNBn)]− was titrated with a
total of 1 equiv of [H-DMF]OTf, followed by the addition of
t
the strong phosphazine base BuP2(dma) (dma = dimethyl-
amine; N‴-tert-butyl-N,N,N′,N′-tetramethyl-N″-[tris-
(dimethylamino)phosphoranylidene]phosphorimidic triamide).
Additions were monitored optically, and the resulting spectra
indicate that protonation with a single equivalent of acid is
Cobalt(II) and zinc(II) complexes were prepared with the
(TNBn)3− and (TEt)3− ligands. The [Zn(L)]− complexes act as
1
redox neutral and diamagnetic derivatives that allow H NMR
9248
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
spectroscopy to be employed as a tool to track proton
movement. Cobalt was chosen for its synthetic accessibility, and
cobalt(II) species are known to reduce dioxygen through either
a 4e− or a more common 2e− pathway. This was seen as an
opportunity to rigorously test the efficacy of the pendant basic
group of [Co(TNBn)]− to alter selectivity in the ORR relative
to [Co(TEt)]−. The cobalt complexes with the (TNBn)3− and
(TEt)3− ligands both have four-coordinate, coordinatively
unsaturated, trigonal-monopyramidal structures. In [Co-
(TNBn)]−, the amine groups are distant from the cobalt(II)
center. The two complexes are similar structurally and
spectroscopically, displaying expected characteristics for cobalt-
(II) complexes with this primary-coordination sphere. Values
for the structural parameter τ′ close to 1 indicate that very little
distortion from the expected trigonal-monopyramidal coordi-
nation geometry is found in the solid state, and analysis of
frozen solutions by EPR spectroscopy points toward C3-
symmetric cobalt(II) species in an axial environment. The blue
color of both complexes and UV−vis spectra strongly support
the presence of four-coordinate complexes in solution. The data
suggest that the primary-coordination spheres of [Co(TNBn)]−
and [Co(TEt)]− are similar and that the former is a reasonable
control complex without basic groups in the second-
coordination sphere. The zinc derivatives with these ligands
appear to be structurally similar, based on the solid-state
structure of [Zn(TNBn)]− and solution-state NMR analyses,
indicating that these are reasonable diamagnetic analogues of
the cobalt complexes.
and correlation functional,78 with the Stuttgart SDD basis set
for the cobalt atom, the 6-311G** basis set for dioxygen atoms,
and 6-31G for all other atoms (see the SI for additional details).
Calculations indicate that an end-on coordination of dioxygen
is the most favored structure for the adduct, but the dioxygen
binding free energy is endoergic by about +10 kcal mol−1 for
both [Co(TNBn)]− and [Co(TEt)]−. This thermodynamic bias
against dioxygen coordination is qualitatively consistent with
the low conversion to the cobalt superoxo complex that was
observed experimentally. Quantitatively, the magnitude is
higher than anticipated from the experimental Keq values,
which is a typical problem for M−O2 binding energies obtained
from DFT calculations.79,80 The computed Co−O and O−O
bond lengths of 1.92 and 1.27 Å are within the ranges of 1.86−
1.97 and 1.24−1.30 Å typically observed for the respective
distances in reported six-coordinate cobalt(III) superoxo
species.75,81 Relative to molecular oxygen,75 the O−O bond is
only lengthened by ca. 0.06 Å, indicating that charge donation
into the dioxygen moiety is small. Characterization as a
superoxo ligand is also indicated by the computed Mulliken
charges for proximal and distal oxygen atoms of −0.13 and
−0.22 e−, respectively, and corresponding spin charges of 0.39
and 0.63 e−, respectively. All of these data are consistent with a
weak interaction of dioxygen with the metal center. Still, strong
dioxygen coordination is not a prerequisite for electrocatalytic
dioxygen reduction, as evidenced by Arnold’s studies of a
tripodal cobalt tris(2-pyridylmethyl)amine complex.31
These species are among only a few50 reported five-
coordinate cobalt superoxo adducts in the literature. Classic
studies in cobalt dioxygen reactivity were predominantly
focused on six-coordinate products, which also coordinate
oxygen reversibly, but in contrast are accessible in high
conversion.72,75 The reported octahedral complexes are
characterized as low-spin cobalt(III) species, with the unpaired
electron residing on the superoxo moiety. The five-coordinate
superoxo complexes reported here likewise have unpaired
electron density on the superoxo moiety, as judged by EPR
spectroscopy. In contrast to the octahedral systems, DFT
analysis here indicates that the S = 2 cobalt spin state is favored,
albeit only by 3 kcal mol−1 (within the accuracy of the
The cobalt derivatives react with dioxygen only to a small
extent, with low conversion of [Co(TNBn)]− and [Co(TEt)]−,
1
as evidenced by H NMR, EPR, and UV−vis spectroscopies.
Related dioxygen adducts have been reported to have intense
charge-transfer bands in the UV−vis spectrum, with molar
absorptivities of ε ∼ 104 M−1 cm−1.69,76 In the reactions of
dioxygen with [Co(TNBn)]− or [Co(TEt)]−, little depletion of
the starting material is observed, but a small amount of product
is visible because of its high ε.77 With the rough assumption
that the product ε values are approximately 104 M−1 cm−1, as
observed for octahedral adducts, the equilibrium constants for
dioxygen binding would be ca. 10−3 atm−1 for both compounds.
EPR studies indicated that sparging of dioxygen-saturated
solutions with dinitrogen results in decoordination of dioxygen
to regenerate the trigonal-monopyramidal starting material.
The experimental and computational (see below) results are all
consistent with equilibrium dioxygen binding to [Co(TNBn)]−
or [Co(TEt)]−, which lies strongly in favor of the open-site
species (Scheme 6).
2
calculation). This electronic structure requires that the dz
orbitalor σ* orbital for the Co−O2 bondis partially
occupied. The σ-bonding interaction is thus rather weak, and
this is likely the origin of the positive ΔG° for dioxygen
binding. In related systems, Borovik et al. have shown that
tripodal cobalt complexes with hydrogen-bond donor groups in
the second-coordination sphere react with dioxygen to generate
CoIIIOH products through an unknown pathway.50,82 The
difference in the reactivity to [Co(TNBn)]− could be ascribed
to the different nature of the pendant group (hydrogen-bond
donor vs acceptor), the proximity of the pendant group, the
electron density at the metal center, or steric accessibility of the
open coordination site.
We originally targeted a tripodal ligand design as a promising
modular system that would lead to improved electrocatalysts of
the ORR and perhaps other proton−electron reactions. We
succeeded in preparing tripodal complexes that contain an open
coordination site with noncoordinating amine functionalities, to
facilitate proton transfer. However, the addition of 3 equiv of
[H-DMF]OTf to each of the cobalt or zinc complexes
generated the fully protonated pro-ligands H3(TNBn) or
H3(TEt) in quantitative yield (Scheme 7). Protonation occurs
at the amide groups rather than the pendant amines, even
The low conversion precludes detailed experimental
examination of the dioxygen adducts, so we turned to
computational analysis. Density functional theory (DFT)
calculations were performed using Thrular’s M06L exchange
Scheme 6. Reversible Coordination of Dioxygen to
[Co(TNBn)]− and [Co(TEt)]−, Generating Cobalt(III)
Superoxo Products
9249
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
Scheme 7. Instability of the Tripodal Complexes [M(L)]− to
a Total of 1 and 3 equiv of Acid
analysis of the diamagnetic zinc analogues reveals that the
complexes are not stable to acidic conditions and undergo M-L
proteolysis.
This study reveals that the tripodal ligand family reported
here is not a suitable framework to identify relationships
between the primary/secondary-coordination sphere structure
and catalyst performance (e.g., rate, selectivity, overpotential).
Future ligand design should balance the requirements of
modularity, for systematic structure variation, with the basicity
of the primary-coordination sphere. The reactivity of the
coordinating groups with added acid is a serious problem in the
design of ligands with pendant, nonbinding acid/base groups.
The P2 N2 ′ ligands avoid this problem because they bind to
relatively soft metal centers with soft donor ligands, leaving the
harder amine ligands available to bind to the hard proton.
However, soft donors are not typically compatible with the
oxidative conditions of the ORR. When hard donors are
needed, there is a concern that the basicity of these donors will
be greater than the pendant base groups, leading to ligand
displacement from the metal in the presence of protons. The
“hangman” porphyrin and corrole ligands developed by Nocera,
and our pyridyl and carboxyphenyl porphyrin examples,
prevent protonation of the metal-coordinating groups and
binding of the potential proton relays through the rigidity of
the ligand framework. The strategy of kinetically inhibiting
protonation of the primary-coordination sphere through more
rigid ligand structures is worth probing further.
though the amide groups are coordinated to the metal center in
[M(L)]−. This protonation location even appears to hold for
R
R
1
the stoichiometric addition of a single proton, in which H
NMR analysis indicated that 14−44% H3L is generated (Table
S2 in the SI). In all cases, the amide moiety rather than the
pendant tertiary amine is the preferred site of protonation.
Even use of weaker acids results in protonation of the amide
moiety. This is a serious problem because the normal
conditions for dioxygen reduction require that the acid
substrate be present in large excess to the catalyst.
The monoanionic pyridyl ligand was synthesized in an effort
to decrease the basicity of the nitrogen donors in the primary-
coordination sphere of a resulting metal complex. The
attenuated basicity of pyridine relative to amide was expected
to limit complex decomposition by proteolysis with excess acid.
Unfortunately, attempts thus far to generate [M(py2NBn)]+ (M
= Zn, Co, Fe) have been unsuccessful.
It was anticipated that protonation of the pendant base(s) in
[Co(TNBn)]− would increase the affinity of the complex for
dioxygen because of the formation of intramolecular hydrogen
bonds. Changes in the optical spectrum were observed upon
treatment of an oxygenated solution of [Co(TNBn)]− with [H-
DMF]OTf at −20 °C. The resulting absorption bands are
distinct from those observed upon the addition of acid to
anaerobic solutions, and the final spectrum is identical when the
order of the addition is reversed (i.e., acid and then dioxygen).
However, warming the solution to room temperature resulted
in decomposition, which underscores the instability of this
complex to the conditions required for the ORR.
The combination of a low equilibrium constant for dioxygen
binding and poor acid stability severely limits the utility of
[Co(TNBn)]− or [Co(TEt)]− for electrocatalysis of the ORR.
Still, these ligands could be useful for catalysis. The remarkable
tripodal molybdenum complexes developed by Schrock for
catalytic dinitrogen reduction show similar sensitivity to excess
acid.83,84 In this case, catalytic turnover is achieved by the slow
addition of an insoluble acid using a syringe pump and utilizing
a chemical reductant.
EXPERIMENTAL SECTION
■
Synthesis of H3(TEt). Tris(2-aminoethyl)amine (tren; 1.392 g,
9.519 mmol) was combined with NEt3 (3.855 g, 0.03810 mol) in
CH2Cl2 (20 mL) under dinitrogen and cooled in an ice bath.
Propionyl chloride (3.524 g, 0.03809 mmol) was added dropwise. A
white solid (HNEt3Cl) formed immediately, and an additional portion
of CH2Cl2 (20 mL) was added to facilitate stirring. The reaction was
allowed to warm to room temperature and was stirred for 21 h.
Filtration removed some of HNEt3Cl. The remainder of the salt was
removed through successive crystallizations from CH2Cl2 and a final
crystallization from acetone. The solvent from the final crystallization
was removed, and the resulting oil was purified by column
chromatography [4% methanol (MeOH) in CH2Cl2] with silica
(neutralized with 1% NEt3 in CH2Cl2). The purified product slowly
solidified to a white solid over 1 day. Yield: 1.133 g (38%). RF = 0.24
in 10% MeOH in CH2Cl2. 1H NMR (CDCl3, 499 MHz): δ 6.79 (br s,
3H, NH), 3.26−3.25 (m, 6H, NCH2CH2NH), 2.53 (m, 6H,
NCH2CH2NH), 2.23 (q, J = 5.0 Hz, 6H, COCH2CH3), 1.12 (t, J =
5.0 Hz, 9H, COCH2CH3). 13C NMR (CDCl3, 126 MHz): δ 174.8 (s,
NHC(O)CH2), 54.6 (s, NCH2CH2NH), 37.7 (s, NCH2CH2NH), 29.5
(s, COCH2CH3), 10.0 (s, COCH2CH3). HR-MS. Calcd ([M + Na]+):
m/z 337.2215. Obsd: m/z 337.2209. IR (KBr pellet, cm−1): 3284 [br s,
ν(NH)], 1653 [s, ν(CO)].
CONCLUSIONS
■
Synthesis of (Dibenzylamino)acetic Acid Methyl Ester.
Glycine methyl ester hydrochloride (6.072 g, 48.40 mmol) and
NaHCO3 (24.397 g, 290.42 mmol) were combined with dimethyl
sulfoxide (DMSO; 25 mL) and tetrahydrofuran (THF; 180 mL) in a
round-bottomed flask. Benzyl bromide (25.661 g, 150.03 mmol) was
added, followed by THF (20 mL), and the reaction was stirred at 60
°C for 19 h. The reaction was filtered and washed with THF. The
filtrate was collected, and the solution was concentrated under
vacuum. Water (20 mL) was added, and the pH was increased to 9.
The mixture was extracted with CH2Cl2 (3 × 25 mL). The organic
layers were combined and dried with MgSO4, and the solvent was
removed under vacuum to afford a yellow oil. The pure product (RF =
0.24; 10% MeOH in CH2Cl2) was obtained by column chromatog-
raphy (0−2% EtOAc in hexanes) using neutralized silica (1% NEt3 in
Four entries in a new family of modular tripodal ligands have
been synthesized for applications in proton−electron reactions
such as the ORR. The pro-ligand H3(TNBn) was designed to
contain a pendant tertiary amine that could participate in
proton-transfer events, a functionality that positively influences
both rate and selectivity in related catalytic systems. An
analogue without a pendant basic functionality, H3(TEt), was
prepared as a control for reactivity studies. Cobalt(II) and
zinc(II) complexes of these ligands were prepared, and all
species assumed the desired coordinatively unsaturated
trigonal-monopyramidal coordination mode in both the
solution and solid state. However, the cobalt complexes only
1
1
bind dioxygen to a very small extent. More critically, H NMR
hexanes). Yield: 11.189 g (86%). H NMR (CDCl3, 499 MHz): δ
9250
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
7.43−7.41 (m, 4H, Ph), 7.36−7.33 (m, 4H, Ph), 7.29−7.26 (m, 2H,
Ph), 3.84 (s, 4H, NCH2Ph), 3.71 (s, 3H, OCH3), 3.33 (s, 2H,
C(O)CH2N). 13C NMR (CDCl3, 126 MHz): δ 172.0 (s, OC(O)-
CH2), 139.1 (s, Ph), 129.0 (s, Ph), 128.4 (s, Ph), 127.3 (s, Ph), 57.9 (s,
NCH2Ph), 53.5 (s, C(O)CH2N), 51.4 (s, OCH3). HR-MS. Calcd ([M
+ H]+): m/z 270.1494. Obsd: m/z 270.1501. IR (KBr pellet, cm−1):
1749 [s, ν(CO)].
115.8 (s, Ar), 52.8 (s, NCH2CH2NH), 40.8 (s, NCH3Ph), 37.4 (s,
NCH2CH2NH). HR-MS. Calcd ([M + H]+): m/z 774.4132. Obsd: m/
z 774.8. IR (KBr pellet, cm−1): 3383 [m, ν(NH)], 1653 [s, ν(CO)].
Synthesis of H(py2NBn). N,N-Dibenzylglycine hydrochloride (433
mg, 1.48 mmol) was added to a 250 mL flask under dinitrogen,
followed by CH2Cl2 (200 mL) and NEt3 (220 mg, 2.16 mmol, 1.46
equiv). The hydrochloride salt became soluble within 10 min. To the
reaction was added hydroxybenzotriazole (227 mg, 1.48 mmol, 1.0
equiv). The reaction was cooled to 0 °C in an ice bath, and
dicyclohexylcarbodiimide (DCC; 306 mg, 1.48 mmol, 1.0 equiv) was
added as a solution in CH2Cl2 (5 mL). After 15 min, (2-
aminoethyl)bis(2-pyridylmethyl)amine (300 mg, 1.23 mmol, 0.83
equiv) was added, and the reaction was allowed to warm to room
temperature. After stirring for 72 h, a white precipitate had formed on
top of the solution. The solids are byproducts from the coupling
reaction (including a DCC hydration product, dicyclohexylurea) that
are partially soluble in CH2Cl2. The solvent was removed, and MeCN
(50 mL) was added to fully precipitate the byproducts. The solution
was filtered through Celite, the filtrate was dried with Na2SO4, and the
solvent was removed under vacuum. The pure product (RF = 0.27;
2.5% MeOH in CH2Cl2) was obtained as a yellow oil by column
chromatography (2.5% MeOH in CH2Cl2) using alumina (Fluka, pH
N,N-Dibenzylglycine Hydrochloride.
A solution of
(dibenzylamino)acetic acid methyl ester (10.4413 g, 38.7656 mmol)
in MeOH (120 mL) was cooled in an ice bath. LiOH (9.289 g, 388.0
mmol) dissolved in water (120 mL) was slowly added to the MeOH
solution. The reaction was allowed to warm to room temperature and
was then heated to 60 °C for 18 h. The reaction was cooled, and HCl
(64 mL, 6 M) was added to bring the pH to 3. At this point, the
product hydrochloride salt precipitated as a white powder. The
powder was filtered, washed with C6H6 (2 × 15 mL) and Et2O (3 × 15
mL), and dried. A second crop of product precipitated from the
filtrate. This solid was filtered, washed with water (2 × 10 mL), C6H6
(2 × 10 mL), and Et2O (2 × 15 mL), and dried. The combined yield
was 10.2033 g (90%). The 1H NMR spectrum of the product matched
literature values.85
Synthesis of H3(TNBn). N,N-Dibenzylglycine hydrochloride (5.021
g, 17.21 mmol) was added to a 250 mL flask under dinitrogen,
followed by CH2Cl2 (15 mL) and NEt3 (1.915 g, 18.92 mmol, 1.1
equiv). The hydrochloride salt became mostly soluble within 10 min.
To the reaction was added N-hydroxysuccinimide (NHS; 2.969 g,
25.80 mmol, 1.5 equiv), 1-ethyl-3-[3-(dimethylamino)propyl]-
carbodiimide hydrochloride (EDAC; 4.946 g, 25.80 mmol, 1.5
equiv), and CH2Cl2 (20 mL), and the reaction was stirred for 24 h.
Tren (856 mg, 5.73 mmol, 0.33 equiv) was added to the clear yellow
reaction, and the solution was stirred for 6 days. The reaction was
extracted with pH 9 NaOH (2 × 10 mL) and water (10 mL). The
aqueous layer was neutralized and extracted with CH2Cl2 (3 × 10
mL). The combined organic layers were washed with brine (15 mL)
and dried with MgSO4, and the solvent was removed under vacuum.
The pure product (RF = 0.49; 10% MeOH in CH2Cl2) was obtained
by column chromatography (4% MeOH in CH2Cl21) using neutralized
silica (1% NEt3 in CH2Cl2). Yield: 4.12 g (84%). H NMR (CDCl3,
499 MHz): δ 7.32−7.20 (m, 33H, Ph and NHCO), 3.59 (s, 12H,
NCH2Ph), 3.21 (m, 6H, NCH2CH2NH), 3.10 (s, 6H, COCH2N),
2.52 (t, J = 5.0 Hz, 6H, NCH2CH2NH). 13C NMR (CDCl3, 126
MHz): δ 171.3 (s, NHC(O)CH2), 138.0 (s, Ph), 129.0 (s, Ph), 128.7
(s, Ph), 127.7 (s, Ph), 59.3 (s, NCH2Ph), 57.5 (s, NCH2CH2NH), 53.3
(s, C(O)CH2N), 36.9 (s, NCH2CH2NH). HR-MS. Calcd ([M + H]+):
m/z 858.5071. Obsd: m/z 858.5066. IR (KBr pellet, cm−1): 3370 [m,
ν(NH)], 1669 [s, ν(CO)].
1
9.5) as the stationary phase. Yield: 200 mg (34%). H NMR (CDCl3,
500 MHz): δ 8.56 (d, J = 4 Hz, 2H, py-H), 7.60 (br s, 1H,
NHCOCH3), 7.54 (m, 2H, py-H), 7.43 (m, 2H, py-H), 7.38 (m, 4H,
Ph-H), 7.28 (m, 6H, Ph-H), 7.16 (m, 2H, py-H), 3.87 (s, 4H,
pyCH2N), 3.66 (s, 4H, PhCH2N), 3.40 (m, 2H, NCH2CH2NH), 3.15
(s, 2H, COCH2NBn2), 2.77 (t, J = 5.3 Hz, 2H, NCH2CH2NH). 13C
NMR (CDCl3, 500 MHz): δ 170.0 (s, NHC(O)CH2), 159.0 (s, Ar),
149.0 (s, Ar), 138.0 (s, Ar), 136.0 (s, Ar), 129.0 (s, Ar), 128.0 (s, Ar)
127.0 (s, Ar) 123.0 (s, Ar), 122.0 (s, Ar), 60.0 (s, pyCH2N), 59.0 (s,
NCH2Ph), 57.0 (s, NCH2CH2NH), 53.0 (s, C(O)CH2N), 36.0 (s,
NCH2CH2NH). HR-MS. Calcd ([M + H]+): m/z 480.2673. Obsd: m/
z 480.27. IR (solution, CCl4, cm−1): 1589 [m, ν(CO)].
General Procedure for the Synthesis of K[M(L)]. Synthesis of
K[Zn(TEt)]. The pro-ligand H3(TEt) (216 mg, 0.688 mmol) was
dissolved in DMF (3 mL) and added to solid KH (110 mg, 2.75
mmol). Dihydrogen bubbles evolved immediately. The reaction was
left to stir at room temperature for 7 h. Zn(OAc)2 (126 mg, 0.688
mmol) was added, and DMF (1 mL) was used to wash the solid into
the reaction vial. The reaction was stirred again at room temperature
for 11 h. The resulting pale-yellow suspension was filtered, and the
solid K(OAc) was washed with DMF (3 × 1 mL). The combined
washings were concentrated to ca. 1−2 mL, and to this was added
Et2O (4 mL). The resulting suspension was cooled to −10 °C to
promote further precipitation. The resulting solid was isolated by
filtration and purified by successive washes with THF (3 mL), MeCN
Synthesis of H3(TNPh). 2-(N-Methyl-N-phenylamino)benzoic acid
(2.190 mg, 9.636 mmol), NHS (2.441 mg, 21.21 mmol), and EDAC
(4.070 g, 21.21 mmol) were added to a reaction flask and placed under
nitrogen, and CH2Cl2 (50 mL) was added to the reactants. The
solution was stirred at room temperature for 6 h. An aqueous solution
of NaHCO3 (35 mL) was added to the reaction and extracted with
CH2Cl2 (3 × 40 mL). The combined organic layers were dried with
MgSO4 and filtered, and the volatile solvents were evaporated under
vacuum to afford a yellow solid (84% of isolated yield). The product
was dried further by washing with diethyl ether (3 × 10 mL) and
drying under vacuum (18 h). A solution of the dried product (500 mg,
1.54 mmol) in CH2Cl2 (70 mL) was added to a solution of tren (75
mg, 0.51 mmol) in CH2Cl2 (2 mL). The reaction was stirred at 40 °C
for 3 h, after which the volatile solvents were evaporated under
vacuum to afford a light-brown solid. The crude product was purified
by column chromatography (3% MeOH in CH2Cl2) with neutralized
silica (washed with 1% NEt3 in CH2Cl2). The product H3(TNPh) was
isolated as an off-white solid. Yield: 355 mg (76% over two steps). 1H
NMR (CDCl3, 301 MHz): δ 8.20−8.08 (m, 6H, Ar and NHCO),
7.47−7.39 (m, 3H, Ar), 7.36−7.28 (m, 3H, Ar), 7.16−7.07 (m, 9H,
Ar), 3.13−3.08 (m, 15H, NCH2CH2NH and NCH3Ph), 2.34 (t, J =
6.0 Hz, 6H, NCH2CH2NH). 13C NMR (CDCl3, 126 MHz): δ 166.1
(s, C(O)NH), 149.1 (s, Ar), 147.9 (s, Ar), 132.6 (s, Ar), 131.4 (s, Ar),
131.3 (s, Ar), 129.2 (s, Ar), 127.9 (s, Ar), 126.6 (s, Ar), 119.8 (s, Ar),
1
(3 mL), and Et2O (3 mL). Yield: 138 mg (48%). H NMR (DMF-d7,
500 MHz): δ 3.32 (t, J = 5 Hz, C(O)NCH2, 6H), 2.56 (t, J = 5 Hz,
C(O)NCH2CH2N, 6H), 2.13 (q, J = 5 Hz, CH2CH3, 6H), 1.07 (t, J =
5 Hz, CH2CH3, 9H). 13C{1H} NMR (DMF-d7, 126 MHz): δ 176.2 (s,
NC(O)Et), 53.0 (s, C(O)NCH2CH2N), 42.9 (s, C(O)NCH2CH2N),
35.1 (s, CH2CH3), 12.2 (s, CH2CH3). IR (KBr pellet, cm−1): 1665 [s,
ν(CO)], 1569 [s, ν(CO)]. MALDI MS (pyrene matrix): Calcd
([Zn(TEt)]−): m/z 375.1. Obsd: m/z 375.3.
Characterization of K[Zn(TNBn)]. Yield: 102 mg (29%). 1H NMR
(DMF-d7, 499 MHz): δ 7.18−7.39 (m, Ar-H, 30H), 3.61 (s, NCH2Ph,
12H), 3.36 (m, C(O)NCH2CH2, 6H), 3.14 (s, C(O)CH2N, 6H), 2.53
(m, C(O)NCH2CH2N, 6H). 13C{1H} NMR (CD3CN, 126 MHz): δ
173.3 (s, C(O)N), 141.7 (s, Ar), 130.4 (s, Ar), 129.4 (s, Ar), 128.0 (s,
Ar), 62.8 (s, CH2), 58.8 (s, CH2), 54.6 (s, CH2), 42.7 (s, CH2). IR
(KBr pellet, cm−1): 1675 [s, ν(CO)], 1576 [s, ν(CO)]. MALDI
MS (pyrene matrix): Calcd ([Zn(TNBn)]−): m/z 918.4. Obsd: m/z
918.5.
K[Co(TEt)]. Yield: 116 mg (39%).
K[Co(TNBn)]. Yield: 176 mg (43%).
General Procedure for the Synthesis of NEt4[Co(L)]. Syn-
thesis of NEt4[Co(TEt)]. A solution of NEt4Cl (47 mg, 0.28 mmol) in
CH3CN (4 mL) was added to solid K[Co(TEt)] (116 mg, 0.283
mmol). The suspension was stirred at room temperature for 3 days.
9251
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
The reaction was filtered to obtain KCl as a white solid and the
product NEt4[Co(TEt)] as a blue filtrate. The filtrate was
concentrated to ca. 2 mL and passed through a microfiber glass filter
plug to remove any residual KCl, and Et2O (12 mL) was added to the
solution. Pure product slowly crystallized after cooling to −10 °C.
Combining metalation with cation exchange is feasible; however, these
reactions consistently resulted in lower overall purity. This is due to
the similar relative solubilities of NEt4[Co(L)] and any contaminating
ligand, which hinders reprecipitation or crystallization attempts. Yield:
(2) Gewirth, A. A.; Thorum, M. S. Inorg. Chem. 2010, 49, 3557.
(3) Saveant, J.-M. Chem. Rev. 2008, 108, 2348.
́
(4) DuBois, D. L.; Bullock, R. M. Eur. J. Inorg. Chem. 2011, 2011,
1017.
(5) Rakowski Dubois, M.; Dubois, D. L. Acc. Chem. Res. 2009, 42,
1974.
(6) Helm, M. L.; Stewart, M. P.; Bullock, R. M.; DuBois, M. R.;
DuBois, D. L. Science 2011, 333, 863.
(7) Wiese, S.; Kilgore, U. J.; Ho, M.-H.; Raugei, S.; DuBois, D. L.;
Bullock, R. M.; Helm, M. L. ACS Catal. 2013, 3, 2527.
(8) Mock, M. T.; Chen, S.; O’Hagan, M.; Rousseau, R.; Dougherty,
W. G.; Kassel, W. S.; Bullock, R. M. J. Am. Chem. Soc. 2013, 135,
11493.
(9) Yang, J. Y.; Bullock, R. M.; Dougherty, W. G.; Kassel, W. S.;
Twamley, B.; DuBois, D. L.; Rakowski DuBois, M. Dalton Trans. 2010,
39, 3001.
(10) Tronic, T. A.; Rakowski DuBois, M.; Kaminsky, W.; Coggins,
M. K.; Liu, T.; Mayer, J. M. Angew. Chem., Int. Ed. 2011, 50, 10936.
(11) Tronic, T. A.; Kaminsky, W.; Coggins, M. K.; Mayer, J. M. Inorg.
Chem. 2012, 51, 10916.
(12) McGuire, R., Jr.; Dogutan, D. K.; Teets, T. S.; Suntivich, J.;
Shao-Horn, Y.; Nocera, D. G. Chem. Sci. 2010, 1, 411.
(13) Dogutan, D. K.; Stoian, S. A.; McGuire, R.; Schwalbe, M.; Teets,
T. S.; Nocera, D. G. J. Am. Chem. Soc. 2011, 133, 131.
(14) Carver, C. T.; Matson, B. D.; Mayer, J. M. J. Am. Chem. Soc.
2012, 134, 5444.
1
45 mg (38%). H NMR (CD3CN, 499 MHz): δ 90.02 (br s), 84.05
(br s), 3.35 (s, NCH2CH3, 8H), 1.37 (s, NCH2CH3, 12H), −0.98 (br
s). μB (Evans’ method, DMF-d7, 298 K): μeff = 4.36. IR (KBr pellet,
cm−1): 1676 [m, ν(CO)], 1655 [m, ν(CO)], 1558 [s, ν(CO)].
UV [1:1 THF/DMF; λmax, nm (ε, M−1 cm−1)]: 404 (30), 584 (100),
609 (93). X-band EPR (1:1 THF/DMF): 77 K, g = 3.78; 10 K, g⊥ =
4.28, g|| = 2.00 (Az = 96 × 10−4 cm−1). MALDI MS (pyrene matrix):
Calcd ([Co(TEt)]−): m/z 370.1. Obsd: m/z 370.2.
1
Characterization of NEt4[Co(TNBn)]. Yield: 173 mg (39%). H
NMR (CD3CN, 499 MHz): δ 78.78 (br), 6.81 (br s), 6.00 (br s), 5.25
(br s), 3.27 (br s, NCH2CH3, 8H), 1.30 (br s, NCH2CH3, 12H),
−6.05 (br s). μB (Evans’ method, DMF-d7, 298 K): μeff = 4.21. IR (KBr
pellet, cm−1): 1675 [s, ν(CO)], 1576 [s, ν(CO)]. UV [1:1 THF/
DMF; λmax nm (ε, M−1 cm−1)]: 412 (31), 588 (104), 614 (103). X-
band EPR (1:1 THF/DMF, 20 mM): 77 K, g = 3.83; 10 K, g⊥ = 4.30,
g|| = 2.00 (Az = 98 × 10−4 cm−1). MALDI MS (pyrene matrix): Calcd
([Co(TNBn)]−): m/z 913.4. Obsd: m/z 913.5.
(15) Matson, B. D.; Carver, C. T.; Von Ruden, A.; Yang, J. Y.; Raugei,
S.; Mayer, J. M. Chem. Commun. 2012, 48, 11100.
(16) Collman, J. P.; Wagenknecht, P. S.; Hutchison, J. E. Angew.
Chem., Int. Ed. 1994, 33, 1537.
(17) Chang, C. K.; Liu, H. Y.; Abdalmuhdi, I. J. Am. Chem. Soc. 1984,
106, 2725.
(18) Halime, Z.; Kotani, H.; Li, Y.; Fukuzumi, S.; Karlin, K. D. Proc.
Natl. Acad. Sci. U. S. A. 2011, 108, 13990.
(19) Boulatov, R.; Collman, J. P.; Shiryaeva, I. M.; Sunderland, C. J. J.
ASSOCIATED CONTENT
* Supporting Information
■
S
X-ray crystallographic data in CIF format, additional exper-
imental details (including procedures for reactivity studies),
computational details, NMR, IR, and MALDI MS spectra, UV−
vis and 1H NMR titration data, and additional XRD tables. This
material is available free of charge via the Internet at http://
pubs.acs.org.
Am. Chem. Soc. 2002, 124, 11923.
(20) Rosenthal, J.; Nocera, D. G. Acc. Chem. Res. 2007, 40, 543.
(21) Kadish, K. M.; Shen, J.; Fremond, L.; Chen, P.; Ojaimi, M. E.;
́
Chkounda, M.; Gros, C. P.; Barbe, J.-M.; Ohkubo, K.; Fukuzumi, S.;
Guilard, R. Inorg. Chem. 2008, 47, 6726.
(22) Askarizadeh, E.; Yaghoob, S. B.; Boghaei, D. M.; Slawin, A. M.
Z.; Love, J. B. Chem. Commun. 2010, 46, 710.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: jblacqu2@uwo.ca.
*E-mail: james.mayer@yale.edu.
■
Author Contributions
J.M.B. led the project and designed, prepared, and characterized
the C3-symmetric ligands and their complexes, in collaboration
with A.F. M.L.P. performed all of the studies of the pyridyl
ligand. S.R. performed all of the computational studies and
W.K. the crystallographic studies. S.A.C. and T.T. performed
and interpreted the EPR experiments. The manuscript was
written through contributions of all authors. All authors have
given approval to the final version of the manuscript.
(23) Devoille, A. M. J.; Love, J. B. Dalton Trans. 2012, 41, 65.
(24) Collman, J. P.; Boulatov, R.; Sunderland, C. J.; Fu, L. Chem. Rev.
2004, 104, 561.
(25) Holm, R. H.; Kennepohl, P.; Solomon, E. I. Chem. Rev. 1996,
96, 2239.
(26) Thorseth, M. A.; Letko, C. S.; Tse, E. C. M.; Rauchfuss, T. B.;
Gewirth, A. A. Inorg. Chem. 2013, 52, 628.
(27) Jones, M. B.; MacBeth, C. E. Inorg. Chem. 2007, 46, 8117.
(28) Das, U. K.; Bobak, J.; Fowler, C.; Hann, S. E.; Petten, C. F.;
Dawe, L. N.; Decken, A.; Kerton, F. M.; Kozak, C. M. Dalton Trans.
2010, 39, 5462.
Notes
The authors declare no competing financial interest.
(29) Searls, C. E.; Kleespies, S. T.; Eppright, M. L.; Schwartz, S. C.;
Yap, G. P. A.; Scarrow, R. C. Inorg. Chem. 2010, 49, 11261.
(30) Harman, W. H.; Harris, T. D.; Freedman, D. E.; Fong, H.;
Chang, A.; Rinehart, J. D.; Ozarowski, A.; Sougrati, M. T.; Grandjean,
F.; Long, G. J.; Long, J. R.; Chang, C. J. J. Am. Chem. Soc. 2010, 132,
18115.
ACKNOWLEDGMENTS
■
Drs. T. A. Tronic, M. T. Mock, Z. M. Heiden, and C. J. Weiss
of the Center for Molecular Electrocatalysis and Dr. A. S.
Borovik of the University of CaliforniaIrvine are thanked for
helpful discussions. This research was supported as part of the
Center for Molecular Electrocatalysis, an Energy Frontier
Research Center funded by the U.S. Department of Energy,
Office of Science, Office of Basic Energy Sciences. The National
Science and Engineering Research Council of Canada is
thanked for a PDF award for J.M.B.
(31) Ward, A. L.; Elbaz, L.; Kerr, J. B.; Arnold, J. Inorg. Chem. 2012,
51, 4694.
(32) Fukuzumi, S.; Kotani, H.; Lucas, H. R.; Doi, K.; Suenobu, T.;
Peterson, R. L.; Karlin, K. D. J. Am. Chem. Soc. 2010, 132, 6874.
(33) Shook, R. L.; Peterson, S. M.; Greaves, J.; Moore, C.; Rheingold,
A. L.; Borovik, A. S. J. Am. Chem. Soc. 2011, 133, 5810.
(34) Moore, C. M.; Quist, D. A.; Kampf, J. W.; Szymczak, N. K. Inorg.
Chem. 2014, 53, 3278.
REFERENCES
(1) Winter, M.; Brodd, R. J. Chem. Rev. 2004, 104, 4245.
■
(35) Moore, C. M.; Szymczak, N. K. Dalton Trans. 2012, 41, 7886.
9252
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253

Inorganic Chemistry
Article
(36) Sickerman, N. S.; Park, Y. J.; Ng, G. K. Y.; Bates, J. E.; Hilkert,
M.; Ziller, J. W.; Furche, F.; Borovik, A. S. Dalton Trans. 2012, 41,
4358.
(37) Park, Y. J.; Sickerman, N. S.; Ziller, J. W.; Borovik, A. S. Chem.
Commun. 2010, 46, 2584.
(69) Drago, R. S. Physical Methods for Chemists, 2nd ed.; Surfside
Scientific Publishers: Gainesville, FL, 1992.
(70) Pfaff, F. F.; Kundu, S.; Risch, M.; Pandian, S.; Heims, F.;
Pryjomska-Ray, I.; Haack, P.; Metzinger, R.; Bill, E.; Dau, H.; Comba,
P.; Ray, K. Angew. Chem., Int. Ed. 2011, 50, 1711.
(71) Drago, R. S.; Corden, B. B. Acc. Chem. Res. 1980, 13, 353.
(72) Tovrog, B. S.; Kitko, D. J.; Drago, R. S. J. Am. Chem. Soc. 1976,
98, 5144.
(73) Izutsu, K. Acid−Base Dissociation Constants in Dipolar Aprotic
Solvents; Blackwell Scientific: Oxford, U.K., 1990.
(38) Matsumoto, J.; Suzuki, T.; Kajita, Y.; Masuda, H. Dalton Trans.
2012, 41, 4107.
(39) Szajna-Fuller, E.; Chambers, B. M.; Arif, A. M.; Berreau, L. M.
Inorg. Chem. 2007, 46, 5486.
(40) Berreau, L. M. Eur. J. Inorg. Chem. 2006, 2006, 273.
(41) Lee, J. H.; Park, J.; Lah, M. S.; Chin, J.; Hong, J.-I. Org. Lett.
2007, 9, 3729.
(42) Metteau, L.; Parsons, S.; Mareque-Rivas, J. C. Inorg. Chem. 2006,
45, 6601.
(43) Rivas, J. C. M.; Hinchley, S. L.; Metteau, L.; Parsons, S. Dalton
Trans. 2006, 2316.
(44) Matson, E. M.; Bertke, J. A.; Fout, A. R. Inorg. Chem. 2014, 53,
(74) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.;
̈
̈
̈
Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019.
(75) Jones, R. D.; Summerville, D. A.; Basolo, F. Chem. Rev. 1979, 79,
139.
(76) Drago, R. S.; Cannady, J. P.; Leslie, K. A. J. Am. Chem. Soc. 1980,
102, 6014.
(77) To the best of our knowledge, ε values for trigonal-bipyramidal
cobalt(III) superoxo complexes are not available. We use a typical
value here for related octahedral species to obtain a rough estimate of
Keq for dioxygen coordination in the systems reported herein.
(78) Zhao, Y.; Truhlar, D. G. J. Chem. Phys. 2006, 125.
(79) Radon, M.; Pierloot, K. J. Phys. Chem. A 2008, 112, 11824.
́
(80) Ali, M. E.; Sanyal, B.; Oppeneer, P. M. J. Phys. Chem. B 2012,
116, 5849.
(81) Smith, T. D.; Pilbrow, J. R. Coord. Chem. Rev. 1981, 39, 295.
(82) MacBeth, C. E.; Hammes, B. S.; Young, V. G.; Borovik, A. S.
Inorg. Chem. 2001, 40, 4733.
(83) Yandulov, D. V.; Schrock, R. R. Science 2003, 301, 76.
(84) Schrock, R. R. Acc. Chem. Res. 2005, 38, 955.
(85) Breitenmoser, R. A.; Heimgartner, H. Helv. Chim. Acta 2001, 84,
786.
4450.
(45) Lacy, D. C.; Mukherjee, J.; Lucas, R. L.; Day, V. W.; Borovik, A.
S. Polyhedron 2013, 52, 261.
(46) Ng, G. K. Y.; Ziller, J. W.; Borovik, A. S. Chem. Commun. 2012,
48, 2546.
(47) Parsell, T. H.; Yang, M.-Y.; Borovik, A. S. J. Am. Chem. Soc.
2009, 131, 2762.
(48) Mukherjee, J.; Lucas, R. L.; Zart, M. K.; Powell, D. R.; Day, V.
W.; Borovik, A. S. Inorg. Chem. 2008, 47, 5780.
(49) Shook, R. L.; Borovik, A. S. Chem. Commun. 2008, 6095.
(50) Lucas, R. L.; Zart, M. K.; Murkerjee, J.; Sorrell, T. N.; Powell, D.
R.; Borovik, A. S. J. Am. Chem. Soc. 2006, 128, 15476.
(51) Borovik, A. S. Acc. Chem. Res. 2004, 38, 54.
(52) Kim, S.; Saracini, C.; Siegler, M. A.; Drichko, N.; Karlin, K. D.
Inorg. Chem. 2012, 51, 12603.
(53) Syuhei, Y.; Hideki, M. Sci. Technol. Adv. Mater. 2005, 6, 34.
(54) Cheruzel, L. E.; Cecil, M. R.; Edison, S. E.; Mashuta, M. S.;
Baldwin, M. J.; Buchanan, R. M. Inorg. Chem. 2006, 45, 3191.
(55) Wada, A.; Yamaguchi, S.; Jitsukawa, K.; Masuda, H. Angew.
Chem., Int. Ed. 2005, 44, 5698.
(56) Yamaguchi, S.; Wada, A.; Funahashi, Y.; Nagatomo, S.;
Kitagawa, T.; Jitsukawa, K.; Masuda, H. Eur. J. Inorg. Chem. 2003,
2003, 4378.
(57) Alliger, G. E.; Muller, P.; Cummins, C. C.; Nocera, D. G. Inorg.
̈
Chem. 2010, 49, 3697.
(58) Smith, N. D.; Wohlrab, A. M.; Goodman, M. Org. Lett. 2005, 7,
255.
(59) Kim, Y. K.; Lee, S. J.; Ahn, K. H. J. Org. Chem. 2000, 65, 7807.
(60) Silverstein, R. M.; Webster, F. X.; Kiemle, D. J. Spectrometric
Identification of Organic Compounds, 7th ed.; John Wiley & Sons:
Hoboken, NJ, 2005.
(61) Suzuki, T.; Imai, K.; Nakagawa, H.; Miyata, N. ChemMedChem
2006, 1, 1059.
(62) Schatz, M.; Leibold, M.; Foxon, S. P.; Weitzer, M.; Heinemann,
F. W.; Hampel, F.; Walter, O.; Schindler, S. Dalton Trans. 2003, 1480.
(63) Several samples of [Co(L)]− were analyzed by elemental
analysis, and all were slightly outside the accepted level of error. We do
not attribute the deviation to a systematic impurity given that the
magnitude and type (i.e., %C vs %H) of error were not consistent
across the samples.
(64) Niklas, N.; Wolf, S.; Liehr, G.; Anson, C. E.; Powell, A. K.;
Alsfasser, R. Inorg. Chim. Acta 2001, 314, 126.
(65) Girolami, G. S.; Rauchfuss, T. B.; Angelici, R. J. Synthesis and
Technique in Inorganic Chemistry: A Laboratory Manual, 3rd ed.;
University Science Books: Sausalito, CA, 1999.
(66) Ray, M.; Hammes, B. S.; Yap, G. P. A.; Rheingold, A. L.;
Borovik, A. S. Inorg. Chem. 1998, 37, 1527.
(67) Eelman, M. D.; Blacquiere, J. M.; Moriarty, M. M.; Fogg, D. E.
Angew. Chem., Int. Ed. 2008, 47, 303.
(68) Lacy, D. C.; Park, Y. J.; Ziller, J. W.; Yano, J.; Borovik, A. S. J.
Am. Chem. Soc. 2012, 134, 17526.
9253
dx.doi.org/10.1021/ic5013389 | Inorg. Chem. 2014, 53, 9242−9253