Syntheses and electronic structure of bimetallic complexes containing a flexible redox-active bridging ligand

Article abstract of DOI:10.1021/ic302087f

The new ligand L¹, 1-N,1-N-bis(pyridine-2-ylmethyl)-3-N- (pyridine-2-ylmethylidene)benzene-1,3-diamine, was synthesized as a platform for the study of bimetallic complexes containing redox-active ligands. The asymmetric L¹ contains a redox-active α-iminopyridine unit bridged to redox-inert bis(2-pyridylmethyl)amino counterpart and offers two distinct coordination sites. The coordination chemistry of L¹ with Fe, Cu, and Zn was examined. Reaction with zinc afforded the asymmetric binuclear complex [(L¹)Zn₂Cl₄] (1), whereas the symmetric [(L¹)₂Fe₂(OTf)₂](OTf) ₂ (2) and [(L¹)₂Cu₂](OTf) ₄ (3) were isolated in reactions with iron and copper. Both metal-and ligand-centered redox processes are available to the series of metal compounds. EPR and Moessbauer spectroscopy and magnetic susceptibility studies establish that both 2 and 3 are paramagnetic; the vanishingly small ferromagnetic interaction produces decoupled high-spin Fe^II (S = 2) ions in 2. DFT calculations provide further insight into the nature of the exchange interactions in the dimeric systems.

Full text of DOI:10.1021/ic302087f

Article
pubs.acs.org/IC
Syntheses and Electronic Structure of Bimetallic Complexes
Containing a Flexible Redox-Active Bridging Ligand
Stacey Lindsay,† Siu K. Lo,† Oliver R. Maguire,† Eckhard Bill,‡ Michael R. Probert,† Stephen Sproules,§
and Corinna R. Hess*,†
†Department of Chemistry, University of Durham, South Rd, Durham, DH1 3LE, United Kingdom
‡Max-Planck-Institut fur Chemische Energiekonversion, Stiftstrasse 34-36, D-45470 Mulheim an der Ruhr, Germany
̈
̈
§School of Chemistry and Photon Science Institute, The University of Manchester, Oxford Road, Manchester M13 9PL, United
Kingdom
S
* Supporting Information
ABSTRACT: The new ligand L1, 1-N,1-N-bis(pyridine-2-ylmethyl)-
3-N-(pyridine-2-ylmethylidene)benzene-1,3-diamine, was synthesized
as a platform for the study of bimetallic complexes containing redox-
active ligands. The asymmetric L1 contains a redox-active α-
iminopyridine unit bridged to redox-inert bis(2-pyridylmethyl)amino
counterpart and offers two distinct coordination sites. The
coordination chemistry of L1 with Fe, Cu, and Zn was examined.
Reaction with zinc afforded the asymmetric binuclear complex
[(L1)Zn2Cl4] (1), whereas the symmetric [(L1)2Fe2(OTf)2](OTf)2
(2) and [(L1)2Cu2](OTf)4 (3) were isolated in reactions with iron
and copper. Both metal- and ligand-centered redox processes are
available to the series of metal compounds. EPR and Mossbauer
̈
spectroscopy and magnetic susceptibility studies establish that both 2
and 3 are paramagnetic; the vanishingly small ferromagnetic interaction produces decoupled high-spin FeII (S = 2) ions in 2.
DFT calculations provide further insight into the nature of the exchange interactions in the dimeric systems.
[ONO] pincer-type ligands also have been isolated;10,15 the
INTRODUCTION
■
ligand plays a key role in oxidative formation of diazenes from
the Ta2 complex.16 The latter example nicely illustrates that the
electron storage capabilities of ligands can provide a powerful
tool for multielectron transformations. However, the ramifica-
tions for the reactivity of binuclear compounds containing
organic cofactors have yet to be fully explored.
The coordination chemistry of redox-active ligands has received
considerable attention in recent years.1 Metal complexes
containing molecules such as the α-diimines, dioxolenes, and
dithiolenes have been shown to not only defy conventional
oxidation state assignments but exhibit unique electronic
structures and rich chemistry.2,3 Notably, the additional
electron storage site afforded by the coordinated organic
molecules can be advantageous for reactivity:4 among the
examples, ligand-centered redox processes have been implicated
in C−C bond formation, alkene addition, and nitrene transfer
reactions by transition metal and actinide complexes containing
the aforementioned noninnocent moieties.5−10
Examples of bimetallic complexes containing redox-active
ligands are fewer but offer further intrigue with respect to both
reactivity and magnetism. The most common binuclear motif
consists of two metal ions connected by a redox-active bridge
that directly moderates exchange interactions between the
sites.11−13 These, predominantly symmetric, compounds have
attracted attention due to their relevance for molecular
electronics. Other examples include the dimeric [Fe(TIM)]2
(TIM = 2,3,9,10-tetramethyl-1,4,8,11-tetraazacyclotetradeca-
1,3,8,10-tetraene) in which ligand radicals contribute to
metal−metal bond formation and mixed valency.14 Imido-
bridged Zr and Ta dimers containing redox-active [NNN] and
We have now synthesized the asymmetric ligand L1 (1-N,1-
N-bis(pyridine-2-ylmethyl)-3-N-(pyridine-2-ylmethylidene)-
benzene-1,3-diamine; Scheme 1), as part of our initial efforts to
explore the redox properties and reactivity of binuclear
compounds containing redox-active ligands. In contrast to the
aforementioned binuclear systems, L1 was designed with two
distinct coordination sites, supplied by an α-iminopyridine and
a bis(2-pyridylmethyl)amino group. The two-electron redox
series available to α-iminopyridines, as a consequence of low-
lying π* orbitals, previously has been established,17 whereas the
pyridylmethylamino groups serve as innocent counterparts.
The divergent nature of the two functional groups offers a
strategy for selective reduction of one site in a binuclear
complex, thus permitting mixed valency unconstrained by the
accessibility of a particular metal oxidation state. In the case of
Received: September 26, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1H NMR (400 MHz, CDCl3): δ 8.58 (ddd, J = 5, 2, 1 Hz, 2H), 7.62
(td, J = 8, 2 Hz, 2H), 7.25 (d, 2H), 7.16 (ddd, J = 7, 5, 1 Hz, 2H), 7.07
(t, J = 8 Hz, 1H), 6.89 (br, 1H) 6.58 (t, J = 2 Hz, 1H), 6.38 (dd, J = 8,
3 Hz, 1H), 6.35 (s, br, 1H) 4.79 (s, 4H) 1.46 (s, 9H). 13C NMR (176
MHz, CDCl3): δ 158.9, 152.8, 149.8, 149.1, 139.6, 137.0, 130.0, 122.2,
121.1, 108.1, 107.8, 102.8, 80.42, 57.31, 28.46. MS (LRMS+; ESI; m/
z): 391.4 [M + H], 185.2 [M+ −2Me-Py], 102.2 [M+ − (BOC + 2Me-
Py)].
Scheme 1. Two-Electron Redox Series for the α-
Iminopyridine Group with Characteristic Bond Distances
(Angstroms)17
1-N,1-N-Bis(pyridin-2-ylmethyl)benzene-1,3-diamine (III).
IIb (1.00 g; 2.56 mmol) was dissolved in ethyl acetate (30 mL) and
then treated dropwise with 12 M HCl (10 mL). The reaction mixture
was left to stir for 1 h. The resultant solution was combined with 5 M
NaOH (50 mL), and the organic layer was extracted with CH2Cl2 (3 ×
50 mL). Solvent was removed in vacuo to yield the brown solid III
(0.7 g, 98% yield).
1H NMR (400 MHz, CDCl3): δ 8.58 (ddd, J = 5, 2, 1 Hz, 2H) 7.62
(td, J = 8, 2 Hz, 2H) 7.28 (d, 2H) 7.16 (dd, J = 7, 5 Hz, 2H) 6.95 (t, J
= 8 Hz, 1H) 6.14 (dd, J = 8, 2 Hz, 1H), 6.09 (dd, J = 8, 2 Hz, 1H),
6.02 (t, J = 2 Hz, 1H) 4.78 (s, 4H), 3.51 (s, 2H). 13C NMR (176 MHz,
CDCl3): δ 159.2, 149.8, 149.6, 147.7, 137.0, 130.3, 122.1, 121.0, 104.9,
103.7, 99.49, 57.38. MS (LRMS+; ESI; m/z): 291.3 [M + H], 200.2
[M+ − Me-Py], 109.2 [M+ − 2Me-Py].
ligand-centered reduction, as opposed to metal-centered
reduction, mixed valency refers to ‘formal’ rather than ‘physical’
oxidation states18 but equally describes charge localization in
the bimetallic complex.
The coordination chemistry of L1 with Fe, Cu, and Zn is
described herein. The series of metal complexes includes both
symmetric and asymmetric binuclear complexes. Electro-
chemical measurements provide an indication of the nature of
1-N,1-N-Bis(pyridine-2-ylmethyl)-3-N-(pyridine-2-
ylmethylidene)benzene-1,3-diamine (L1). A solution of 2-
pyridinecarboxaldehyde (180 μL; 1.89 mmol) and III (0.50 g; 1.72
mmol) in dry toluene (90 mL) was heated to reflux at 120 °C, over 4
Å molecular sieves, under an argon atmosphere for 5 h. Solvent was
removed in vacuo to give the yellow solid L1 (0.60 g, 90% yield).
1H NMR (400 MHz, CDCl3): δ 8.68 (d, J = 4 Hz, 1H), 8.59 (d, J =
4 Hz, 2H), 8.51 (s, 1H) 8.11 (d, J = 8 Hz, 1H) 7.78 (t, J = 8 Hz, 1H)
7.63 (t, J = 7 Hz, 2H) 7.34 (td, J = 6 Hz, 1H) 7.29 (d, J = 8 Hz, 2H)
7.17 (m, 3H) 6.64 (m, 3H) 4.85 (s, 4H). 13C NMR (176 MHz,
CDCl3): δ 160.3, 158.5, 154.5, 152.3, 149.7, 149.7, 149.3, 136.8, 136.6,
130.0, 125.0, 122.1, 122.0, 120.8, 110.9, 109.2, 105.9, 57.1. MS
(LRMS+; ESI; m/z): 380.6 [M + H]. An accurate mass of the molecule
was determined using high-resolution MS and found to be within 6
ppm (HRMS; ASAP; m/z: 380.1898 [M + H]). UV−vis: λmax, nm (ε,
M−1 cm−1) in CH2Cl2: 254 (3.8 × 104).
the redox processes available to these complexes. The L1 −Cu2
2
and L1 −Fe2 complexes were further characterized by
2
spectroscopic, magnetic susceptibility, and DFT computational
studies, which denote weak ferromagnetic coupling mediated
by L1.
EXPERIMENTAL SECTION
■
Iron trifluoromethanesulfonate was purchased from Strem Chemicals;
all other reagents were obtained from Sigma Aldrich and used as
received. Metal compounds were synthesized in an inert atmosphere
glovebox, under nitrogen, using anhydrous solvents. Solvents were
dried by passage over activated alumina columns from Innovative
Technology, Inc. (Amesbury, MA) and stored over activated 3 Å
molecular sieves.
[(L1)Zn2Cl4] (1). Colorless ZnCl2 (26 mg; 0.19 mmol) was added to
a solution of L1 (36 mg; 0.09 mmol) in THF or MeCN (10 mL). The
mixture was stirred overnight, and the resultant yellow precipitate was
filtered to give 1 (43 mg, 70% yield). Diffraction-quality single crystals
were obtained by slow evaporation of a concentrated solution of 1 in
MeCN.
(3-Aminophenyl)carbamic Acid tert-Butyl Ester (I). Synthesis
of the BOC-protected 1,3-benzenediamine was analogous to
procedures previously described.19 A solution of di-tert-butyl-
dicarbonate (10.6 mL; 0.046 mol) in 1,4-dioxane (15 mL) was
added dropwise to a solution of m-phenylenediamine (5.00g; 0.046
mol) in 1,4-dioxane (15 mL). Triethylamine (6.4 mL; 0.046 mol) was
then added to the reaction mixture, and the solution was heated at 60
°C for 24 h. Solvent was removed in vacuo to yield a red-brown oil.
Crude product was purified via flash column chromatography (2:3
ethyl acetate/hexanes; Rf = 0.2) to give the desired product, (3-
aminophenyl)carbamic acid tert-butyl ester (6.5 g, 67% yield).
1H NMR (400 MHz, CDCl3): δ 7.04 (t, J = 9 Hz, 1H), 6.97 (s,
1H), 6.55 (ddd, J = 8, 2, 1 Hz, 1H), 6.37 (br, 1H), 6.37 (ddd, J = 8, 2,
1,1H), 3.67 (s, br, 2H), 1.51 (s, 9H). 13C NMR (100 MHz, CDCl3): δ
152.8, 147.4, 139.5, 129.8, 110.0, 108.8, 105.3, 80.5, 28.5.
{3-[Bis(pyridine-2-ylmethyl)amino]phenyl}amino-2,2-dime-
thylpropanoate (IIb). To a solution of (3-aminophenyl)carbamic
acid tert-butyl ester (I: 4.00 g; 0.019 mol) in anhydrous ethanol (60
mL) was added Na2CO3 (8.14 g; 0.077 mol). 2-(Chloromethyl)-
pyridine hydrochloride (6.3 g; 0.038 mol) was then added, and the
mixture was heated to reflux at 80 °C for 24 h. The resulting brown
suspension was filtered, and the solvent was removed in vacuo to yield
a brown oil. The brown oil was diluted with CH2Cl2 (40 mL) and
treated with 2 M NaOH (50 mL). Product was extracted with CH2Cl2
(3 × 50 mL), then the organic layer was washed with brine (2 × 50
mL) and dried over MgSO4, and the CH2Cl2 was removed under
vacuum. The crude product was run through a silica plug (125 cm3
silica; ethyl acetate; Rf = 0.2) to give the desired product as an orange
oil (4.1 g, 44% yield).
1H NMR (700 MHz, CD2Cl2): δ 9.24 (d, J = 5 Hz, 2H), 9.01 (s,
1H), 8.81 (d, J = 5 Hz,1H), 8.27 (td, J = 8, 2 Hz, 1H) 8.08 (d, J = 8
Hz, 1H) 7.93 (m, 3H) 7.87 (ddd, J = 8, 5, 1 Hz, 1H), 7.52 (t, J = 6 Hz,
2H) 7.51 (d, J = 8 Hz, 2H) 7.40 (m, 3H). Anal. Calcd for
C24H21Cl4N5Zn2: C, 44.21; H, 3.25; N, 10.74. Found: C, 43.99; H,
3.25; N, 10.80. UV−vis: λmax, nm (ε, M−1 cm−1) in CH2Cl2: 262 (∼1.6
× 104), 331 (∼1.6 × 104).
[(L1)2Fe2(OTf)2](OTf)2 (2). A suspension of Fe(OTf)2 (23 mg,
0.065 mmol) in THF (5 mL) was added to a solution of L1 (25 mg;
0.066 mmol) in THF (5 mL; THF was stored over Na; the yield of
product obtained was highly dependent on the quality of the reagents
and solvent) upon which the mixture immediately turned green. The
reaction mixture was stirred for 72 h, during which time a series of
color changes was observed, from green to blue to black, with the
appearance of a pink solid. The pink solid 2 was collected by filtration
(26 mg, 54% yield), leaving a charcoal gray filtrate. Additional products
of this reaction were not identified. Single crystals were obtained by
slow diffusion of diethyl ether into a solution of 2 in MeCN.
Anal. Calcd for C52H42N10O12S4F12Fe2: C, 42.58; H, 2.89; N, 9.55.
Found: C, 42.37; H, 2.96; N, 9.61. UV−vis: λmax, nm (ε, M−1 cm−1) in
MeCN: 515 (750).
[(L1)2Cu2](OTf)4 (3). In THF (5 mL) was dissolved L1 (25 mg;
0.066 mmol) and Cu(OTf)2 (25 mg; 0.066 mmol). The solution was
stirred overnight, resulting in a forest green suspension, which was
filtered to give the green solid 3 (36 mg, 74% yield). Single crystals
B
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 2. Synthesis of L1
were obtained by slow diffusion of diethyl ether into an acetonitrile
solution of 3.
Crystallography. Structure determinations were carried out from
single-crystal X-ray diffraction data collected at 100 K using Cu Kα
radiation (λ = 1.54178 Å) on a Bruker Pt135-CCD Proteum
diffractometer with multilayer focusing optics or at 120 K using Mo
Kα radiation (λ = 0.71073 Å) on a Bruker SMART 6K-CCD
diffractometer. Sample temperature was controlled and maintained
using an Oxford Cryosystems open-flow N2 cooling device.21 A series
of narrow φ and/or ω scans (0.5° or 0.3°) was performed at various
setting angles to maximize data coverage. Unit cell parameters were
determined and refined inside the APEX222 software suite, and raw
data were integrated using the SAINT program.23 Structures were
solved, refined, and publication material produced using the OLEX224
interface to the SHELXTL suite of programs.25 All structures
presented were affected, to variable degrees, by low-resolution data
being achievable, due to crystal size and quality. This was also coupled
in many cases with solvent and/or ligand disorder (further details are
included in the crystallographic information files). In particular, the
small size of 1 and the low crystal symmetry (P-1), coupled to the
orientation of the sample, limited the completeness of the diffraction
data for 1 at high resolution. In spite of the less than ideal
completeness, diffraction data were collected to a θmax of 68.33° with
Cu Kα radiation, giving a data to parameter ratio of 4179/316. Lower
than optimal completeness to high angle does marginally affect the
accuracy of the derived parameters but is correctly reflected in the
standard uncertainties recorded in the crystallographic information file
and reported parameters herein.
Density Functional Theory (DFT) Calculations. All DFT
calculations were performed with the ORCA program package.26
Geometry optimizations of the complexes were performed at the
B3LYP27−29 level of DFT. The all-electron Gaussian basis sets were
those developed by the Ahlrichs group.30,31 Triple-ζ quality basis sets
TZV(P) with one set of polarization functions on the metals and on
the atoms directly coordinated to the metal center were used.31 For
the carbon and hydrogen atoms, slightly smaller polarized split-valence
SV(P) basis sets were used that were of double-ζ quality in the valence
region and contained a polarizing set of d functions on the non-
hydrogen atoms.30 Auxiliary basis sets used to expand the electron
density in the resolution-of-the-identity (RI) approach were
chosen,32,33 where applicable, to match the orbital basis. SCF
calculations were tightly converged (1 × 10−8 Eh in energy, 1 ×
10−7 Eh in the density change, and 1 × 10−7 in maximum element of
the DIIS error vector). Geometry optimizations for all complexes were
carried out in redundant internal coordinates without imposing
symmetry constraints. In all cases the geometries were considered
converged after the energy change was less than 5 × 10−6 Eh, the
gradient norm and maximum gradient element were smaller than 1 ×
10−4 and 3 × 10−4 Eh Bohr−1, respectively, and the root-mean square
and maximum displacements of all atoms were smaller than 2 × 10−3
Anal. Calcd for C52H42N10O12S4F12Cu2: C, 42.13; H, 2.86; N, 9.45.
Found: C, 42.20; H, 3.01; N, 9.46. UV−vis: λmax, nm (ε, M−1 cm−1) in
MeCN: 596 (520). MS (LRMS+; ESI; m/z): 1333.3 [M − OTf], 592.2
[M − 2OTf], 221.2 [M − 4OTf].
Physical Measurements. NMR spectra were recorded on a
1
Bruker Avance-400 (400 MHz H, 100 MHz 13C, 376 MHz 19F) or a
Varian VNMRS-700 (700 MHz 1H, 176 MHz 13C) spectrometer.
Electronic spectra were recorded on a Perkin-Elmer Lambda 900
spectrophotometer. Diffuse reflectance spectra were obtained by
illumination of the sample using an Energetiq LDLS EQ-99 broad-
band lamp and collected at 20° to the excitation using an Ocean
Optics Maya Pro 2000 spectrometer. Data was recorded using the
Ocean Optics software, and integration times were adjusted to afford
maximum response of the spectrometer without saturation of the
detector. Teflon was used as the white standard. Mass spectra were
measured using a Waters TQD instrument for ESI (1 μL of a ∼1 mg/
mL sample in methanol or acetonitrile injected into a flow (0.2 mL/
min) of methanol or acetonitrile; capillary voltage 3000 V, cone
voltage 30 V) or on a Xevo QToF for high-resolution spectra
(atmospheric pressure solids analysis probe ionization experiments
(ASAP)). Microanalyses were carried out in the Chemistry Depart-
ment at Durham University. Electrochemical measurements were
carried out using an Autolab PG-STAT 30 potentiostat using a three-
electrode cell equipped with a Pt working electrode and Pt wire
counter and reference electrodes. Potentials are reported with
reference to an internal standard of ferrocenium/ferrocene (Fc+/0).
Magnetic susceptibility data (2−290 K) were recorded using a SQUID
magnetometer (MPMS7, Quantum Design) in a 1 T external field.
Data were corrected for underlying diamagnetism using tabulated
Pascal’s constants and fit using julX (Dr. E. Bill). Mossbauer data were
̈
recorded on an alternating constant-acceleration spectrometer. The
minimum experimental line width was 0.24 mm s−1 (full width at half-
height). Sample temperature was maintained constant in an Oxford
Instruments Variox or an Oxford Instruments Mossbauer-Spectromag
̈
2000 cryostat, which is a split pair superconducting magnet system for
applied fields (up to 8 T). The field at the sample is oriented
perpendicular to the γ-beam. The 57Co/Rh source (1.8 GBq) was
positioned at room temperature inside the gap of the magnet system at
a zero-field position. Isomer shifts are quoted relative to iron metal at
300 K; data were simulated using mfit (Dr. E. Bill). Multifrequency
EPR measurements were carried out at the EPSRC National UK EPR
Facility and Service in the Photon Science Institute at The University
of Manchester. X-Band spectra were collected using a Bruker EMX
Micro spectrometer, K-band spectra with a Bruker E580 spectrometer,
and Q-band spectra on a Bruker EMX spectrometer. Simulations were
performed using Bruker’s Xsophe software package.20
C
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 3. Synthesis of Compounds 1−3
Figure 1. Comparison of the electronic spectra of L1 (black) and 1 (blue) in CH2Cl2 (left) and spectra of 2 (red) and 3 (green) in MeCN (right).
and 4 × 10−3 Bohr, respectively. Orbital and spin density plots were
Reaction of the BOC-protected diamine (I)19 with 2-
(chloromethyl)pyridine hydrochloride affords a mixture of
both the mono- (IIa) and the bis(pyridine-2-ylmethyl)benzene-
1,3-diamine (IIb), which are readily separated by chromatog-
raphy. Deprotection of IIb followed by reaction with 2-
pyridinecarboxaldehyde cleanly generates the target molecule.
Subsequent examination of the coordination chemistry of L1
with Fe, Cu, and Zn allowed for comparison of the structural
preferences and redox properties of the resultant first-row, late
transition metal complexes. The unique electronic structures of
iron−diimine complexes have been noted previously, with
ligand-centered reduction predominantly favored over metal-
centered reduction.17,35,36 The paucity of low-valent iron
compounds contrasts with the prevalence of monovalent
copper, while the α-diimine group must be invoked in the
reduction of Zn-containing compounds.
created using GaussView.34
RESULTS AND DISCUSSION
■
Synthesis and Characterization. The new ligand L1 was
designed as a bimetallic scaffold, endowed with a redox-active
α-iminopyridine that could compete with coordinated metal
ions for electrons (Scheme 1). The meta disposition of the
bridging benzene moiety disfavors formation of a mononuclear
complex, while the asymmetry of the molecule imposed by the
disparate set of N-donor groups offers distinctive properties to
each intended coordination site. Importantly, ligand-centered
redox processes available to one coordination site could permit
charge localization in reduced binuclear compounds. The two-
electron redox series available to α-iminopyridines implies that
both one- and two-electron charge-separated states could
theoretically be accessed (Scheme 1). Thus, L1 offers an
alternative strategy to the rare metal-based two-electron mixed-
valence complexes.
Bimetallic products were obtained in all reactions of L1
(Scheme 3), irrespective of the M:L ratio (2:1 or 1:1).
However, an asymmetric binuclear complex, [(L1)Zn2Cl4] (1),
was obtained in the reaction with ZnCl2, while a symmetric
dimer, [(L1)2Fe2(OTf)2](OTf)2 (2), was the only identified
L1 was synthesized in four steps starting from the
commercially available m-phenylenediamine (Scheme 2).
D
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
1
Figure 2. Overlay of the aromatic region of the H NMR spectrum (700 MHz, CD2Cl2) of L1 (top) and 1 (bottom) with labeling scheme.
Assignments based on COSY spectra (see Supporting Information).
product upon reaction of L1 with Fe(OTf)2. An analogous
dimeric copper-containing complex, [(L1)2Cu2](OTf)4 (3),
also was obtained upon reaction of L1 with Cu(OTf)2.
The binuclear zinc complex, [(L1)Zn2Cl4] (1), was obtained
upon reaction of L1 with 2 equiv of ZnCl2. Compound 1 is
mildly soluble in CH2Cl2, and its electronic spectrum is clearly
dissimilar to that of L1, with a distinct absorption band at 331
nm (Figure 1). The proton resonances in the aryl region of the
1H NMR spectrum of 1 in CD2Cl2 are shifted downfield
significantly in comparison to the L1 spectrum (Figure 2).
Products obtained from the analogous reactions of L1 with
FeCl2 and CuCl2 likewise proved highly insoluble, which
obscured their characterization. Therefore, we sought alter-
native starting reagents to further investigate the ligand
coordination chemistry. Addition of 1 equiv of Fe(OTf)2 to a
solution of L1 in THF led to precipitation of 2 as a pink solid.
The series of color changes throughout the reaction suggests
formation of various intermediates en route to or products in
addition to 2; the yield of 2 also did not exceed 54%. However,
we have not identified any other products, and compound 2
was isolated in all reactions irrespective of the Fe:ligand
stoichiometry (2:1, 1:1, 1:2 examined) when the reaction was
carried out with Fe(OTf)2.
The electronic spectrum of 2 features a single absorption
band in the visible region, with λmax = 515 nm (Figure 1), which
we attribute to a metal-to-ligand charge transfer (MLCT)
transition. The MLCT band in the solid state spectrum of 2 is
slightly red shifted (λmax = 523 nm) in comparison to the
absorption band in acetonitrile solutions (Figure S7, Support-
ing Information). 19F NMR of 2 shows a single, broad peak at
−76 ppm (Figure S8, Supporting Information), indicative of
noncoordinated triflate anions; signal broadening is a result of
fast exchange of the triflate ions.37−39 Therefore, although 2
was crystallized from MeCN/Et2O mixtures, we cannot exclude
the possibility that in MeCN solutions the triflate ligands may
be substituted by solvent molecules.
The analogous copper dimer, [(L1)2Cu2](OTf)4 (3), was
synthesized in reactions with equimolar amounts of Cu(OTf)2
and L1 in THF. The electronic spectrum of 3 in MeCN (Figure
1) exhibits an absorption band at λmax = 596 nm (ε = 520 M−1
cm−1 ⇒ 260 M−1 cm−1 per Cu), which we assign to the CuII
d−d transition. The absorbance value is within the typical range
observed for related tetragonal Cu(II)−pyridine com-
pounds.40,41 Furthermore, a MLCT transition would be
expected to occur at higher energy than for the corresponding
diiron complexes.
Electrochemistry. Cyclic voltammograms (CV) of L1 and
1−3 are shown in Figure 3, and redox potentials are referenced
versus the Fc+/0 couple. The [L1]/[L1•]− couple associated
with the one-electron reduction of the α-diimine unit of L1
(Scheme 1) is observed at −0.86 V. The peak-to-peak
separation is ∼180 mV, suggesting structural rearrangement
upon reduction.42 A partially chemically reversible reduction is
observed at E1/2 = −1.31 V in the CV of 1 and readily described
as ligand centered.
An irreversible oxidation is seen in the CV of 2 (Figure S9,
Supporting Information) along with two closely spaced
reduction events at −1.2 and −1.3 V, respectively. Reduction
of the diiron complex occurs at slightly more negative
potentials than observed for L1. In contrast to 1, both metal-
and ligand-centered redox processes are feasible for 2. Previous
studies of iron complexes containing diimine ligands have
demonstrated that reduction of the ligand commonly is favored
over formation of low-valent metal oxidation states.17,35,43
However, numerous intriguing possibilities exist with respect to
the nature of the [(L1)2Fe2]4+/3+ and [(L1)2Fe2]3+/2+ couples
and the distribution of two additional electrons among the four
redox-active centers in 2.
The [(L1)2Cu2]4+/3+ couple of 3 appears at a significantly
more positive potential (E1/2 = −0.44 V) than for 2, denoting
metal-centered reduction. This is in agreement with the
observed trend of increasing metal character in a series of
first-row bis(iminopyridine) complexes across the period from
E
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
accommodate two distinct metal coordination sites. The two
zinc sites are distorted tetrahedral, with a constrained 80° bite
angle imposed by the α-iminopyridine at Zn(1). The
coordination geometry at the Zn(2) site is analogous to a
related symmetric binuclear zinc complex, [Zn2(1,3-tpbd)Cl4]
(1,3-tpbd = N,N,N′,N′-tetrakis(pyridine-2-ylmethyl)benzene-
1,3-diamine),44 with only a weak interaction between Zn(2)
and the amino nitrogen (Zn(2)−N(3) = 2.624(2)). The
chloride ligands prohibit close approach of the two zinc ions,
leading to a Zn···Zn separation of ∼5.7 Å.
In the iron and copper dimers each metal ion is coordinated
by an α-iminopyridine group of one L1 unit and the
bis(pyridylmethyl)amino group of a second ligand. Two
solvatomorphs of [(L1)2Fe2(OTf)2](OTf)2 were isolated
(Figure 5 and Table 1, denoted by 2·4MeCN and
2·2MeCN), which differ in the orientation of the coordinated
triflate ligand and the number of solvent molecules. The FeII
centers in both structures are octahedral: the N atoms of the
bis(pyridylmethyl)amino groups adopt a facial arrangement, the
opposing face comprised of the α-iminopyridine, and a
coordinated triflate anion (Fe−O ≈ 2.1 Å). The Fe−N bond
distances are indicative of high-spin FeII centers, as would be
expected given the coordinated triflate ion. The coordination
geometry at the iron center coincides with the ‘slipped’
arrangement of the L1 bridging benzenes and a 7.5−7.7 Å
separation between the two metal sites.
Figure 3. Cyclic voltammograms of L1 and 1−3, 0.2 V s−1, 0.1 M
[N(n-Bu)4]PF6 (L1 and 1) or N(n-Bu)4]OTf (2 and 3). CV of 1 was
obtained in CH2Cl2, all others were obtained in MeCN.
Each CuII site in 3 adopts a distorted square pyramidal
geometry (τ = 0.06−0.12),45 with lengthening of the Cu−Nim
bond (∼2.26 Å) along the Jahn−Teller axis. The bis-
(pyridylmethyl)amino group assumes a pincer-like arrange-
ment, with two short Cu−Npy (∼1.97 Å) bonds and one
slightly longer Cu−Nam (∼2.1 Å) bond. The pyridyl nitrogen of
the α-iminopyridine supplies the fourth ligand of the square
planar base (Cu−N4 plane = 0.214 Å), which constrains the
weakly coordinated axial imine nitrogen at a 77° Npy−Cu−Nim
angle. The two L1 benzene rings of the dicopper complex are
Cr to Ni.17 The lack of reversibility of this process can be
rationalized by the expected changes in coordination geometry
that typically accompany CuI formation. Further reduction of
the dicopper complex is obscured by copper stripping beyond
−0.75 V (Figure S9, Supporting Information).
Solid State Structures. The molecular structures of 1 and
the complex cations [(L1)2Cu2]4+ and [(L1)2Fe2(OTf)2]2+ are
shown in Figures 4−6. The structure of 1, although derived
from less than optimal data completeness (see Experimental
Section and CIF), substantiates the ability of L1 to
Figure 4. Structure of 1 (50% probability ellipsoids). Hydrogens omitted for clarity. Selected bond lengths (Angstroms) and angles (degrees): Zn1−
Cl1 2.1917(5), Zn2−Cl3 2.2499(5), Zn1−Cl2 2.2176(6), Zn2−Cl4 2.2866(5), Zn1−N1 2.0714(16), Zn2−N3 2.624(2), Zn1−N2 2.088(2), Zn2−
N4 2.067(2), Zn2−N5 2.070(2), N1−C5 1.352(3), C5−C6 1.468(3), N2−C6 1.287(3), N1−Zn1−N2 80.77(6), N4−Zn2−N5 99.23(7), Cl1−
Zn1−Cl2 121.46(2), Cl3−Zn2−Cl4 103.56(2), Zn1···Zn2 5.6978(3).
F
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Molecular structure of 2·4MeCN (top) and 2·2MeCN (bottom); 50% probability ellipsoids. Hydrogen atoms and triflate counteranions
have been omitted for clarity. Selected bond lengths (Angstroms) and angles (degrees). 2·4MeCN: Fe−N1 2.177(3), Fe−N2 2.214(3), Fe−N3
2.321(2), Fe−N4 2.141(3), Fe−N5 2.161(3), Fe−O1T 2.103(2), N1−C5 1.351(4), N2−C6 1.273(4), C5−C6 1.460(5), N1−Fe−N2 75.96(9),
N3−Fe−N4 78.64(9), N3−Fe−N5 76.22(9), N4−Fe−N5 91.2(1), N1−Fe−N4 174.36(9), N2−Fe−N5 166.8(1), N3−Fe−O1 162.2(1), Fe···Fe
7.4782(5). 2·2MeCN: Fe−N1 2.170(1), Fe−N2 2.257(1), Fe−N3 2.350(1), Fe−N4 2.163(1), Fe−N5 2.180(1), Fe−O1T 2.113(1), N1−C5
1.353(2), N2−C6 1.286(2), C5−C6 1.463(2), N1−Fe−N2 75.73(4), N3−Fe−N4 77.92(4), N3−Fe−N5 75.40(5), N4−Fe−N5 88.76(5), N1−
Fe−N4 177.46(5), N2−Fe−N5 159.46(5), N3−Fe−O1 165.37(4), Fe···Fe 7.7444(8).
nearly superimposed and within π-stacking distance (3.4−3.7
Å). The Cu···Cu separation consequently is significantly shorter
(∼6.9 Å) than in the diiron complex.
and SQUID magnetometry. These studies provide insight into
the ability of L1 to modulate communication between the two
coordination sites. The zero-field Mossbauer spectrum of 2
̈
One consequence of the structural preferences of the
different metal ion is the orientation of the α-iminopyridine
group with respect to the benzene ring. The α-iminopyridine is
coplanar with the benzene in 1 but rotated by 24° in 2·2MeCN,
52° in 2·4MeCN, and nearly orthogonal (∼85°) in 3. In all
three structures, the C−N and C−C bond distances of the α-
iminopyridine groups are as expected for the fully oxidized form
(Scheme 1). The distances provide a benchmark for establish-
ing ligand-centered reduction in future studies with these
compounds.
exhibits a single quadrupole doublet and establishes the
electronic equivalence of the two metal centers (Figure 7).
The fit of the data affords an isomer shift of δ = 1.12 mm s−1
and a quadrupole splitting of |ΔEQ| = 1.93 mm s−1, typical of
high-spin FeII (S = 2) compounds.46 Magnetic susceptibility
measurements (Figure 8) show a temperature-independent
magnetic moment in the range from 100 to 290 K, decreasing
below 100 K to a value of ∼4.8 μB. The effective magnetic
moment of 7.9 μB at room temperature is higher than the spin-
only value for two uncoupled S = 2 Fe centers (μeff = 6.9 μB for
g = 2) but below the spin-only values of 8.9 μB for an S = 4
ground state arising from two strongly ferromagnetically
Electronic Structure. The electronic structures of 2 and 3
and the nature of the metal−metal interactions in these
complexes were probed by EPR and Mossbauer spectroscopies
̈
G
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 6. Structure of 3 (50% probability ellipsoids). Hydrogen atoms have been omitted, and only two triflate anions are shown for clarity. Selected
bond lengths (Angstroms) and angles (degrees): Cu1−N1A 2.076(3), Cu1−N2A 2.247(3), Cu1−N3 2.106(3), Cu1−N4 1.978(3), Cu1−N5
1.968(3), Cu2−N1 2.042(3), Cu2−N2 2.282(3), Cu2−N3A 2.086(3), Cu2−N4A 1.975(3), Cu2−N5Aavg 1.945(9), N1−C5 1.348(5), C5−C6
1.473(6), N2−C6 1.265(5), N1A−Cu1−N2A 76.7(1), N3−Cu1−N4 83.1(1), N3−Cu1−N5 82.9(1), N1A−Cu1−N3 166.7(1), N4−Cu1−N5
163.1(1), N1−Cu2−N2 76.9(1), N3A−Cu2−N4A 83.4(1), N3A−Cu2−N5A 85.5(4), N1−Cu2−N3A 166.3(1), N4A−Cu2−N5Aavg 161.9(6),
Cu1···Cu2 6.9283(8).
Table 1. Crystallographic Data for 1−3
a
1
2·4MeCN
2·2MeCN
3
chem. formula
fw
C24H21Cl4N5Zn2
652.00
C62H57F12Fe2N15O12S4
1672.17
C56H48F12Fe2N12O12S4
1549.00
C52H42Cu2F12N10O12S4
1482.28
cryst syst
space group
a (Å)
triclinic
triclinic
triclinic
monoclinic
C2/c
P−1
P−1
P−1
7.1827(2)
9.4410(2)
20.2740(4)
77.8560(10)
87.9490(10)
72.3700(10)
1280.35(5)
2
10.6568(2)
12.9150(2)
15.0100(3)
106.9800(10)
109.4480(10)
97.5000(10)
1802.68(6)
1
8.9929(11)
12.9736(18)
14.7541(16)
69.606(4)
86.687(5)
76.892(4)
1571.0(3)
1
42.0617(8)
10.7977(2)
28.7341(5)
90.00
b (Å)
c (Å)
α (deg)
β (deg)
97.1390(10)
90.00
γ (deg)
V (Å3)
12949.0(4)
8
Z
cryst size (mm3)
ρcalcd (mg/mm3)
0.13 × 0.07 × 0.05
1.691
0.12 × 0.11 × 0.07
1.540
0.4 × 0.23 × 0.18
1.637
0.16 × 0.12 × 0.06
1.521
μ (mm−1
)
6.338
5.198
0.702
2.896
Rint
0.0357
0.0393
0.0228
0.0523
data/restraints/params
R1 [I ≥ 2σ(I)]
4179/0/316
0.0241
5831/23/523
0.0526
9589/22/483
0.0388
10141/22/849
0.0591
wR2 [I ≥ 2σ(I)]
R1 [all data]
0.0638
0.1550
0.1012
0.1730
0.0247
0.0555
0.0451
0.0686
wR2 [all data]
0.0645
0.1591
0.1061
0.1798
goodness-of-fit on F2
1.058
1.092
1.069
1.088
Δρmin,max (e Å−3
)
−0.31, 0.34
−0.742, 0.591
−0.634, 1.023
−0.919, 1.424
a
b
MeCN > 4, but could not be determined precisely due to high mobility of the final solvent molecule. Observation criterion: I > 2σ(I). R1 = Σ||Fo|
c
d
− |Fc||/Σ|Fo|. wR2 = [Σ[w(Fo − Fc2)2]/Σ[w(Fo )2]]1/2, where w = 1/σ2(Fo ) + (aP)2 + bP, P = (Fo + 2Fc2)/3. GoF = [Σ[w(Fo − Fc2)2]/(n −
2
2
2
2
2
p)]1/2
.
coupled high-spin FeII ions. Data were fit according to the
following spin Hamiltonian
assess the relative contribution of each parameter and yielded
values for J = 0−0.2 cm−1 and |D| = 14−22 cm−1.47 The best fit
was obtained with g = 2.2, J = 0.04 cm−1, |D| = 19.7 cm−1, and
E/D = 0.26, denoting negligible interaction between the two
iron centers in 2. The magnitude of these parameters is
confirmed by the absence of any EPR signal at Q-band
frequency (∼34 GHz). The large single ion contribution to the
zero-field splitting and its pronounced rhombicity shifts the
̂
̂
̂
̂
Η = −2JS1·S2 + gβ(S1 + S2)·B
⎡
⎤
2
z,i
1
3
E
D
2
2
y,i
̂
̂
̂
+
∑
D S
−
Si(Si + 1) + (Sx,i − S )
⎢
⎥
⎦
⎣
i=1,2
including exchange coupling and zero-field splitting terms, with
S1 = S2 = 2. Fits with either J or D fixed at zero were included to
H
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
absolute value. The X-band EPR spectrum recorded in MeCN/
CH2Cl2 displayed a signal consistent with the interpretation of
the magnetic data, that is, two weakly coupled CuII (S = 1/2)
ions (Figure 9). The spectral profile is dominated by a large g
Figure 7. Zero-field Mossbauer spectrum of 2 at 80 K, where the open
̈
circles depict experimental data and the red line the fit (δ = 1.12 mm
s−1, |ΔEQ| = 1.93 mm s−1).
Figure 9. X-band EPR spectrum of 3 recorded in MeCN/CH2Cl2 at
10 K (experimental conditions: frequency, 9.4740 GHz; modulation,
0.2 mT; power, 0.2 mW). Experimental spectrum shown in black, and
simulation depicted by the red trace: g = 2.081, 2.088, 2.196; A = 0, 40,
156 × 10−4 cm−1; J = −0.2 cm−1 (fixed); dCu···Cu = 6.9 Å; χ = 34°; ρ =
7°. (Inset) Experimental and simulated EPR spectra in the half-field
region (conditions: frequency, 9.4741 GHz; modulation, 1.0 mT;
power, 63 mW).
splitting synonymous with CuII; plus addition of the minuscule
exchange coupling gives rise to a poorly resolved 7-line
hyperfine pattern in the g|| region characteristic of coupled
63,65Cu (I = 3/2, 100% abundant) nuclei. This pattern is more
clearly displayed at K-band frequency (∼24 GHz) where the
enhanced Zeeman interaction separates the parallel component
centered on g|| = 2.196 from the broad perpendicular feature at
g⊥ ≈ 2.08 that is devoid of hyperfine splitting (Figure S11,
Supporting Information). The inset in Figure 9 shows a half-
field feature arising from the forbidden “ΔMS = 2” transition of
the spin triplet (S = 1) formed by coupling of the two CuII ions.
The 7-line pattern is poorly resolved as its intensity is
proportional to the magnitude of D,48,49 and therefore,
substantially higher power (63 mW) was required to expose
this feature.
Simulation of both the X- and the K-band spectrum was
achieved with the same spin Hamiltonian parameters: g =
(2.081, 2.088, 2.196) and A = (0, 40, 156) × 10−4 cm−1. The
spin−spin interaction was included in the simulation using a J
interaction matrix, which incorporates both exchange and
dipolar coupling elements. The best fit was achieved for J =
−0.2 cm−1 (fixed from magnetic susceptibility) for an interspin
distance (r) of 6.9 Å retrieved from the crystal structure and is
consistent with very weakly coupled CuII ions. Inclusion of two
Euler angles, χ = 34° and ρ = 7°, defined as the relative
orientation of the local g axes which are mostly determined by
the orientation of the magnetic orbital of the second CuII site
with respect to the first and enabled the experimental hyperfine
line shape to be successfully modeled. Simulation was less
sensitive to the magnitude of ρ, and the χ angle is close to the
Figure 8. Temperature dependence of the magnetic moment μeff, μB,
of powdered samples of 2 and 3. Filled circles are experimental data;
blue line represents the best fit. For 2: S1 = S2 = 2, g1 = g2= 2.21, J =
0.043 cm−1, |D1| = |D2| = 19.7 cm−1, E/D1 = E/D2 = 0.26, TIP = 0.027
emu, including a 12% impurity with S = 2. For 3: S1 = S2 = 1/2, g1 =
g2= 2.12 (fixed), J = −0.241 cm−1, TIP = 50 × 1e−6 emu.
transitions to resonant field positions beyond the available
magnetic field (1.8 T).
The magnetic susceptibility data for 3 shows weak
ferromagnetic coupling between the two CuII centers (Figure
8). The temperature-independent (20−290 K) effective
magnetic moment of ∼2.6 μB matches the expected value for
two uncoupled CuII centers with the hallmark g ≈ 2.1. Data
were fit for g = 2.12 and J = −0.2 cm−1. This very small
exchange coupling precludes assignment of the total spin
ground state in 3, and therefore, J should be treated as an
I
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
crystallographically defined dihedral angle (∼40°) between the
CuN4 basal planes orthogonal to the Jahn−Teller axis. The
zero-field splitting parameter, D, can be estimated from a two-
2 −3
point dipole model, D = −(3/2)g2μB r ≈ −0.005 cm−1, a
value consistent with the extremely weak half-field transition
(Figure 9, inset).
Ferromagnetic interactions between the metal ions in 2 and
3 might be presupposed, based on previous studies with
bimetallic phenylenediamine systems.50,51 Ferromagnetic cou-
pling is expected for systems containing an odd number of
bridging atoms between two paramagnetic centers (i.e., with
meta-substituted benzene linkers), while an even number of
bridging atoms promotes antiferromagnetic interactions.
Spectroscopic, magnetic, and DFT computational studies
demonstrated that the coupling is orchestrated by spin
polarization through the aromatic bridge. The magnitude of
the exchange interactions in the CuII 1,3-[bis(2-pyridylmethyl)-
amino]benzene complexes was further shown to be dependent
on the orientation of the metal SOMOs with respect to the
phenylenediamine unit.50
DFT calculations on 3, likewise, illustrate spin polarization of
the phenylenediamine system of L1. The spin density map
depicts an alternating pattern on the carbon atoms of the
central aromatic ring, with delocalization of the metal spin
toward the amino nitrogen of the bis(pyridylmethyl) group
(Figure 10). The SOMOs of 3 (Figure 11) consist of the in-
Figure 11. DFT-derived (B3LYP) SOMOs for 3.
between the two copper centers.52 Further inspection of the
remaining metal-based MOs of 3 also reveals pervasive
covalency, with contributions from the ligand π system and
admixture of the copper d orbitals. Spin−orbit coupling
contributions, therefore, might also account for the weak
coupling in 3.
The iron centers in 2 are essentially uncoupled. Several
factors impede strong communication between the iron sites, as
illustrated by both the molecular structure and the DFT
calculations.53 The metal separation (7.5−7.7 Å) is substantially
greater than that in 3. The α-iminopyridine is canted at 24°
(2′) or 52° (2) with respect to the plane of the benzene bridge,
which results in poor overlap with the aromatic π system. The
DFT-derived spin density map for 2 (Figure 12) depicts
Figure 10. DFT-derived (B3LYP) spin density plot for 3 based on
Lowdin population analysis.
̈
2
2
phase and out-of-phase σ* orbitals, comprising ∼34% Cu dx −y
character, directed toward the Nam p orbital, with minor
contributions from the π system of the benzene ring. However,
due to the orientation of the α-iminopyridine ligands, the Nim p
2
2
orbitals lie orthogonal to the dx −y -based σ* orbitals, severing
the conventional pathway for π-type exchange through the
Nim−C−C−C−Nam bridge. Efficient π-type overlap of the Cu
SOMOs with both N p orbitals in the related 1,3-phenylenedi-
amine bridged cupric metallocyclophane (Cu···Cu = 6.8 Å)51
and 1,3-[bis(2-pyridylmethyl)amino]benzene complexes
(Cu···Cu = 5.9 Å)50 led to substantial coupling of the two
metal centers, with measured exchange constants of 8.4 and 4.6
cm−1, respectively, obtained from magnetic susceptibility
measurements. Vanishing interactions between the metal ions
in 3 can thus be attributed to the lack of a direct π-type
Figure 12. DFT-derived (B3LYP) spin density plot for 2 based on
Lowdin population analysis.
̈
negligible spin density on the phenylenediamine unit, and in
contrast to 3, all SOMOs are strongly metal based. Overlap
2
between the Fe dz orbitals and the extended Nim−C−C−C−
Nam π system (Figure 13) provides a possible exchange
pathway to connect the two Cu dx −y orbitals through one L1
molecule. Interactions between the stacked benzene rings of L1
might provide an alternate pathway for communication
2
2
pathway between the iron centers. However, the benzene π
2
system contributes minimally to these dz -based SOMOs.
J
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
2
Figure 13. DFT-derived (B3LYP) Fe dz -based SOMOs for 2.
CONCLUSIONS
ACKNOWLEDGMENTS
■
■
The new ligand, L1, provides a platform for bimetallic
complexes, with multiple redox sites furnished by the terminal
α-iminopyridine moiety. The high flexibility of the ligand
accommodates the coordination geometry preferences of
various metal ions and formation of both symmetric and
asymmetric binuclear complexes. Electrochemical studies
confirm the redox activity of L1 and provide an indication of
the nature of the reductive processes available to the bimetallic
compounds. The redox behavior of 1−3 appears to follow the
anticipated trend for the series of metals examined in this study.
One-electron reduction of the ligand is observed for 1, in which
L1 is coordinated to redox-inert ZnII, whereas reduction of the
dimeric copper complex, 3, likely involves the CuII/I couple. A
second reduction event only is observed in the CV of 2, which
is the most intriguing of the three metal compounds that we
isolated. Compound 2 offers the option of both metal- and
ligand-centered redox processes; the ensuing spin coupling can
lead to varied electronic structures.
Both 2 and 3 are paramagnetic. The coordination of the
triflate ions in 2 supports a high-spin configuration at each FeII
center. While studies have shown that 1,3-phenylenediamine
bridges in related binuclear compounds can provide an efficient
route for exchange interactions between two metal centers, the
metal ions in both 2 and 3 are virtually uncoupled. Inspection
of the molecular structures and calculated data reveals that the
communication pathway between the metal sites is dismantled
by the orientation of the phenylenediamine groups and the
coordination geometry adopted by the metal. As a result, L1
may provide a means to localize charge upon reduction of the
compounds. Efforts are currently underway to isolate and
characterize the reduced forms of the L1−metal complexes.
We thank the EPSRC, the University of Durham, and the
Durham Energy Institute for financial support. We are grateful
to Mr. Bernd Mienert (MPI) for technical assistance with
Mossbauer and SQUID measurements. We thank Dr. Alan
̈
Kenwright for helpful discussions concerning NMR experi-
ments and Prof. Andy Beeby for the use of his instrument for
diffuse reflectance spectra.
REFERENCES
(1) Chirik, P. J. Inorg. Chem. 2011, 50, 9737.
(2) Sproules, S.; Wieghardt, K. Coord. Chem. Rev. 2011, 255, 837.
(3) Kaim, W. Eur. J. Inorg. Chem. 2012, 343.
(4) Hindson, K.; de Bruin, B. Eur. J. Inorg. Chem. 2012, 340.
(5) Haneline, M.; Heyduk, A. F. J. Am. Chem. Soc. 2006, 128, 8410.
(6) Sylvester, K. T.; Chirik, P. J. J. Am. Chem. Soc. 2009, 131, 8772.
(7) Tondreau, A. M.; Milsmann, C.; Patrick, A. D.; Hoyt, H. M.;
Lobkovsky, E.; Wieghardt, K.; Chirik, P. J. J. Am. Chem. Soc. 2010, 132,
15046.
(8) Kraft, S. J.; Williams, U. J.; Daly, S. R.; Schelter, E. J.; Kozimor, S.
A.; Boland, K. S.; Kikkawa, J. M.; Forrest, W. P.; Christensen, C. N.;
Schwarz, D. E.; Fanwick, P. E.; Clark, D. L.; Conradson, S. D.; Bart, S.
C. Inorg. Chem. 2011, 50, 9838.
■
(9) Ouch, K.; Mashuta, M. S.; Grapperhaus, C. A. Inorg. Chem. 2011,
50, 9904.
(10) Nguyen, A. I.; Zarkesh, R. A.; Lacy, D. C.; Thorson, M. K.;
Heyduk, A. F. Chem. Sci. 2011, 2, 166.
(11) Kaim, W. Inorg. Chem. 2011, 50, 9752.
(12) Nawn, G.; Waldie, K. M.; Oakley, S. R.; Peters, B. D.; Mandel,
D.; Patrick, B. O.; McDonald, R.; Hicks, R. G. Inorg. Chem. 2011, 50,
9826.
(13) Arumugam, K.; Shaw, M. C.; Chandrasekaran, P.; Villagran, D.;
Gray, T. G.; Mague, J. T.; Donahue, J. P. Inorg. Chem. 2009, 48, 10591.
(14) Hess, C. R.; Weyhermuller, T.; Bill, E.; Wieghardt, T. Angew.
̈
Chem., Int. Ed. 2009, 48, 3703.
(15) Szigethy, G.; Heyduk, A. F. Inorg. Chem. 2011, 50, 125.
(16) Zarkesh, R. A.; Ziller, J. W.; Heyduk, A. F. Angew. Chem., Int. Ed.
2008, 47, 4715.
(17) Lu, C. C.; Bill, E.; Weyhermuller, T.; Bothe, E.; Wieghardt, K. J.
Am. Chem. Soc. 2008, 130, 3181.
ASSOCIATED CONTENT
■
̈
S
* Supporting Information
Crystallographic data files (CIF format), additional electro-
chemical, spectroscopic, and computational data, and NMR
spectra of new compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
(18) Parkin, G. J. Chem. Educ. 2006, 83, 791.
(19) Hah, J.-M.; Martas
Chem. 2003, 46, 1661.
́
ek, P.; Roman, L. J.; Silverman, R. B. J. Med.
(20) Hanson, G. R.; Gates, K. E.; Noble, C. J.; Griffin, M.; Mitchell,
A.; Benson, S. J. Inorg. Biochem. 2004, 98, 903.
(21) Cosier, J.; Glazer, A. M. J. Appl. Crystallogr. 1986, 19, 105.
(22) APEX2, X-ray software suite, version 2012.4-3; Bruker Analytical
X-ray Instruments Inc.: Madison, WI, 2012.
AUTHOR INFORMATION
■
Corresponding Author
(23) SAINT, Data Reduction Software, version 8.18c; Bruker
Analytical X-ray Instruments Inc., Madison, WI, 2012.
(24) Dolomonov, O.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. J. Appl. Crystallogr. 2009, 42, 339.
*E-mail: c.r.hess@durham.ac.uk.
Notes
The authors declare no competing financial interest.
K
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(25) Sheldrick, G. M. J. Acta Crystallogr., Sect A: Found. Crystallogr.
2008, 64, 112.
(53) DFT calculations were carried out on both solvatomorphs of 2
and yielded identical results (see Supporting Information).
(26) Neese, F. An Ab initio, DFT and Semiempirical SCF-MO Package,
Version 2.9; Max Planck Institute for Bioinorganic Chemistry,
Mulheim an der Ruhr, Germany, Jan 2012.
̈
(27) Becke, A. D. J. Chem. Phys. 1986, 84, 4524.
(28) Becke, A. D. J. Chem. Phys. 1993, 98, 5648.
(29) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785.
(30) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571.
̈
(31) Schafer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
̈
5829.
(32) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
̈
Chem. Phys. Lett. 1995, 240, 283.
̈
(33) Eichkorn, K.; Treutler, O.; Ohm, H.; Haser, M.; Ahlrichs, R.
̈
Chem. Phys. Lett. 1995, 242, 652.
(34) GaussView 4.1.2; Gaussian, Inc.: Wallington, CT, 2006.
(35) Hess, C. R.; Weyhermuller, T.; Bill, E.; Wieghardt, T. Inorg.
̈
Chem. 2010, 49, 5686.
(36) Ray, K.; Petrenko, T.; Wieghardt, K.; Neese, F. Dalton Trans.
2007, 1552.
(37) Ward, A. L.; Elbaz, L.; Kerr, J. B.; Arnold, J. Inorg. Chem. 2012,
51, 4694.
(38) England, J.; Gondhia, R.; Bigorra-Lopez, L.; Peterson, A. R.;
White, A. J. P.; Britovsek, G. J. P. Dalton Trans. 2009, 5319.
(39) Borzel, H.; Comba, P.; Hagen, K. S.; Lampeka, Y. D.; Lienke, A.;
̈
Linti, G.; Merz, M.; Pritkow, H.; Tsymbal, L. V. Inorg. Chim. Acta
2002, 337, 407.
(40) Schindler, S.; Szalda, D. J.; Creutz, C. Inorg. Chem. 1992, 31,
2255.
(41) Murthy, N. N.; Mahroof-Tahir, M.; Karlin, K. D. Inorg. Chem.
2001, 40, 628.
(42) Compton, R. G.; Banks, C. E. Understanding Voltammetry; 2nd
ed.; Imperial College Press: London, 2011.
(43) Muresan, N.; Lu, C. C.; Ghosh, M.; Peters, J. P.; Abe, M.;
Henling, L. M.; Weyhermuller, T.; Bill, E.; Wieghardt, K. Inorg. Chem.
̈
2008, 47, 4579.
(44) Foxon, S.; Xu, J.-Y.; Turba, S.; Leibold, M.; Hampel, F.;
Heinemann, F. W.; Walter, O.; Wurtele, C.; Holthausen, M.;
̈
Schindler, S. Eur. J. Inorg. Chem. 2007, 429.
(45) τ = 0 represents a perfect square pyramid, τ = 1 corresponds to
perfectly trigonal bipyramidal geometry. See: Addison, A. W.; Rao, T.
N.; Reedijk, J.; van Rijn, J.; Verschoor, G. C. J. Chem. Soc., Dalton
Trans. 1984, 1349.
(46) Greenwood, N. N.; Gibb, T. C. Mossbauer Spectroscopy;
̈
Chapman and Hall Ltd.: London, 1971.
(47) A reasonable fit of the data could only be obtained by including
a TIP value of 0.027 emu, which is unlikely to correspond to
temperature-independent paramagnetism of the diiron complex. The
Mossbauer spectrum of the bulk solid (Figure S10, Supporting
̈
Information) shows the presence of a second iron-containing species
in the powder samples. The presence of the secondary component has
been taken into account in the fit of the magnetic data (12%, S = 2)
but does not account for the large TIP values. Although we do not
know the origin, the TIP-like background magnetism suggests the
presence of magnetic nanoparticles that constitute a minor component
of bulk solid samples of 2.
(48) Eaton, S. S.; More, K. M.; Sawant, B. M.; Eaton, G. R. J. Am.
Chem. Soc. 1983, 105, 6560.
(49) Eaton, S. S.; Eaton, G. R. Coord. Chem. Rev. 1988, 83, 29.
(50) Foxon, S. P.; Torres, G. R.; Walter, O.; Pederson, J. Z.;
Toftlund, H.; Huber, M.; Falk, K.; Haase, W.; Cano, J.; Lloret, F.;
̈
Julve, M.; Schindler, S. Eur. J. Inorg. Chem. 2004, 335.
(51) Fernan
́
dez, I.; Ruiz, R.; Faus, J.; Julve, M.; Lloret, F.; Cano, J.;
Ottenwaelder, X.; Journaux, Y.; Munoz, M. C. Angew. Chem., Int. Ed.
2001, 40, 3039.
̃
(52) Martín, S.; Grace, I.; Bryce, M. R.; Wang, C.; Jitchati, R.;
Batsanov, A. S.; Higgins, S. J.; Lambert, C. J.; Nichols, R. J. J. Am.
Chem. Soc. 2010, 132, 9157.
L
dx.doi.org/10.1021/ic302087f | Inorg. Chem. XXXX, XXX, XXX−XXX