Hydrazine-mediated strongly coupled Re(CO)3 dimers

Article abstract of DOI:10.1039/c5dt02821c

Dimeric metal complexes can often exhibit coupling interactions via bridging ligands. In this report, we present two Re(CO)₃ dimers, where the metals are linked via a bis(pyca) hydrazine (pyca = pyridine-2-carbaldehyde imine) Schiff base ligand. For the dimeric compounds 4 and 5, we observe strong coupling across the dimer as measured by cyclic voltammetry: ～480 mV separations between the first and the second reduction waves that correspond to comproportionation constants close to 1.5 × 10⁸. Evidence for a mixed valence state upon one electron reduction was also observed by spectroelectrochemistry in which a clear inter-valence charge-transfer (IVCT) band was observed in [4]- and [5]-complexes. The electronic structures of all target compounds were probed by DFT and TDDFT computational methods. DFT calculations indicate that reduction takes place at the diimine units, and that the observed coupling is a ligand-based phenomenon, rather than one that involves metal-based orbitals.

Full text of DOI:10.1039/c5dt02821c

Dalton
Transactions
View Article Online
View Journal
PAPER
Hydrazine-mediated strongly coupled Re(CO)3
dimers†
Cite this: DOI: 10.1039/c5dt02821c
A. Hasheminasab,a H. M. Rhoda,b L. A. Crandall,a J. T. Ayers,c V. N. Nemykin,*b
R. S. Herrick*c and C. J. Ziegler*a
Dimeric metal complexes can often exhibit coupling interactions via bridging ligands. In this report, we
present two Re(CO)3 dimers, where the metals are linked via a bis(pyca) hydrazine (pyca = pyridine-2-
carbaldehyde imine) Schiff base ligand. For the dimeric compounds 4 and 5, we observe strong coupling
across the dimer as measured by cyclic voltammetry: ∼480 mV separations between the first and the
second reduction waves that correspond to comproportionation constants close to 1.5 × 108. Evidence
for a mixed valence state upon one electron reduction was also observed by spectroelectrochemistry in
which a clear inter-valence charge-transfer (IVCT) band was observed in [4]- and [5]-complexes. The
electronic structures of all target compounds were probed by DFT and TDDFT computational methods.
DFT calculations indicate that reduction takes place at the diimine units, and that the observed coupling is
a ligand-based phenomenon, rather than one that involves metal-based orbitals.
Received 23rd July 2015,
Accepted 2nd September 2015
DOI: 10.1039/c5dt02821c
www.rsc.org/dalton
Re(CO)3 diimines can be used to initiate chemical reactions or
to induce electron transfer processes. Coupled photochemi-
Introduction
Dimeric Re(CO)3 compounds have been attracting interest as cally-active transition metal compounds can be used as com-
synthetic targets, both as components of materials and for ponents of materials; these have been employed with several
possible biological applications.1–7 There are many examples d6 metal ion systems. Although the electronic interactions of
of Re(CO)3 dimers where the two metal ions are bridged by metals such as ruthenium have been extensively explored in
halides, hydroxides or alkoxides, organic fragments (such as a dimeric complexes, similar experiments have not been carried
phenyl ring) or by nitrogenous ligands. A variety of experi- out on Re(CO)3 diimine systems. In this report, we present the
mental and theoretical studies have been carried out to probe synthesis and characterization of two Re(CO)3 dimeric com-
the effects of dimer structure on physical properties, such as pounds that exhibit an extraordinary degree of coupling across
photophysics or electron transfer.3,8–23 Under certain con- the dimers. The coupling can be readily observed via cyclic vol-
ditions, Re(CO)3 complex centers can couple through a brid- tammetry and spectroelectrochemistry, which show a large
ging ligand. Parameters including bridge type, metal–metal separation between sequential reduction waves and the pres-
distance, interplanar distances, degree of saturation, conju- ence of an apparent inter-valence charge transfer (IVCT) band,
gation, and orientation of the bridge moiety can all affect the respectively. We also present several monomeric analogues
degree of coupling and their resultant properties such as redox and monometalated compounds which we can use as controls
potentials.7,12,14,15,18,24–27
to study the specific effects of metal dimer structures on elec-
Rhenium(I) diimines have been studied for their photo- tronic properties. DFT and TDDFT calculations reveal that the
chemical activities, due to the presence of metal-to-ligand reductions are primarily ligand-based rather than metal-cen-
charge transfer bands in these compounds. Irradiation of tered, which is in agreement with the literature on the
reduction of monomeric Re(CO)3 bipyridine complexes.
aDepartment of Chemistry, University of Akron, OH 44325-3601, USA.
E-mail: ziegler@uakron.edu
bDepartment of Chemistry & Biochemistry, University of Minnesota Duluth, Duluth,
Experimental
General information
MN 55812, USA. E-mail: vnemykin@d.umn.edu
cDepartment of Chemistry, College of the Holy Cross, Box C, Worcester, MA 01610-
2395, USA. E-mail: rherrick@holycross.edu
†Electronic supplementary information (ESI) available: Spectroscopic results,
Materials and methods. Reagents were purchased from
Strem, Acros Organics, TCI America or Sigma-Aldrich and used
as received without further purification. Syntheses were per-
formed to minimize exposure to air or water. All solvents were
electrochemical experiments, X-ray structural data and calculations. CCDC
1414734–1414738 for 3–5, 8, 9. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c5dt02821c
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
dried and deoxygenated by alumina and copper columns in 0.05 µm alumina. The electrode was rinsed with ethanol and
the Pure Solv solvent system (Innovative Technologies, Inc.); deionized water after each polishing and wiped with a
dried solvents were stored over molecular sieves.
Kimwipe. The non-aqueous Ag/Ag+ reference electrode was pre-
NMR spectra were recorded on Varian Mercury 300 MHz pared by soaking the silver wire in the degassed DMF solution
and 500 MHz instruments. Chemical shifts were reported with of 0.01 M AgClO4: 0.1 M TBAPF6. Approximated formal poten-
respect to residual solvent peaks as internal standard (1H: d6- tials for a reversible redox couples (E°′ = (Epc + Epa)/2) were all
DMSO, δ = 2.50 ppm; 13C: d6-DMSO, δ = 39.7 ppm). Infrared referenced against Ag/Ag+. ΔEp = Epa − Epc, ΔE = (E1°′ − E2°′)
spectra were collected on Thermo Scientific Nicolet iS5 which and peak currents were calculated by use of an extrapolated
was equipped with iD5 ATR. Electronic absorption spectra were baseline current28,29 Potentials were reported against Ag/AgCl
recorded on Hitachi U-2000 UV-vis spectrophotometer. Elemental reference electrode and compared with previous data reported
Analyses were performed by Atlantic Microlab of Norcross, GA by Geiger and coworker.30 The linear dependence between the
30091. Mass Spectrometric analyses were carried out at the Mass square root of the scan rate (ν1/2) and both the cathodic (ipc)
Spectrometry center at The University of Akron in Akron, OH.
and anodic (ipa) currents was examined between 0.1 to 0.07 V
X-ray data collection and structure determination. X-ray s−1, and the formal potentials are independent of scan rate. At
intensity data for compounds 3 and 9 were measured at 100 K a 0.10 V s−1 sweep rate, the FcH/FcH+ occurs at 0.48 0.05 V s−1
on a CCD-based X-ray diffractometer system equipped with a (ΔEp = 71 mV; ipa/ipc = 1.02).
Mo-target X-ray tube (Mo Kα radiation, λ = 0.71073 Å) operated
Spectroelectrochemistry. Spectroelectrochemical data were
at 2000 W power. Data for 4, 5 and 8 were collected on a CCD- collected on a JASCO-720 spectrophotometer at room tempera-
based diffractometer with dual Cu/Mo ImuS microfocus optics ture. The experiments were conducted using a CHI-620C elec-
(Cu Kα radiation, λ = 1.54178 Å). Crystals were mounted on a trochemical analyzer using a custom-made 1 mm cell with a
cryoloop using Paratone oil and placed under a steam of nitro- platinum mesh-working electrode. In order to suppress over-
gen at 100 K. The detector was placed at a distance of 5.009 cm tones in the NIR region of the optical spectra, measurements
from the crystal. Crystals data and structure refinement para- were conducted in 0.3 M TBAP in DMF.
meters are summarized in Table 1.
DFT and TDDFT calculations. The starting geometries of all
Cyclic voltammetry. Cyclic voltammograms were obtained compounds were taken from X-ray structures. They were opti-
using a standard three electrode cell and Electrochemical ana- mized using the B3LYP exchange–correlation functional31,32
lyzer BAS 100B from Bioanalytical systems and were recorded coupled with the relativistic DZP basis set for the Re atom33
at 298 K under the following conditions: 10−3 M samples in di- and the 6-311G(d)34 basis set for the remaining atoms. Energy
methylformamide (DMF) in the presence of 0.1 M tetrabutyl- minima in optimized geometries were confirmed by frequency
ammonium
hexafluorophosphate
(TBAPF6)
or calculations. DMF was used as a solvent in all of the single
tetrabutylammonium perchlorate (TBAP) as a supporting elec- point DFT-PCM and TDDFT-PCM calculations; solvent effects
trolyte, Ag/Ag+ reference electrode, 0.79 mm2 gold working were calculation using the polarized continuum model
electrode, and platinum wire auxiliary electrode. The working (PCM).35 The first 30 states of each compound were calculated
electrode was polished first with 3 µm fine diamond, then in all TDDFT-PCM calculations. All DFT calculations were con-
Table 1 X-ray crystal data and structure parameters for compounds 3–5, 8 and 9
Compound
3
4
5
8
9
Empirical formula
Formula weight
Crystal system
Space group
C15H10BrN4O3Re
560.38
C18H10Cl2N4O6Re2
821.60
Monoclinic
P2/c
C18H10Br2N4O6Re2
910.52
C9H7BrClN3O3Re
426.83
Monoclinic
P2(1)
C9H7BrClN3O3Re
471.29
Triclinic
Triclinic
Triclinic
ˉ
ˉ
ˉ
P1
P1
P1
a/Å
b/Å
c/Å
α(°)
7.036(3)
9.082(4)
7.036(3)
81.144(4)
84.259(4)
72.527(4)
801.9(6)
2
14.4660(9)
11.4063(8)
12.9730(10)
90
96.727(5)
90
2125.9(3)
4
2.567
11.4508(7)
11.8036(7)
11.8641(7)
75.213(4)
67.040(3)
65.574(4)
2084.2(2)
2
6.4068(3)
8.8341(5)
14.4493(8)
90
101.947(3)
90
800.09(7)
2
1.772
6.477(3)
8.265(3)
12.881(5)
95.607(4)
93.095(4)
97.211(4)
679.3(5)
2
β(°)
γ(°)
Volume (Å3)
Z
Dc (Mg m−3
)
2.321
10.090
524
2.264
21.280
828
2.304
11.885
432
µ (mm−1
F(000)
)
24.653
1512
16.419
396
Reflns collected
Indep. reflns
5781
2810
1.033
0.0201
0.0512
0.0209
0.0517
13 976
3266
1.016
0.0577
0.1507
0.0703
0.1630
4003
4003
1.042
0.0468
0.1253
0.0537
0.1303
4931
2043
1.165
0.0638
0.1864
0.0642
0.1866
4937
2677
1.115
0.0415
0.1164
0.0457
0.1206
GOF on F2
R1 (on Fo2, I > 2σ(I))
wR2 (on Fo2, I > 2σ(I))
R1 (all data)
wR2 (all data)
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
ducted using the Gaussian 09 software package,36 and the 131.0 ppm. CHN Anal. Calc. for Re2C18H10O6N4Cl2: C, 26.31;
QMForge program37 was used for the molecular orbital analysis. H, 1.22; N, 6.82. Found: C, 26.25; H, 1.25; N, 6.72. IR (CO
stretch, cm−1): 2020, 1885. MS (ESI): m/z
Re2C18H10O6N4Cl2Na: 844.90 found 844.9. UV-Vis spectrum in
= calc. for
Synthetic procedures
Synthesis of 1,2-pyridinecarboxaldehydeazine, (pyca–pyca, CH2Cl2 λmax 293 nm (ε = 2.0 × 104 M−1 cm−1) and λmax 458 nm
pyca = pyridine-2-carbaldehyde imine). This compound was (ε = 6.0 × 103 M−1 cm−1).
prepared as previously described.38
5: Yield 42 mg (75%). 1H NMR (300 MHz, d6-DMSO,
Synthesis of 2 and 3 [Re(CO)3(pyca–pyca)X; 2, X = Cl; 3 = δ (ppm)): 9.57 (d, 2H, NvCH), 9.1 (d, J = 6.0 Hz, 2H, H on py),
Br]. The procedure for the synthesis of 2 is representative for 8.74 (d, J = 6.0 Hz, 2H, H on py), 8.42 (m, 2H, H on py), 7.94
both compounds. 2: Re(CO)5Cl (0.050 g, 0.141 mmol) and 1 (m, 2H on py).13C NMR (75 MHz, d6-DMSO, δ (ppm)): 195.9,
(0.030 g, 0.141 mmol) were refluxed in 25 mL of methanol for 195.7, 185.4, 167.2, 165.2, 154.2, 153.9, 151.6, 151.4, 140.8,
4 h. After cooling to room temperature, a red solid was filtered 140.6, 132.2, 132.0, 130.0. CHN Anal. Calc. for
and washed with 10 mL hexane. Crystals suitable for X-ray Re2C18H10O6N4Br2: C, 23.74; H, 1.11; N, 6.15. Found: C, 24.00;
diffraction were prepared by slow vapor diffusion of water into H, 1.09; N, 6.03. IR (CO stretch, cm−1): 2014, 1913, and 1880.
1
a DMSO solution. Yield: 52 mg (71%). H NMR (500 MHz, d6- MS (ESI): m/z = calc. for Re2C18H10O6N4Br2Na: 932.79 found
DMSO, δ (ppm)): 9.40 (s, 1H, metal-bound NvCH), 9.07 (d, J = 932.8. UV-Vis spectrum in CH2Cl2 λmax 289 nm (ε = 2.0 × 104
4.5 Hz, 1H, H on metal-bound py), 8.86 (s, 1H, NvCH), 8.81 M−1 cm−1) and λmax 470 nm (ε = 5.4 × 103 M−1 cm−1).
(d, J = 5.0 Hz 1H, H on metal-bound py), 8.35 (m, 2H, on metal
Synthesis of 8 and 9 [Re(CO3)X(pyca-NH2); 8, X = Cl; 9 =
bound and free py), 8.20 (d, J = 8.0 Hz, 1H, on free py), 8.05 Br]. The procedure for the synthesis of 8 is representative for
(m, 1H, H on metal bound py), 7.84 (m, 1H, H on metal-bound both compounds.
py), 7.65 (m, 1H, H on metal-bound py) ppm. 13C NMR
8: Re(CO)5Cl (0.050 g, 0.141 mmol) and pyridine-2-carboxal-
(75 MHz d6-DMSO, δ (ppm)) 196.8, 186.0, 165.0, 163.4, 153.5, dehyde (0.0151 g, 0.141 mmol) were refluxed in 25 mL of
153.4, 150.5, 150.0, 140.6, 137.7, 130.1, 129.7, 127.0, hexane for 6 h. A red solid (6) was filtered and washed with
122.6 ppm. CHN Anal. Calc. for Re1C15H10O3N4Cl1: C, 34.92; diethyl ether. The solid was then dissolved in THF and hydra-
H, 1.95; N, 10.85. Found: C, 35.11; H, 2.06; N, 10.69. IR (CO zine monohydrate (0.0070 g, 0.141 mmol) was added under
stretch, cm−1): 2020, 1888. MS (ESI): m/z
= calc. for vigorous stirring. The yellow solid was filtered and washed
Re1C15H10O3N4Cl1Na: 538.98 found 539.0. UV-Vis spectrum in with 25 mL diethyl ether. Crystals suitable for X-ray diffraction
CH2Cl2 λmax 314 nm (ε = 1.7 × 104 M−1 cm−1) and λmax 437 nm were prepared by slow vapor diffusion of hexane into a CHCl3
(ε = 2.6 × 102 M−1 cm−1).
solution. Yield 48 mg (80%). 1H NMR (300 MHz, d6-DMSO,
3: Yield: 34 mg (48%). 1H NMR (500 MHz, d6-DMSO, δ (ppm)): 8.78 (d, J = 6.0 Hz, 1H, H on py), 8.41 (s, 2H, H on
δ (ppm)): 9.40 (s, 1H, metal-bound NvCH), 9.09 (d, J = 5.0 Hz, –NH2), 8.34 (s, 1H, NvCH), 8.11 (m, 1H, H on py), 7.90 (d, J =
1H, H on metal-bound py), 8.87(s, 1H, NvCH), 8.81(d, J = 9.0 Hz, 1H on py), 7.48 (m, 1H, H on py).13C NMR (75 MHz d6-
4.5 Hz 1H, H on metal-bound py), 8.35(m, 2H, on metal bound DMSO) 197.6, 196.0, 188.4, 155.9, 151.9, 141.7, 139.6, 125.2,
and free py), 8.19 (d, J = 8.0 Hz, 1H, on free py), 8.05 (m, 1H, H 124.4 ppm. CHN Anal. Calc. for Re1C9H7O3N3Cl1: C, 25.32; H,
on metal bound py), 7.83(m, 1H, H on metal-bound py), 7.65 1.65; N, 9.84. Found: C, 25.47; H, 1.80; N, 9.77. IR (CO stretch,
(m, 1H, H on metal-bound py) ppm. 13C NMR (75 MHz d6- cm−1): 2025, 1892. MS (ESI): m/z = calc. for Re1C9H7O3N3Cl1Na:
DMSO, δ (ppm)) 196.8, 165.1, 163.4, 153.5, 153.4, 150.5, 150.0, 450.00 found 449.8. UV-Vis spectrum in CH2Cl2 λmax 296 nm (ε =
140.6, 137.7, 130.1, 129.7, 127.0, 122.6 ppm. CHN Anal. Calc. 8.6 × 103 M−1 cm−1) and λmax 388 nm (ε = 3.5 × 103 M−1 cm−1).
1
for Re1C15H10O3N4Br1: C, 32.15; H, 1.79; N, 9.99. Found: C,
9: Yield 36 mg (61%). H NMR (300 MHz, CDCl3, δ (ppm)):
32.04; H, 1.82; N, 9.73. IR (CO stretch, cm−1): 2021, 1893. MS 8.92 (d, 1H, J = 5.5 Hz, H on py), 8.18 (s, 1H, NvCH), 8.95 (m,
(ESI): m/z = calc. for Re1C15H10O3N4Br1Na: 582.93 found 582.9. 1H, H on py), 7.65 (d, J = 7.5 Hz, 1H on py), 7.42(m, 1H, H on
UV-Vis spectrum in CH2Cl2 λmax 310 nm (ε = 1.7 × 104 M−1 py), 6.54 (s, 2H,
H
on –NH2). CHN Anal. Calc. for
Re1C9H7O3N3Br1: C, 22.93; H, 1.49; N, 8.91. Found: C, 22.52;
cm−1) and λmax 433 nm (ε = 2.7 × 102 M−1 cm−1).
Synthesis of 4 and 5 [Re(CO3)X(pyca–pyca)Re(CO)3X; 4, X = H, 2.05; N, 8.95. IR (CO stretch, cm−1): 2019, 1873. MS (ESI):
Cl; 5 = Br]. The procedure for the synthesis of 4 is representa- m/z = calc. for Re1C9H7O3N3Br1Na: 493.91 found 493.8. UV-Vis
tive for both compounds. 4: Re(CO)5Cl (0.050 g, 0.141 mmol) spectrum in CH2Cl2 λmax 296 nm (ε = 8.2 × 103 M−1 cm−1) and
and 1 (0.015 g, 0.0705 mmol) were refluxed in 25 mL of n-pro- λmax 393 nm (ε = 2.9 × 103 M−1 cm−1).
panol for 12 h. After cooling to room temperature, a dark red
solid was filtrated and washed with 10 mL hexane. Yield:
47 mg (82%).Crystals suitable for X-ray diffraction were pre-
pared by slow vapor diffusion of water into a DMSO solution.
Results and discussion
1H NMR (300 MHz, d6-DMSO, δ (ppm)): 9.58 (d, 2H, NvCH), In an earlier report,39 we outlined the one-pot synthesis of a
9.10 (d, J = 6.0 Hz, 2H, H on py), 8.72 (d, J = 6.0 Hz, 2H, H on series of phenylene-bridged dinuclear Re(I) complexes, as well
py), 8.44 (m, 2H, H on py), 7.95 (m, 2H on py).13C NMR as their structures, spectroscopic features and electrochemical
(75 MHz d6-DMSO, δ (ppm)) 196.6, 195.8, 186.0, 167.3, 165.51, properties. These compounds could be readily synthesized via
154.17, 153.82, 151.7, 151.5, 141.0, 140.8, 132.1, 131.9, the reaction of phenylene diamines, pyridine-2-carboxyalde-
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
hyde and Re(CO)5X (X = Cl, Br). We surmised that we could meric analogues to 4 and 5, Re(CO)3 diimine with a terminal
generate “phenylene-free” analogs via the reaction of hydrazine amine group. Compounds 8 and 9, can be produced from pyri-
and pyridine-2-carboxyaldehdye. This reaction, shown in dine-2-carboxyaldehyde and an excess of hydrazine by first
Scheme 1, affords a bis-Schiff base diazine ligand (compound 1) forming the rhenium pyridine-2-carboxaldehyde adducts
which can coordinate one or two equivalents of Re(CO)3 and 6 and 7 in situ,41,42 followed by addition of hydrazine
can be considered structurally similar to the phenylene- (Scheme 1).
bridged ligands that we reported previously. This allowed us to
All new compounds synthesized for this study were fully
not only generate the desired dimeric compounds 4 and 5, but characterized, including via X-ray crystallography. Fig. 1 shows
also the monomeric species 2 and 3. Compound 2 was the structures of the bromide compounds (3, 5, and 9); the
reported previously.40 The degree of metalation was controlled chloride analogs 2, 4, and 8 effectively have nearly identical
by the choice of solvent; the higher boiling alcohol, n-propa- structures (see ESI†). The compounds exhibit the expected
nol, afforded the bis-metalated species, but reaction in metha- facial Re(CO)3 coordination mode observed for bidentate pyri-
nol resulted in the formation of only the singly-metalated dine imine chelates. The structure of the free ligand, 1, has
product. The monometalated products 2 and 3 can be used as been previously elucidated38 and the N–N and CvN bond dis-
controls to determine the effect of dimeric coupling on the tances of the dimeric Schiff base unit clearly show single bond
properties of 4 and 5. We also synthesized alternative mono- (1.413(1) Å) and double bond (1.266(1) Å), characters, respect-
ively. In compounds 2–5, we observe similar N–N and Schiff
base CvN bond distances in the metal-bound chelates, with
the only difference being a slight lengthening of the metal-
bound CvN double bonds (∼1.27–1.29 Å) due to metal–ligand
backbonding. These types of complexes are well known and
have readily observable spectroscopic features that can be used
to establish their identity.40,43–50 For example, the IR spectra of
these compounds exhibit a1 and
e type CO stretching
vibrations that result from the facial coordination geometry
and local C3v symmetry. In the 1H NMR spectra, the diagnostic
CH resonance of the imine confirms the presence of the
desired compounds.
Isomerization is clearly seen in the NMR spectra of 4 and 5,
which we have observed previously in similar compounds. Iso-
merization in 4 and 5 occurs because the halide can be placed
on either side of the diimine plane, leading to two stereo-
centers. Furthermore, as seen in the crystal structures, there
are chiral C2 twists coincident with the N–N bonds. Rotation
about the N–N bond is expected to be sterically restricted. For
5, only one diastereomer was crystallographically observed. For
4, a single chiral twist conformation was observed, but both
enantiomers were observed for each stereocenter. A similar be-
havior was observed for the m-phenylenediimine dimers that
we reported previously.39
A second, but indirect, way to observe the electronic effects
of forming a dimer from two Re diimine compounds via a
hydrazine bridge is by UV-visible spectroscopy. Fig. 2 shows
Scheme 1
Fig. 1 Structures of 3, 5 and 9 with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
shows the voltammograms for compounds 4 and 5 (top) versus
those from the monometallated compounds 2 and 3. The
scans were carried out in DMF with 0.1 M tetrabutyl-
ammonium hexafluorophosphate (TBAPF6) as the electrolyte.
These complexes exhibit reversible reductions in the cathodic
region, and the electrochemical data is summarized in ESI
Table S2.† Compounds 2 and 3 show a single reversible peak
at ∼−0.74 V, but for the bimetallic systems 4 and 5, two revers-
ible reductions are observed. These well-defined reductions are
separated by 485 and 482 mV for 4 and 5, respectively. These
separations can be converted to comproportionation constants
(Kc)27,39,51,52 of 1.54 × 108 and 1.48 × 108 for the two com-
pounds. These values, which correspond to the degree of coup-
ling in 4 and 5, are on the same order of magnitude as strongly
coupled Ru(bpy)2 dimers, such as those that incorporate tri-
azoleate and imidazolate bridges.46,51 We also investigated the
cyclic voltammetry of the monomeric compounds 8 and 9. The
voltammograms, shown in the ESI,† exhibit irreversible
reductions similar to those seen for the corresponding mono-
mers 2 and 3. The reductions can be assigned as taking place
on the diimine ligand regions of these dimers (vide infra).
In order to evaluate spectroscopic signatures of the reduced
species in monomeric complexes 2 and 3 as well as dimeric
analogues 4 and 5, we have conducted spectroelectrochemical
experiments in the DMF/0.3 M TBAP system. In the case of the
monomeric chloro-derivative 2, reduction under spectro-
electrochemical conditions results in an intensity decrease
and the low-energy shift of charge-transfer band at 309 nm to
327 nm. In addition, the initial band at 409 nm disappears,
while two new bands at 384 and 483 nm appear in the visible
region (Fig. 4). This reduction is reversible as neutral complex
2 can be regenerated upon oxidation of [2]− (ESI Fig. S29†).
Because of the lower stability of Re–Br bond, spectroelectro-
chemical data for complex 3 are more complicated (Fig. 5).
Although initial UV-vis spectra transformation during the first
reduction process looks similar to those observed in chloro
Fig. 2 UV-visible spectra for compounds 2, 4 and 8.
the spectra of the chloride variants of the monomers (2 and 8)
as well as the coupled dimeric species, 4. The spectra of the
bromide compounds are nearly identical in appearance (ESI†).
As in other examples of Re(CO)3 diimine complexes, all six
compounds exhibit low-energy metal-to-ligand charge transfer
(MLCT) bands in the visible region. We can readily determine
the identity of these transitions since they are sensitive to the
dielectric constant of the solvent. For the monomeric com-
pounds 8 and 9, the MLCT band is in the UV region of the
spectrum at 388 and 393 nm (ε = 3.5 × 103 M−1 cm−1). With
the monomeric compounds 2 and 3, the additional pyridine
imine unit induces a significant red shift of the MLCT absorp-
tion band to 437 nm and 433 (ε = 2.6 × 102 M−1 cm−1), due to
the presence (and thus overlap) of the two π systems. For the
dimeric systems 4 and 5, we observe an even greater batho-
chromic shift to 458 and 470 nm (ε = 6.0 × 103 M−1 cm−1). The
presence of two coupled MLCT transitions result in an approxi-
mately 20–30 nm shift in the UV-visible absorbance spectrum;
similar shifts have been observed in coupled Ru(II) systems.51
We can observe the electronic coupling between the
Re(CO)3(pyca) units directly by use of cyclic voltammetry. Fig. 3
Fig. 4 UV-vis spectra transformation of 2 into [2]− during the first
reduction under spectroelectrochemical conditions in DMF/0.3 M TBAP
system.
Fig. 3 Cyclic voltammograms of monomers and dimers in 0.1
TBAPF6/DMF at 0.25 V and 10.0 μA V−1 sensitivity versus AgCl.
M
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
intensities of 294 and 436 nm bands and appearance of four
new bands at 327, 465, 578, and 1130 nm in mixed-valence [4]−
(Fig. 6). During second single-electron reduction, initial tran-
sitions at 327 and 465 nm shift to 331 and 474 nm, while bands
at 578 and 1130 nm disappear (Fig. 6). Appearance of the broad
NIR band at 1130 nm during the first reduction and its dis-
appearance during the second reduction process is clearly
indicative of its intra-valence charge-transfer (IVCT) character.
On a qualitative level, spectroelectrochemical transform-
ations of the dibromo complex 5 are similar to those observed
in its dichloro analogue 4. In particular, both initial bands at
282 and 443 nm disappear, while four new transitions at 388,
467, 577, and 1130 nm appear in the spectrum of [5]− during
the first reduction process (Fig. 7). The NIR band observed at
1130 nm in mixed-valence [5]− is significantly more intense
than IVCT observed in dichloro analogue [4]− and clearly has
an asymmetric shape with a prominent shoulder at ∼1020 nm.
During second reduction, four bands at 388, 467, 577, and
1130 nm observed in UV-vis-NIR spectrum of [5]− disappear,
while two new bands at 355 and 482 nm appear in the spec-
trum of [5]2− (Fig. 7). Thus, the asymmetric NIR band in the
mixed-valence [5]− can be assigned as IVCT transitions. The
monomeric compounds 7 and 8 could not be investigated by
spectroelectrochemical methods due to the lack of reversibility
in their cyclic votammograms. The IVCT bands in the mixed-
valence [4]− and [5]− dimers were analyzed in the borders of
Hush theory formalism.54–57 Since the information on the
transition dipole moments for intra-ligand IVCT is uncertain,
we can only estimate the lower and the upper limits of the
coupling matrix element Hab. For IVCT analysis, the crystallo-
graphically determined Re–Re distance was used to estimate
the upper, and the pyridine(N)–pyridine(N) distance was used
to estimate the lower borders of Hab. Band deconvolution ana-
lysis of IVCT (ESI Fig. S31†) suggest that both [4]− and [5]−
complexes belong to the class III (completely delocalized)
systems in Robin–Day classification.57 The estimated Hab
Fig. 5 UV-vis spectra transformation of 3 into [3]− followed possible
dimerization during the first reduction under spectroelectrochemical
conditions in DMF/0.3 M TBAP system.
analogue 2, reoxidation of the electrochemically generated [3]−
anion results in a UV-vis spectrum which is different from
neutral complex 3. This spectrum is very similar to pyridine-
containing rhenium dimers with a direct Re–Re bond.53 Trans-
formation of complex 3 into possible dimer was further con-
firmed by continuous CV experiments (ESI Fig. S30†). Indeed,
during such experiments, the initial reduction peak of 3 dis-
appeared, while a new peak appears in the CV curve suggesting
transformation of 3 into new species under electrochemical
conditions.
In the case of dimeric complexes 4 and 5, the first and the
second reduction processes are well separated and reversible
and thus spectroscopic signatures of all reduced species can
be accurately characterized by spectroelectrochemical experi-
ments (Fig. 6 and 7). In the case of chlorine-containing dimer
4, the first, single-electron reduction results in decrease of
Fig. 6 UV-vis spectra transformation of 4 into [4]− during the first reduction (left) and [4]− into [4]2− during the second reduction (right) under
spectroelectrochemical conditions in DMF/0.3 M TBAP system.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
Fig. 7 UV-vis spectra transformation of 5 into [5]− during the first reduction (left) and [5]− into [5]2− during the second reduction (right) under
spectroelectrochemical condition in DMF/0.3 M TBAP system.
values are 307–420 cm−1 for [4]− and 1145–1647 cm−1 for [5]−
To elucidate spectroscopic and redox properties of the
complexes, while estimated Γ = 0.59 ([4]− complex) and 0.63 monomeric complexes 2, 3, 7, and 8 as well as dimers 4 and 5,
([5]− complex).
we have conducted DFT and TDDFT calculations of all target
complexes following procedures similar to those used by us for
molybdenum carbonyl compounds.58 The energy diagram for
these compounds is shown in Fig. 8, molecular orbital compo-
sitions are listed in the ESI,† and selected frontier orbitals are
shown in Fig. 9 and ESI Fig. S32–37.† The DFT-predicted elec-
tronic structure of all target compounds is quite similar,
which correlate well with their redox properties and UV-vis
spectra. First, occupied frontier orbitals are clearly indicative
of the low-spin d6 configuration of the rhenium centers. In
particular, in the case of monomeric complexes 2, 3, 7, and 8,
the HOMO, HOMO−1, and HOMO−2 are dominated by
rhenium dxz, dyz, and dxy orbital contributions with the first
two orbitals being close to each other in energy (Fig. 8). For
dimeric complexes 4 and 5, HOMO and HOMO−1 are degene-
rate and dominated by the equivalent contribution from both
rhenium centers forming standard pair of “in” (dxz + dxz) and
“out” (dxz − dxz) of phase combination of dxz orbitals. A similar
Fig. 8 DFT predicted energy diagram for target rhenium compounds.
Fig. 9 DFT predicted frontier MOs for target rhenium compounds.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
pattern of “in” and “out” of phase combinations of rhenium gies of all complexes are in reasonable agreement with the
dyz and dxy orbitals was also observed for HOMO−2/HOMO−3 experimental data (Fig. 10). In the case of all four mono-
and HOMO−4/HOMO−5 pairs, respectively. Again, predomi- nuclear rhenium complexes, two major regions of the spectra
nant dxz and dyz orbital combinations are closely spaced, while were predicted by TDDFT calculations. The low-energy region
combinations of dxy orbitals are energetically well separated. is described by the MLCT transitions dominated by the
In all target complexes, rhenium-centered occupied frontier intense band mostly originating from HOMO−1 → LUMO exci-
orbitals follow ligand-based π-MOs and then more stable tation. The UV region of the spectra has a major contribution
halogen-centered orbitals (Fig. 8). The LUMO and LUMO+1 in from intra-ligand π–π* excitations. Because of the presence of
all rhenium compounds are predominantly ligand-centered π*- a larger number of rhenium-centered occupied orbitals in
orbitals with the LUMO well-separated in energy (∼1 eV) from dimeric complexes 4 and 5, TDDFT predicted UV-vis spectra of
the LUMO+1. The CO-centered π* orbitals have significantly these complexes have a larger number of excited states,
higher energy than LUMO and LUMO+1. Overall, the electronic although qualitative picture remains the same: low-energy
structure of the rhenium complexes discussed in this paper is transitions have predominantly MLCT character, while UV
in agreement with the previously reported rhenium–pyridine region is dominated by the intra-ligand π–π* excitations.
carbonyls59,60 and suggestive that the first oxidation should be
metal-centered, while the first reduction should be centered
on the aromatic ligand, which correlate well with the electro-
and spectroelectrochemical data presented above. Thus,
Conclusions
associated with the mixed-valence state in [4]− and [5]− the In conclusion, dimeric Re(CO)3 pyridine imine complexes can
IVCT band represents the intra-ligand rather than metal-to- be readily generated by reaction of ligand 1 with Re(CO)5X pre-
metal charge transfer and in the absence of information on cursors. Both monomeric and dimeric species can be gene-
the transition dipole moments, its analysis should be viewed rated by use of alcoholic solvents with different boiling points.
as only a rough estimate.
Simple monomeric compounds (those with ligand 1 and only
In order to improve tentative assignments in the UV-vis one equivalent of metal and monomeric Schiff base com-
spectra of all rhenium complexes, we have conducted a set of plexes) can also be generated and provide useful controls for
TDDFT calculations on these systems. TDDFT predicted ener- understanding the electronic interactions between the two
halves of the dimeric compounds 4 and 5. We observe strong
coupling in compounds 4 and 5 as evidenced by cyclic voltam-
metry as well as UV-visible spectroscopy. Spectroelectrochem-
ical measurements support these conclusions with the
observation of IVCT bands in 4 and 5 upon single electron
reduction. DFT and TDDFT calculations indicate that the
reductions are ligand-centered, which support intra-ligand
transitions for the IVCT bands We are continuing our work on
Re(CO)3 dimeric and polymeric systems, and are investigating
their properties as possible components of photonic materials.
Acknowledgements
RSH would like to thank the Petroleum Research Fund (51085-
UR3) for financial support of this research. CJZ would like to
acknowledge the University of Akron for support of this work
as well as and the National Science Foundation (CHE-9977144)
for funds used to purchase the diffractometer used in this
work. Generous support from Minnesota Supercomputing
Institute for VNN is greatly appreciated.
References
1 A. Juris, S. Campagna, I. Bidd, J. M. Lehn and R. Ziessel,
Inorg. Chem., 1988, 27, 4007–4011.
2 M. Mauro, C.-H. Yang, C.-Y. Shin, M. Panigati, C.-H. Chang,
G. D’Alfonso and L. De Cola, Adv. Mater., 2012, 24, 2054–
Fig. 10 TDDFT predicted UV-vis spectra for target rhenium
compounds.
2058.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Dalton Transactions
Paper
3 G. Pereiras-Gabián, E. M. Vázquez-López and U. Abram, 27 P. J. Ball, T. R. Shtoyko, J. A. Krause Bauer, W. J. Oldham
Z. Anorg. Allg. Chem., 2004, 630, 1665–1670. and W. B. Connick, Inorg. Chem., 2004, 43, 622–632.
4 T. Rajendran, B. Manimaran, F.-Y. Lee, G.-H. Lee, 28 P. T. Kissinger, W. Lafayette and W. R. Heineman, J. Chem.
S.-M. Peng, C. M. Wang and K.-L. Lu, Inorg. Chem., 2000,
39, 2016–2017.
Educ., 1983, 60, 702–706.
29 A. D. Le and L. Yu, J. Electrochem. Soc., 2011, 158, F10.
5 Y.-H. Tseng, D. Bhattacharya, S.-M. Lin, P. Thanasekaran, 30 N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877–
J.-Y. Wu, L.-W. Lee, M. Sathiyendiran, M.-L. Ho, 910.
M.-W. Chung, K.-C. Hsu, P.-T. Chou and K.-L. Lu, Inorg. 31 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
Chem., 2010, 49, 6805–6807.
6 E. Ferri, D. Donghi, M. Panigati, G. Prencipe, L. D’Alfonso,
32 C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens.
Matter, 1988, 37, 785–789.
I. Zanoni, C. Baldoli, S. Maiorana, G. D’Alfonso and 33 A. Canal Neto and F. E. Jorge, Chem. Phys. Lett., 2013, 582,
E. Licandro, Chem. Commun., 2010, 46, 6255–6257. 158–162.
7 M. Panigati, M. Mauro, D. Donghi, P. Mercandelli, 34 A. D. McLean and S. G. Chandler, J. Chem. Phys., 1980, 72,
P. Mussini, L. De Cola and G. D’Alfonso, Coord. Chem. Rev.,
2012, 256, 1621–1643.
8 J. A. Baiano, D. L. Carlson, G. M. Wolosh, D. E. DeJesus,
5639.
35 J. Tomasi, B. Mennucci and R. Cammi, Chem. Rev., 2005,
105, 2999–3093.
C. F. Knowles, E. G. Szabo and W. R. Murphy, Inorg. Chem., 36 M. J. M. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
1990, 29, 2327–2332. et al., Gaussian 09, 2009.
9 P. J. Blower, J. R. Dilworth, J. P. Hutchinson and J. Zubieta, 37 A. L. Tenderholt, QMForge.
Transition Met. Chem., 1982, 7, 354–355.
10 T. Tabeya, M. Abe, A. Mitani, K. Tsuge and Y. Sasaki,
J. Nucl. Radiochem. Sci., 2005, 6, 157–159.
11 J. W. M. Van Outersterp, F. Hartl and D. J. Stufkens, Orga-
nometallics, 1995, 14, 3303–3310.
12 A. Vogler and J. Kisslinger, Inorg. Chim. Acta, 1986, 115,
193–196.
13 V. W. Yam, K. M. Wong and K. Cheung, Organometallics,
1997, 16, 1729–1734.
38 E. C. Kesslen, W. B. Euler and B. M. Foxman, Chem. Mater.,
1999, 11, 336–340.
39 A. Hasheminasab, J. T. Engle, J. Bass, R. S. Herrick and
C. J. Ziegler, Eur. J. Inorg. Chem., 2014, 2014, 2643–2652.
40 N. M. Shavaleev, Z. R. Bell, G. Accorsi and M. D. Ward,
Inorg. Chim. Acta, 2003, 351, 159–166.
41 R. Costa, K. Chanawanno, J. T. Engle, B. Baroody,
R. S. Herrick and C. J. Ziegler, J. Organomet. Chem., 2013,
734, 25–31.
14 B. J. Yoblinski, M. Stathis and T. F. Guarr, Inorg. Chem., 42 L. A. García-Escudero, D. Miguel and J. A. Turiel, J. Organo-
1992, 31, 5–10. met. Chem., 2006, 691, 3434–3444.
15 S. Van Wallendael, R. J. Shaver, D. P. Rillema, 43 D. A. Edwards, G. M. Hoskins, M. F. Mahon,
B. J. Yoblinski, M. Stathis and T. F. Guarr, Inorg. Chem.,
1990, 29, 1761–1767.
K. C. Malloy and G. R. G. Rudolph, Polyhedron, 1998, 17,
2321–2326.
16 C. M. Álvarez, R. García-Rodríguez, J. M. Martín-Alvarez, 44 W. J. Stratton and D. H. Busch, J. Am. Chem. Soc., 1960, 82,
D. Miguel and J. A. Turiel, Inorg. Chem., 2012, 51, 3938–
3940.
17 G. Böhm, K. Wieghardt, B. Nuber and J. Weiss, Angew.
Chem., Int. Ed. Engl., 1990, 29, 787–790.
4834–4839.
45 W. J. Stratton and D. H. Busch, J. Am. Chem. Soc., 1958, 80,
3191–3195.
46 P. Cai, M. Li, C.-Y. Duan, F. Lu, D. Guo and Q.-J. Meng, New
J. Chem., 2005, 29, 1011.
18 C. Bruckmeier, M. W. Lehenmeier, R. Reithmeier,
B. Rieger, J. Herranz and C. Kavakli, Dalton Trans., 2012, 47 M. Chandra, A. N. Sahay, S. M. Mobin and D. S. Pandey,
41, 5026–5037. J. Organomet. Chem., 2002, 658, 43–49.
19 J. A. Cabeza, A. Llamazares, V. Riera, R. Trivedi and 48 V. Chandrasekhar, T. Hajra, J. K. Bera, S. M. W. Rahaman,
F. Grepioni, Organometallics, 1998, 17, 5580–5585.
20 J. Granifo, Polyhedron, 1999, 18, 1061–1066.
N. Satumtira, O. Elbjeirami and M. A. Omary, Inorg. Chem.,
2012, 51, 1319–1329.
21 H. Qayyum, R. S. Herrick and C. J. Ziegler, Dalton Trans., 49 R. Karmakar, C. R. Choudhury, S. R. Batten and S. Mitra,
2011, 40, 7442–7445. J. Mol. Struct., 2007, 826, 75–81.
22 E. Spaltenstein and J. M. Mayer, J. Am. Chem. Soc., 1991, 50 G. Mahmoudi and A. Morsali, Cryst. Growth Des., 2008, 8,
113, 7744–7753. 391–394.
23 J. W. M. Van Outersterp, D. J. Stufkens and A. Vlcek, Inorg. 51 D. E. Richardson and H. Taube, J. Am. Chem. Soc., 1983,
Chem., 1995, 34, 5183–5194. 105, 40–51.
24 F. P. A. Johnson, M. W. George, F. Hartl and J. J. Turner, 52 D. M. D’Alessandro and F. R. Keene, Dalton Trans., 2004,
Organometallics, 1996, 15, 3374–3387. 3950–3954.
25 G. J. Stor, F. Hartl, J. W. M. van Outersterp and 53 D. L. Morse and M. S. Wrighton, J. Am. Chem. Soc., 1976,
D. J. Stufkens, Organometallics, 1995, 14, 1115–1131. 98, 3931–3934.
26 B. D. Rossenaar, F. Hartl and D. J. Stufkens, Inorg. Chem., 54 N. S. Hush, Prog. Inorg. Chem., 1967, 8, 391–444.
1996, 35, 6194–6203.
55 C. Creutz, Prog. Inorg. Chem., 1983, 30, 1–73.
This journal is © The Royal Society of Chemistry 2015
Dalton Trans.

View Article Online
Paper
Dalton Transactions
56 N. S. Hush, Coord. Chem. Rev., 1985, 64, 135–157.
57 M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem.,
1968, 10, 247–422.
58 V. N. Nemykin, J. G. Olsen, E. Perera and P. Basu, Inorg.
Chem., 2006, 45, 3557–3568.
59 A. Drozdz, M. Bubrin, J. Fiedler, S. Záliš and W. Kaim,
Dalton Trans., 2012, 41, 1013–1019.
60 R. Heydová, E. Gindensperger, R. Romano, J. Sýkora,
A. Vlček, S. Záliš and C. Daniel, J. Phys. Chem. A, 2012, 116,
11319–11329.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2015