Examples of reductive azo cleavage and oxidative azo bond formation on Re2(CO)10 template: Isolation and characterization of Re(III) complexes of new azo-aromatic ligands

Article abstract of DOI:10.1021/ic201200r

A new example of simultaneous reductive azo bond cleavage and oxidative azo bond formation in an azo-aromatic ligand is introduced. The chemical transformation is achieved by the reaction of Re₂(CO)₁₀ with the ligand 2-[(2-N-Arylamin

Full text of DOI:10.1021/ic201200r

ARTICLE
pubs.acs.org/IC
Examples of Reductive Azo Cleavage and Oxidative Azo Bond
Formation on Re2(CO)10 Template: Isolation and Characterization of
Re(III) Complexes of New Azo-Aromatic Ligands
Nanda D. Paul,† Subhas Samanta,† Tapan K. Mondal,‡ and Sreebrata Goswami*,†
†Department of Inorganic Chemistry, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700 032, India
‡Department of Chemistry, Jadavpur University, Jadavpur, Kolkata 700 032, India
S Supporting Information
b
ABSTRACT: A new example of simultaneous reductive azo
bond cleavage and oxidative azo bond formation in an azo-
aromatic ligand is introduced. The chemical transformation is
achieved by the reaction of Re2(CO)10 with the ligand 2-[(2-N-
Arylamino)phenylazo]pyridine (HL1). A new and unexpected
mononuclear rhenium complex [Re(L1)(L3)] (1) was isolated
from the above reaction. The new azo-aromatic ligand, H2L3
(H2L3 = 2, 20-dianilinoazobenzene) is formed in situ from HL1.
A similar reaction of Re2(CO)10 and a closely related azo-ligand,
2,4-ditert-butyl-6-(pyridin-2-ylazo)-phenol (HL2), resulted in a seven coordinated compound [Re(L2){(L4)•ꢀ}2] (2; HL4 =
2-amino-4,6-ditert-butyl-phenol) via reductive cleavage of the azo bond. The complexes have been characterized by using a host of
physical methods: X-ray crystallography, nuclear magnetic resonance (NMR), cyclic voltammetry, ultravioletꢀvisible (UVꢀvis),
electron paramagnetic resonance (EPR) spectroscopy, and density functional theory (DFT). The experimental structures are well
reproduced by density functional theory calculations and support the overall electronic structures of the above compounds.
Complex 1 is a closed shell singlet, while complex 2 exemplifies a singlet diradical complex where the two partially oxidized
aminophenoleto ligands are coupled to each other, yielding the observed diamagnetic ground state. Complexes 1 and 2 showed two
successive one-electron redox responses. EPR spectral studies in corroboration with DFT results indicated that all of the redox
processes occur at the ligand center without affecting the trivalent state of the metal ion.
Scheme 1. Electron Transfer Series
’ INTRODUCTION
Metathesis reactions1 in unsaturated organic molecules con-
stitute an important class of chemical reactions that occur with
simultaneous bond cleavage and bond formation processes.
Accordingly, metathesis in olefins has gained widespread use in
research and industry for making products ranging from medi-
cines and polymers to enhanced fuels.1,2 In comparison, such a
reaction in azo compounds is not known. Diazo compounds, in
general, and azo-aromatics, in particular, are high-value chemicals
widely used in industry as organic dyes, indicators, pigments,
food additives, radical reaction initiators, and therapeutic agents.3
In recent times, we have been in search4 of synthetic protocols
suitable for the isolation of transition metal complexes of azo-anion
radical ligands. In this respect, zerovalent metal carbonyl complexes
were found4aꢀc to be useful precursors where the electron-rich
metal center acts as a possible electron pool. Following this synthetic
protocol, we have been successful in isolation of stable complexes of
azo-anion radical ligands. The ligands containing azo functions are
known to undergo reduction4ꢀ6 chemically or electrochemically at
rather accessible potentials to form stable azo-anion radicals (ꢀ1) or
dianionic hydrazido (ꢀ2) species. The further addition of two more
electrons to a hydrazido function leads to the rupture7 of the NꢀN
bond (Scheme 1).
With this available knowledge, we set out to explore the
chemical reactions of Re2(CO)10 with the two extended triden-
tate azoaromatic ligands, HL1 and HL2 (Scheme 2), that are
known to undergo facile reduction and produce stable azo-anion
radical complexes.4 In practice, the reference chemical reaction
with HL1 produced a Re compound [Re(L1)(L3)] (1) of a newly
formed azoaromatic ligand H2L3 (Schemes 3 and 4). The
formation of H2L3 from HL1 is noteworthy and proceeds via
simultaneous azo bond cleavage and NꢀN bond fusion reac-
tions. However, a similar reaction with HL2 (Scheme 3) and
Re2(CO)10 led only to the reductive cleavage of the azo function
with the formation of a complex of a 2-aminophenol derivative,
[Re(L2){(L4)•ꢀ}2] (2; Scheme 4). Notably, metal mediated azo
bond cleavage7 leading to the imido complexes has long been
Received: June 4, 2011
Published: July 20, 2011
r
2011 American Chemical Society
7886
dx.doi.org/10.1021/ic201200r Inorg. Chem. 2011, 50, 7886–7893
|

Inorganic Chemistry
ARTICLE
Scheme 2. Ligands: Reactants
Scheme 4. Ligands: Products
Scheme 3. Reactions
which could not be purified so far. It may be worth noting that
seven coordinated complexes of Re are rare.9
The complexes, 1 and 2, gave satisfactory elemental analyses
(cf., Experimental Section). Electrospray mass spectra of the
complexes corroborate their formulation as [Re(L1)(L3)] (1)
and [Re(L2)(L4)2] (2), respectively. For example, each of the
complexes 1a and 2 showed an intense peak due to the respective
molecular ion [1a-H]+ and [2-H]+ at m/z 822 and 936 amu,
respectively. Notably, the experimental spectral features of the
complexes corroborate very well with their simulated isotopic
pattern for the given formulation. Representative spectra of 1a
and 2 along with their simulations are in the Supporting
Information, Figures S1 and S2.
known in the literature, but unification of metal-imido fragments
to azo arene via NꢀN coupling reactions is scarce.8 And these
two reactions occurring together in a single chemical reaction, as
happened here, is nonexistent in the literature.
’ RESULTS AND DISCUSSION
Complexes 1 and 2 possess S = 0 ground states, as determined
by magnetic susceptibility measurements at room temperature.
Syntheses and Characterization. The ligand, HL1, offers
three very different kinds of coordination atoms, namely, a
pyridyl-N, a reducible azo function containing N, and a diaryla-
mido-N (N, after deprotonation), whereas the ligand HL2 is very
similar to the previous one where the secondary amide-N is
replaced by a phenolate group (Scheme 2). Pyridyl-N and azo-N
are π accepting, whereas deprotonated amido-N and phenolate-
O ligands are strongly donating.
1
They display sharp H NMR signals in the normal range10 for
diamagnetic compounds. The 1H NMR spectra of the represen-
tative complexes are submitted as Supporting Information,
Figures S3ꢀS5. Notably, the protons in the mixed ligand
complexes 1 and 2 are unique and displayed a very large number
of resonances. There has been serious overlap of resonances of
the aromatic region, and it was not possible to assign them.
However, complex 1b showed three methyl proton resonances at
δ 1.99, 2.19, and 2.32 ppm. The two imine NꢀH protons of 2
appeared at δ 8.98 and 9.32 ppm, respectively, which disappeared
on D2O shake.
The reactions of Re2(CO)10 (0.15 mmol) with HL1 (0.60mmol)
in boiling n-octane produced new reddish pink [Re(L1)(L3)]
complexes (1a and 1b) in 45 and 43% yield, respectively
(Scheme 3). While selecting Re(0)-carbonyl as the starting
compound, it was anticipated that the substitution of CO ligands
by [L1]ꢀ with concomitant internal electron transfer from the
electron-rich metal Re(0) to the reducible azo aromatic ligand
[L1]ꢀ would produce the metal radical complex directly.4aꢀc
However, the reference reaction produced a totally unexpected
product. The compound, isolated from the reaction mixture,
contained an undissociated tridentate ligand [L1]ꢀ and a new
dianionic azo-aromatic ligand [L3]2ꢀ coordinated in a tridentate
fashion. Formation of the new ligand [H2L3] is noteworthy, which
involves simultaneous reductive azo bond cleavage and NꢀN bond
formation. These results immediately prompted us to explore a
similar reaction of Re2(CO)10 with another extended azo-aromatic
ligand. We purposefully chose HL2 for carrying out the reaction,
as its conjugate base offers a nearly similar coordination environ-
ment like [L1]ꢀ. The reaction of HL2 and Re2(CO)10, under
identical reaction conditions, however, formed a seven coordi-
nated mixed ligand complex [Re(L2){(L4)•ꢀ}2] (2) in nearly
60% yield (Schemes 3 and 4). In this compound, two partially
oxidized (semiquinone) aminophenolato ligands (L4•ꢀ) are co-
ordinated along with an undissociated [L2]ꢀ ligand. 2-Aminophe-
noleto ligands are formed in situ via the reductive cleavage of the
parent [L2]ꢀ ligand. The above chemical reactions also produced
several low-yield byproducts, as seen on a preparative TLC plate,
Rationale for the Transformation HL1fH2L3. The reac-
tion of Re2(CO)10 with HL1 exemplifies an unprecedented
chemical reaction wherein a dianionic tridentate N,N,N donor
(H2L3) is formed via simultaneous four electron reductive
cleavage and oxidative recombination of the aryl imides. The
conversion HL1fH2L3 occurs within the coordination sphere,
and the overall reaction is a multiple electron transfer process
where the electron-rich dimetallic Re2(CO)10 serves as an
electron pool.
The mechanisms of these chemical reactions are not resolved
as yet. However, a few reactions have been planned, and the
results along with related literature have been used to make a
reasonable assessment of the reference reaction. It has been
argued7a,11 before that both reductive azo cleavage as well as
oxidation of the coordinated organo-imides to a diazo function
are facilitated by the dimetallic mediators. For example, an
oxidative NꢀN bond formation reaction in a Ta complex has
been achieved recently in a bridging imido dimer.8 For compar-
ison, in the present case, the zero-valent Re2 complex has
mediated the two opposite redox processes, reductive bond
cleavage as well as oxidative azo bond fusion, simultaneously. It
is worth noting here that the reactions of the monomeric
Re(CO)5Cl with HL1 under identical conditions failed to bring
7887
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
Scheme 5. Summary of Ligand Transformations
Figure 2. Molecular view of 2. Hydrogen atoms are omitted for clarity.
like pap or 4-aryl substituted pap (N/\N) (L5) with Re2(CO)10
(see Experimental Section) under identical reaction conditions,
underwent substitution and electron transfer, producing a dir-
adical complex, [ReII(N/\N•ꢀ)2(CO)2].4a To exclude the pos-
sibility of the formation of H2L3 via multiple oxidative bond
fusion of aromatic mono amine, produced via deamination of
HL1, a reaction with the ꢀMe substituted ligand HL1b
(Scheme 2) was carried out. Interestingly, the product obtained
from this reaction contains the new ligand H2L3b, which is
symmetrical about the azo function. All of these observations
taken together suggest that HL1fH2L3 conversion occurs via a
sort of overall azo bond metathesis. However, the other metath-
esis product could not be identified so far. The observed ligand
transformations in the present work are summarized in
Scheme 5.
X-Ray Crystal Structure. The complexes 1b and 2 form good
quality crystals, whose structures have been solved by single-crystal
X-ray diffraction studies. Molecular views of the compounds 1b
and 2 are displayed in Figures 1 and 2, respectively. Crystal-
lographic data and selected experimental aswellas DFT computed
structural parameters are listed in Tables 1ꢀ3; ORTEPs are
submitted as Supporting Information (Figures S7 and S8).
The X-ray structure of 1b indeed confirmed the formation of
H2L3 from HL1. Complex 1b is hexa-coordinated and is mer-
idional. The two deprotonated ligands, [L1b]ꢀ and [L3b]2ꢀ
,
Figure 1. Molecular view of 1b. Hydrogen atoms are omitted for clarity.
coordinate the central rhenium ion in an N6-fashion using pairs
of pyridyl-N, azo-N, and amido-N (deprotonated secondary
amine) atoms. The chelate bite angle of N(pyridyl)ꢀReꢀN-
(azo) (73.57(5)°) is smaller than the average of N(azo)ꢀReꢀN-
(amide) (81.83(6)°).10 The crystal structure of compound 2, on
the other hand, is hepta-coordinated.9 An anionic tridentate
ligand [L2]ꢀ and two bidentate amino phenolato ligands (HL4)ꢀ
complete the coordination environment about the central Re
atom. The ligand [HL4]ꢀ is formed via reductive cleavage of the
azo bond in HL2. Its coordination geometry is distorted penta-
gonal bipyramidal, and the rhenium atom sits in a basal plane
formed by O1, O2, O3, N1, and N3 atoms. The axial positions are
occupied by N4 and N5 atoms (N4ꢀRe1ꢀN5 angle is 160°)
(Figure 2) of the two [HL4]ꢀ ligands. Notably, the two nitrogen
atoms of the amino phenolate ligands are disposed trans to each
other. We wish to note here that the average d(NꢀN) lengths of
about similar transformation. The latter reaction produced
[Re(pap)(CO)3Cl] (3) (pap =2-(phenylazo)pyridine, N/\N)
as the only isolable product (cf. Experimental Section).12 The
formation of pap from HL1 via deamination4c has been observed
previously in several other reactions. The three-dimensional
X-ray structure (ORTEP) of compound 3 is submitted as
Supporting Information Figure S6. In contrast, a similar reaction
of HL2 with Re2(CO)10 stopped after the cleavage of the azo
bond, producing amino phenol from HL2. We wish to note here
that the two coordinated nitrogen atoms of the two amino
phenolato ligands in compound 2 are disposed in trans orienta-
tions, which is not a favorable geometry for an interligand
recombination reaction for NꢀN bond formation. Moreover,
the reactions of closely related bidentate azo-aromatic ligands,
7888
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
Table 1. Crystallographic Data for 1b, 2, and 3
1b
2
3
empirical formula
molecular mass
temp (K)
C
44H37N8Re
C47H66N5O3Re
963.27
C14H9ClN3O3Re
488.90
864
150 K
150 K
150 K
cryst syst
monoclinic
P21/c
triclinic
monoclinic
P21/n
space group
a (Å)
P1
9.7491(3)
19.0490(6)
21.1424(6)
90
10.579(6)
12.786(7)
18.829(11)
94.595(14)
92.864(16)
102.871(14)
2469(2)
2
12.4199(19)
9.0921(14)
13.818(2)
90
b (Å)
c (Å)
R (deg)
β (deg)
100.0780(10)
90
102.650(5)
90
γ (deg)
V (Å3)
3865.8(2)
4
1522.5(4)
4
Z
Dcalcd (g/cm3)
cryst dimens (mm)
θ range for data collection (deg)
GOF
1.630
1.296
2.133
0.08 ꢁ 0.16 ꢁ 0.18
1.5ꢀ28.6
1.01
0.15 ꢁ 0.17 ꢁ 0.19
1.1ꢀ24.7
1.03
0.08 ꢁ 0.16 ꢁ 0.18
2.0ꢀ24.3
1.01
reflns collected
unique reflns
final R indices [I > 2σ(I)]
43806
26869
16378
9605
8364
2484
R1 = 0.0338
wR2 = 0.0714
R1 = 0.0599
wR2 = 0.1668
R1 = 0.0322
wR2 = 0.0702
Table 2. Summary of Selected Experimental and Calculated
Bond Lengths (Å) of Complexes 1b, [1b]+, and [1b]2+
Table 3. Summary of Selected Experimental and Calculated
Bond Lengths (Å) of Complexes 2, [2]+, and [2]2+
1b
theoretical
[1b]+
[1b]2+
2
[2]+
[2]2+
bonds
X-ray
theoretical
theoretical
bonds
X-ray
theoretical
theoretical
theoretical
Re1ꢀN(1)
Re1ꢀN(3)
Re1ꢀN(4)
Re1ꢀN(5)
Re1ꢀN(7)
Re1ꢀN(8)
N(2)ꢀN(3)
N(6)ꢀN(7)
C(35)ꢀN(8)
2.076(3)
1.972(3)
1.957(3)
1.977(3)
2.037(3)
2.081(3)
1.336(5)
1.330(5)
1.402(5)
2.101
1.990
2.012
2.008
2.077
2.131
1.322
1.296
1.400
2.109
2.020
1.992
2.038
2.062
2.100
1.306
1.304
1.382
2.120
2.073
1.976
2.116
2.047
2.028
1.286
1.316
1.343
Re1ꢀO1
Re1ꢀO2
Re1ꢀO3
Re1ꢀN1
Re1ꢀN3
Re1ꢀN4
Re1ꢀN5
C25ꢀO2
N2ꢀN3
C20ꢀN4
C20ꢀC25
C21ꢀC22
C23ꢀC24
C39ꢀO3
C34ꢀN5
C34ꢀC39
C35ꢀC36
C37ꢀC38
2.016(6)
1.976(6)
2.047(6)
2.066(8)
2.020(8)
1.990(7)
1.971(8)
1.350(12)
1.347(11)
1.353(11)
1.378(14)
1.392(16)
1.389(16)
1.332(11)
1.361(12)
1.435(12)
1.366(14)
1.409(13)
2.069
2.081
2.035
2.098
2.047
2.035
2.019
1.325
1.325
1.365
1.428
1.386
1.392
1.337
1.364
1.418
1.388
1.394
2.063
2.118
2.091
2.131
2.087
2.087
2.052
1.301
1.312
1.340
1.445
1.375
1.379
1.309
1.347
1.441
1.377
1.374
2.024
2.132
2.113
2.159
2.121
2.077
2.076
1.291
1.301
1.326
1.469
1.372
1.374
1.298
1.327
1.464
1.371
1.371
the azoaromatic ligands in both the compounds 1b and 2 are
elongated (for 1b, 1.333(5); 2, 1.347(11)Å) compared to the
uncoordinated ligand salt,13 [HL1a]ClO4 (1.246(5)Å). The
elongation of the ꢀNdNꢀ chromophores in these complexes
is due to extensive Re(dπ)-ligand(π*) back-bonding,10,14 which
has been supported by DFT studies. However, the two amino-
phenolates in complex 2 are partially oxidized: the CꢀN and
CꢀO lengths are in agreement with the bond lengths in its
semiquinone oxidation state.15 Ambiguities of oxidation states in
the above coordinated azo-aromatics have been finally sorted out
by their DFT analysis (cf. below).
The metalꢀnitrogen bonds in these complexes are unexcep-
tional, lying in the range 1.95ꢀ2.07 Å. The bonds from Re to the
neutral 2-pyridyl-N atoms are distinctly longer than those to the
anilido or azo nitrogen atoms.10,14 The effects of the ꢀNdNꢀ
bond elongation due to back bonding are also reflected by the
lowering of vibrational frequencies of νNdN as compared to the
uncoordinated HL1/HL2. For example, the νNdN band in free
ligands appears near 1380 cmꢀ1, whereas it is lowered consider-
ably in the coordinated states and appeared at 1285 and
1260 cmꢀ1 in the complexes 1a and 2, respectively.13,16
Cyclic Voltammetry and EPR. Cyclic voltammograms of 1a,
1b, and 2 were recorded in CH2Cl2 solutions containing 0.1 M
TBAP at 25 °C. The potentials are referenced to the saturated
7889
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
Table 4. Electrochemical Dataa
Scheme 6. Different Electronic Possibilities of H2L4
compound
oxidation E1/2,b V (ΔEp, mV)
[Re(L1a)(L3a)] (1a)
[Re(L1b)(L3b)] (1b)
[Re(L2)(L4•ꢀ)2] (2)
0.43 (80), 0.82 (70)
0.41 (75), 0.79 (80)
0.53 (75), 0.10 (80)
a Conditions: solvent, dichloromethane; supporting electrolyte, NBut4-
ClO4 (0.1M); working electrode, platinum for oxidation and reduc-
tion processes; reference electrode, Ag/AgCl; solute concentration, ca.
10ꢀ3(M); scan rate, 50 mVsꢀ1. b E1/2 = 1/2(Epa+Epc), ΔEp = Epa ꢀ Epc
in mV (Epa = anodic peak potential, Epc = cathodic peak potential).
2, the geometries of the complexes 1b and 2 in both singlet and
triplet states were optimized, using density functional theory
(B3LYP and SDD on Re; 6-31+G(d) on C, N, and O; 6-31G on
H atoms—see Experimental Section). The agreement between
the experimental structural parameters and that obtained
through DFT calculations for 1b and 2 were found to be
excellent. Calculated selected bond lengths and angles are
compared with experimental ones in Tables 2 and 3.
The most stable structure of 1b proves to be a closed-shell
singlet state, [ReIII(L1b)(L3b)], and all attempts to locate open-
shell singlet [ReV(L1b)•ꢀ(L3b)•ꢀ] configurations using the BS-
DFT approach converged back to the closed-shell configuration.
Complex 2, on the other hand, on broken symmetry density
functional theoretical calculations (BS (S = 0)), clearly shows that
the two aminophenoleto ligands (L4) are open-shell, π-radical
monoanions [(L4)•ꢀ], as was shown computationally for the
analogous systems.15 Thesensitive bond distances associated with
the coordinated L4 (CꢀO and CꢀN X-ray bond distances) and
NBO calculated average Wiberg bond order (CꢀO, 1.138 (1.151
and 1.125) and CꢀN, 1.223 (1.229 and 1.219)) in complex 2
suggests its intermediate semiquinone [(L4)•ꢀ] radical state
leading to the valence configuration of [ReIII(L2)ꢀ{(L4)•ꢀ}2].
Notably, one of these two ligands remains in the spin-up
manifold, while the second one is in the spin-down manifold,
yielding the observed antiferromagnetically coupled diamagnetic
singlet (S = 0) ground state. This notion is further confirmed by
the Mulliken spin population analysis (Figure S10, Supporting
Information) and is very similar to the results calculated for
similar systems.15 The elongation of the azo-chromophores in
both complexes 1 and 2 is due to substantial dπ(Re)fπ*(NdN)
back bonding, which indeed is consistent with the substantially
reduced Wiberg index for the NꢀN bonds (1.33 and 1.41 for 1b,
1.29 for 2) compared to the free-ligand value of 1.77.14
The energies and compositions of the DFT calculated frontier
orbitals for both complexes, [Re(L1b)(L3b)] (1b) and
[ReIII(L2){(L4)•ꢀ}2] (2), are listed in the Supporting Informa-
tion (Tables S1 and S2). For both complexes 1b and 2, the
MO representations show the highest occupied MO (HOMO)
to be formed by a combination of d metal orbitals and
ligand orbitals with a major contribution from the ligand orbitals
(1b: L1b, 05%; Re, 09%; L3b, 86% and 2: L2, 11%, Re, 04%; L4,
85%). Closely lying occupied HOMOꢀ1 is again formed by a
major contribution of ligand orbitals. The lowest lying unoccu-
pied MOs (LUMOs) are, to a large extent, delocalized over
the π system of the ligands; metal orbitals contribute to this
orbital only by a very little amount (Tables S1 and S2, Supporting
Information).
Figure 3. Cyclic voltammogram of 1 and 2 in CH2Cl2.
Ag/AgCl electrode. The results are summarized in Table 4, and
the voltammograms (for 1a and 2) are displayed in Figure 3.
Both complexes 1 and 2 showed two reversible anodic responses
in the potential range 0.3ꢀ1.0 V. One-electron stoichiometry of
the reversible redox processes was confirmed by exhaustive
electrolyses of the representative compounds 1a at 0.65 and
1.0 V and 2 at 0.7 and 1.2 V, respectively.
To have an insight into the nature of the electronic levels
associated with the reversible redox processes, electrogenerated
complexes were studied by EPR spectroscopy in CH2CL2/0.1 M
TBAP at 77 K. The one electron oxidized complexes [1a]+ and
[2]+, generated by exhaustive electrolysis of [1a] and [2] at 0.65
and 0.7 V, respectively, displayed ligand-centered sharp single
line spectra at g = 2.006 and g = 2.003, respectively, with peak to
peak line widths of 308 mT to 343 mT, which indicate that each
of the oxidized complexes undergoes ligand based oxidation, and
consequently the unpaired spin is localized primarily on the
coordinated ligand.4,15 The two electron oxidized dipositive
compounds, [1]2+ and [2]2+, are, however, EPR silent, as
expected. EPR spectra of all of the above one electron oxidized
complexes are displayed in Figure S9 of the Supporting Informa-
tion. These results are fully corroborated by density functional
theory (DFT) calculations, which show that the highest occupied
molecular orbitals (HOMO, HOMOꢀ1) of the complexes 1b
and 2 are primarily ligand-centered (cf. below). Thus, one
electron oxidation in these complexes is expected to occur from
the ligand centers and produce monocationic ligand-centered
radical complexes of Re(III), as observed experimentally.
The Electronic Structures and UVꢀVis Spectra. Because of
the redox noninnocent nature of the coordinated ligands, the
assignment of electronic structures of these complexes is an
important issue in view of different electronic structure
possibilities4,15 (Schemes 1 and 6). In order to have a definite
look into the electronic structures of these two complexes 1 and
Geometry optimizations were also carried out on the one- and
two-electron oxidized redox partners of 1b and 2. Optimized
7890
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
Table 5. UVꢀVisꢀNIR Spectral Transition of the Com-
polydentated (tri and higher) azoaromatic substrates. Our stud-
ies in this area are in progress.
plexes 1 and 2
electronic spectra
λmax, nm (ε, Mꢀ1 cmꢀ1
)
’ EXPERIMENTAL SECTION
compound
Materials. Re2(CO)10 is an Aldrich reagent. The ligands L1(a/b) and
L2 were prepared by following the reported procedure.17 All other
chemicals were of reagent grade and used as received. Solvents were
purified and dried prior to use.
[Re(L1a)(L3a)] (1a)
555 (12170), 464 (10873),
390(12776), 334 (17940), 270(22403)
559 (9015), 468 (7785),
[Re(L1b)(L3b)] (1b)
[Re(L2)(L4•ꢀ)2] (2)
390 (9173), 337(14376), 270(18509)
935 (5598), 602 (8458), 490 (20901),
290(18093)
Instrumentation. 1H NMR spectra were taken on a Bruker
Advance DPX 300 spectrometer, and SiMe4 was used as the internal
standard. Infrared spectra were obtained using a Perkin-Elmer 783
spectrophotometer. Cyclic voltammetry was carried out in 0.1 M
Bu4NClO4 solutions using a three-electrode configuration (platinum
working electrode, Pt counter electrode, Ag/AgCl reference) and a PC-
controlled PAR model 273A electrochemistry system. The E1/2 for the
ferroceniumꢀferrocene couple under our experimental conditions was
0.39 V. A Perkin-Elmer 240C elemental analyzer was used to collect
microanalytical data (C, H, N). ESI mass spectra were recorded on a
micro mass Q-TOF mass spectrometer (serial no. YA 263). EPR spectra
in the X band were recorded with a JEOL JES-FA200 spectrometer.
Syntheses. Reaction of Re2(CO)10 and HL1(a/b): Synthesis of
[Re(L1a)(L3a)] (1a). A total of 100 mg of Re2CO10 (0.15 mmol) and
165 mg of HL1a (0.60 mmol) were heated at reflux for 4 days in n-octane.
The color of the solution changed from red to pink during this period.
The solution was filtered, which on evaporation produced a dark mass.
The crude product was then loaded onto a preparative silica TLC plate
for purification. Toluene was used as the eluent. Complex 1a was
separated on a TLC plate as a pink band with several other minor
overlapping products. It was recrystallized by slow diffusion of a
dichloromethane solution of the compound into hexane. Its yields and
characterization data are as follows. Yield: 45%. IR (KBr, cmꢀ1): 1585
[ν(CdN)], 1295 [L1ν(NdN)], 1285 [L2ν(NdN)]. ESI-MS, m/z: 822
[MH]+. Anal. Calcd for C41H31N8Re: C, 59.91; H, 3.80; N, 13.63.
structural parameters and net spin densities for the computed
ground states of the [1b]+, [1b]2+, [2]+, and [2]2+ are collected
in Tables 2 and 3. The one electron oxidized monocationic
complexes, [1b]+ and [2]+, have a doublet ground state, and the
changes in net spin density indicate that the oxidation process
is clearly ligand centered, although in the case of [1b]+, it is
mainly localized on a single ligand (L3b; Re = 0.1857, L1b
=
0.0147, L3b = 0.7996; Figure S11, Supporting Information) while
it is delocalized over the twoaminophenoleto ligands in [2]+ (Re =
ꢀ0.1523, L2 = 0.0445, and L4•ꢀ = 1.1078; Figure S12, Supporting
Information). The two electron oxidized complexes [1b]2+ and
[2]2+ have closed-shell singlet state structures, which again is
consistent with the ligand centered redox process.
Overall, the oxidation states of the metal and the two ligands in
compounds 1b and 2 may be described as [ReIII(L1b)ꢀ(L3b)2ꢀ
]
(1b) and [ReIII(L2)ꢀ{(L4)•ꢀ}2] (2), respectively. The central
Re(III) ion remains essentially invariant in the two redox
processes [1b/2]1+/2, with all redox events localized on the
ligands, which is in line with the EPR data.
All of the compounds 1 and 2 are freely soluble in nonpolar
solvents like dichloromethane and chloroform, and their electro-
nic spectra were recorded in dichloromethane solution. The
experimental spectral results are collected in Table 5, and the
corresponding spectra are displayed in Figure S13 (Supporting
Information). The low intensity tail observed in the experimental
spectrum compares well with the calculated value for the
HOMOfLUMO transition at around 900ꢀ950 nm. The two
midintensity peaks with transition maxima close to 550 (for 1b)
and 600 nm (for 2) can be readily assigned to HOMOꢀ1/
HOMOꢀ2fLUMO transitions, which are computationally
predicted to appear at 592 and 606 nm, respectively. Several
possible transitions were found to lie between 250 and 400 nm,
which are assigned as the intraligand (ligand(π)fligand(π*))
charge transfer transition.4
1
Found: C, 59.87; H, 3.82; N, 13.62. H NMR (500 MHz, d6-DMSO,
300 K): δ 8.15 (d, 1H J = 8.5), 7.62 (d, 1H, J = 8.0 Hz), 7.53 (t, 1H, J1 =
7.0, J2 = 8.0 Hz), 7.41 (t, 1H, J 1= 7, J2 = 8.5 Hz), 7.30ꢀ7.28 (m, 2H),
7.16 (d, 1H, J = 8.0 Hz), 7.10ꢀ7.02 (m, 6H), 6.92 (t, 1H, J1 = 7.5, J2 =
7.5 Hz), 6.77 (t, 1H, J1 = 7.5, J2 = 7.5 Hz), 6.73ꢀ6.69 (m, 7H Hz),
6.65ꢀ6.60 (m, 3H), 6.55 (d, 1H, J = 8.0 Hz), 6.50 (t, 1H, J1 = 7.5, J2 =
7.5 Hz), 6.46 (t, 1H, J1 = 7.5, J2 = 7.0 Hz), 6.39 (d, 1H, J = 7.5 Hz), 6.30
(t, 1H, J1 = 6.5, J2 = 6.5 Hz), 6.23 (d, 1H, J= 8.0 Hz).
Synthesis of [Re(L1b)(L3b)] (1b). This was synthesized following the
same procedure as described for 1a. Its yield and characterization data
are as follows. Yield: 42%. IR (KBr, cmꢀ1): 1585 [ν(CdN)], 1280
[L1ν(NdN)], 1275 [L2ν(NdN)]. ESI-MS, m/z: 864 [MH]+. Anal.
Calcd for C44H37N8Re: C, 61.16; H, 4.32; N, 12.97. Found: C, 61.14; H,
4.34; N, 12.93. 1H NMR (500 MHz, d6-DMSO, 300 K): δ 8.12 (d, 1H
J = 8.0 Hz), 7.58 (d, 1H, J = 8.0 Hz), 7.50ꢀ7.49 (m, 2H), 7.42 (t, 1H, J1
= 7.5, J2 = 8.0 Hz), 7.34 (d, 1H, J = 8.5 Hz), 7.28 (d, 1H, J = 6.0 Hz), 7.14
(d, 2H, J = 8.0 Hz), 7.08 (d, 1H, J = 8.0 Hz), 7.02 (t, 1H, J1= 8.0, J2= 8.5
Hz), 6.77 (d, 1H, J = 7.5 Hz), 6.73ꢀ6.67 (m, 6H), 6.60 (t, 1H, J1= 7.5,
J2=7.5), 6.53ꢀ6.49 (m, 5H), 6.43 (t, 1H, J1 = 7.5, J2 = 8 Hz), 6.27 (t,
1H, J1= 6.5, J2= 7.0), 6.21 (d, 1H, J= 6.0 Hz), 6.13 (d, 1H, J = 8.0 Hz).
Reaction of Re2(CO)10 and HL2: Synthesis of [Re(L2){(L4)•ꢀ}2] (2).
A total of 100 mg of Re2CO10 (0.15 mmol) and 190 mg of HL2
(0.61 mmol) were heated at reflux for 96 h in n-octane. The color of the
solution changed from orange to dull green during this period. The
solution was filtered, which on evaporation produced a dark mass. The
crude product was then loaded onto a preparative silica TLC plate for
purification. Chloroform was used as the eluent. Complex 2 was
separated on a TLC plate as a dull green band as a major product with
some other minor overlapping products. It was recrystallized by slow
evaporation of a methanolꢀdichloromethane solvent mixture of the
’ CONCLUSION
In conclusion, we have introduced an unprecedented type of
multiple step redox process in Re-coordinated azoaromatic
ligands. Attempts have been made to rationalize the formation
of H2L3 from HL1. Considering the products of all of the above-
noted reactions together, one may think of the most simplest
pathway of four electron reductive azo cleavage followed by
oxidation of one of the imide fragments, leading to a symmetrical
azo-aromatic ligand [H2L3]. It is thus concluded that dimetallic
electron-rich mediators can cleave and reform azo bond(s) only
on multidented azo-aromatics. The results have generated a
scope of synthesizing otherwise inaccessible extended new azo-
aromatic compounds by reactions of multimetalꢀcarbonyls with
7891
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
complex. Its yields and characterization data are as follows. Yield: 60%.
IR (KBr, cmꢀ1): 1580 [ν(CdN)], 1260 [L1ν(NdN)]. ESI-MS, m/z:
936 [MH]+. Anal. Calcd for C47H66N5O3Re: C, 60.36; H, 7.11; N, 7.49.
Found: C, 60.32; H, 7.15; N, 7.45. 1H NMR (500 MHz, CDCl3, 300 K):
δ 9.32 (s, 1H (NꢀH)), 8.98 (s, 1H (NꢀH)), 7.84 (s, 1H), 7.79ꢀ7.58
(m, 2H), 7.48 (t, 1H, J1 = 9, J2 = 8.5 Hz), 7.41 (t, 1H, J1 = 8.5, J2 =
8.0 Hz), 7.06 (s, 1H), 6.98 (s, 1H), 6.78 (s, 1H), 6.73 (s, 1H Hz), 6.66 (s,
1H Hz), 1.45ꢀ1.23 (s, 54H).
’ ASSOCIATED CONTENT
S
Supporting Information. Experimental and characteriza-
b
tion details, X-ray crystallographic files in CIF format, and
ORTEP and respective atom numbering schemes for 1ꢀ3 are
provided. This material are available free of charge via the
Internet at http://pubs.acs.org.
Reaction of Re(CO)5Cl and HL1: Synthesis of [Re(pap)(CO)3Cl] (3). A
total of 100 mg of Re(CO)5Cl (0.27 mmol) and 150 mg of HL1 (0.54
mmol) was heated at reflux for 3 days in n-octane. The color of the
solution changed from red to deep blue during this period. The solution
was filtered, which on evaporation produced a dark mass. Recrystalliza-
tion of the crude product from dichloromethane/hexane solution
produced complex 3. Its yield and characterization data are as follows.
Yield: 75%. IR (KBr, cmꢀ1): 1580 [ν(CdN)], 1290 [Lν(NdN)], 1905
[ν(CtO)]. ESI-MS, m/z: 489 [MH]+. Anal. Calcd for C14H9-
ClN3O3Re: C, 34.39; H, 1.86; N, 8.59. Found: C, 34.37; H, 1.89;
N, 8.58.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: icsg@iacs.res.in.
’ ACKNOWLEDGMENT
The research was supported by Department of Science and
Technology, New Delhi (Project: SR/S1/IC/0031/2010). Crys-
tallography was performed at the DST-funded National Single
Crystal Diffractometer Facility at the Department of Inorganic
Chemistry, IACS. N.D.P. and S.S. thank the Council of Scientific
and Industrial Research for fellowship support.
Reaction of Re2(CO)10 and L5: Synthesis of [Re(L5•ꢀ)2(CO)2] (4). A
total of 100 mg of Re2CO10 (0.15 mmol) and 165 mg of HL1a (0.61
mmol) were heated at reflux for 96 h in n-octane. The color of the
solution changed from orange to deep blue during this period. The crude
complex was crystallized from a hot n-octane solution of the reaction
mixture. Its yield and characterization data are as follows. Yield: 75%. IR
(KBr, cmꢀ1): 1195 [ν(NdN)]; 1875, 1940 [ν(CtO)]. ESI-MS, m/z:
791 [MH]+. Anal. Calcd for C36H28N8O2Re: C, 54.67; H, 3.57; N,
14.17. Found: C, 54.63; H, 3.60; N, 14.14.
’ REFERENCES
(1) (a) Handbook of Metathesis; Grubbs, R. H., Ed.; Wiley-VCH:
Weinheim, Germany, 2003; Vols 1ꢀ3. (b) For a major account of Ru
metathesis catalysts, see: Nguyen, S. T.; Trnka, T. M. In Handbook of
Metathesis; Grubbs, R. H., Ed.; Wiley-VCH: Weinheim, Germany, 2003;
Vol. 1, chapter 1.6. (c) Trnka, T. M; Grubbs, R. H. Acc. Chem. Res. 2001,
34, 18 and references therein. (d) Astruc, D. New J. Chem. 2005, 29, 42
and references therein.
(2) (a) Basset, J.-M.; Copꢀeret, C.; Soulivong, D.; Taoufik, M.; Cazat,
J. T. Acc. Chem. Res. 2010, 43, 323. (b) Alcaide, B.; Almendros, P.; Luna,
A. Chem. Rev. 2009, 109, 3817.
(3) (a) Grirrane, A.; Corma, A.; García, H. Science 2008, 322, 1661.
(b) Zhang, C.; Jiao, N. Angew. Chem., Int. Ed. 2010, 49, 6174.(c) Hunger,
K. industrial dyse: Chemistry, properties, Applications; Wiley-VCH:
Weinheim, Germany, 2003. (d) Andersosn, R. G.; Nickless, G. Analyst
1967, 92, 25. (e) Ashutosh, P. N. D.; Mehrotra, J. K. Colourage 1979,
26, 25. (f) Athey, R. D., Jr. Eur. Coatings J. 1998, 3, 146. (g) Sheppard,
C. S. Encycl. Polym. Sci. Eng. 1985, 2, 143. (h) Hoult, J. R. S. Drugs 1986,
97, 2939. (i) Sandborn, W. J. Am. J. Gastroenterol. 2002, 97, 2939.(j)
Cation, S. C.; Farris, E. Consise Encyclopedia of Chemical Technilogy;
Wiley: New York, 1985.
(4) (a) Paul, N. D.; Samanta, S.; Goswami, S. Inorg. Chem. 2009,
49, 2649. (b) Sanyal, A.; Banerjee, P.; Lee, G.-H.; Peng, S.-M.; Hung, C.-
H.; Goswami, S. Inorg. Chem. 2004, 43, 7456. (c) Sanyal, A; Chatterjee,
S; Casti~neiras, A; Sarkar, B; Singh, P; Fiedler, J; Zꢀaliꢁs, S; Kaim, W;
Goswami, S. Inorg. Chem. 2007, 46, 8584. (d) Samanta, S.; Singh, P.;
Fiedler, J.; Zꢀaliꢁs, S.; Kaim, W.; Goswami, S. Inorg. Chem. 2008, 47, 1625.
(e) Sanyal, A.; Banerjee, P.; Lee, G.-H.; Peng, S.-M.; Hung, C.-H.;
Goswami, S. Inorg. Chem. 2004, 43, 7456.
EPR Spectral Studies. The one electron oxidized complexes [1]+
and [2]+, generated by exhaustive electrolysis of the respective com-
plexes 1 and 2 at 0.65 and 0.7 V, respectively, in CH2Cl2/0.1 M TBAP,
were immediately dipped into liquid nitrogen, and the resulting frozen
solutions were used for the EPR measurements at 77 K. The applied
potential for the two electron oxidized complexes [1]2+ and [2]2+ are 1.0
and 1.2 V, respectively.
Crystallography. Crystallographic data for compounds 1b, 2, and
3 are collected in Table S-1 (Supporting Information). Suitable X-ray
quality crystals of these are obtained as follows: 1b, by slow evaporation
of a dichloromethaneꢀhexane solvent mixture of the compound; 2, by
slow evaporation of a methanolꢀdichloromethane solution of the
compound; and 3, by slow evaporation of dichloromethaneꢀhexane
solution of the compound.
All data were collected on a Bruker SMART APEX-II diffractometer,
equipped with graphite monochromated Mo KR radiation (λ = 0.71073
Å), and were corrected for Lorentz-polarization effects. 1b: A total of
43 806 reflections were collected, of which 9605 were unique (Rint
=
0.063), satisfying the (I > 2σ(I)) criterion, and were used in subsequent
analysis. 2: A total of 26 869 reflections were collected, of which 8364
were unique (Rint = 0.057). 3: A total of 16 378 reflections were
collected, of which 2484 were unique (Rint = 0.082).
The structures were solved by employing the SHELXS-97 program
package18 and were refined by full-matrix least-squares based on F2
(SHELXL-97).19 All hydrogen atoms were added in calculated
positions.
(5) Sarkar, B.; Patra, S.; Fiedler, J.; Sunoj, R. B.; Janardanan, D.;
Lahiri, G. K.; Kaim, W. J. Am. Chem. Soc. 2008, 130, 3532 and references
therein. (g) Doslik, N.; Sixt, T.; Kaim, W. Angew. Chem., Int. Ed. 1998,
37, 2403. (h) Kaim, W. Coord. Chem. Rev. 2001, 219, 463.
(6) (a) Sinan, M.; Panda, M.; Banerjee, P.; Shinisha, C. B.; Sunoj,
R. B.; Goswami, S. Org. Lett. 2009, 11, 3219 and references therein.
(7) (a) Laplaza, C. E.; Cummins, C. C. Science 1995, 268, 861. (b)
Sadique, R. A.; Gregory, A. E.; Brennessel, W. W.; Holland, L. J. Am.
Chem. Soc. 2007, 129, 8112. (c) Lentz, M. R.; Vilardo, J. S.; Lockwood,
M. A.; Fanwick, P. E.; Rothwell, I. P. Organometallics 2004, 23, 329 and
references therein. (d) Peters, R. G.; Warner, B. P.; Burns, C. J. J. Am.
Chem. Soc. 1999, 121, 5585 and references therein. (e) Evans, W. J.;
Kozimor, S. A.; Ziller, J. W. Chem. Commun. 2005, 4681. (f) Ohki, Y.;
Takikawa, Y.; Hatanaka, T.; Tatsumi, K. Organometallics 2006, 25, 3111.
(g) Pan, C.-L.; Chen, W; Song, S.; Zhang, H.; Li, X. Organometallics
Computational Details. All calculations described in this paper
were performed using density functional theory as implemented in the
Gaussian 03 package.20 Full geometry optimizations were performed
without symmetry constraints. The vibrational frequency calculations were
performed to ensure that the optimized geometries represent the local
minima and that there are only positive Eigen values. The hybrid B3LYP
exchange-correlation functional was used in conjunction with the SDD
basis set with effective core potential for the Re atom and the 6-31G(d)
basis set for C, H, N, and O atoms. GaussSum21 was used to calculate the
fractional contributions of various groups to each molecular orbital.
7892
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893

Inorganic Chemistry
ARTICLE
2009, 48, 6344. (h) Lahiri, G. K.; Goswami, S.; Falvello, L.; Chakravorty,
A. Inorg. Chem. 1987, 26, 3365. (i) Ghosh, A. K.; Majumdar, P.; Falvello,
R. L.; Mostafa, G.; Goswami, S. Organometallics 1999, 18, 5086.
(8) Zarkesh, R. A.; Ziller, J. W.; Heyduk, A. F. Angew. Chem., Int. Ed.
2008, 47, 4715.
(9) Xu, L.; Setyawati, I. A.; Pierreroy, J.; Pink, M.; Young, V. G., Jr.;
Patrick, B. O.; Rettig, S. J.; Orvig, C. Inorg. Chem. 2000, 39, 5958 and
references therein.
(10) (a) Saha, A.; Ghosh, A. K.; Majumder, P.; Mitra, K. N.; Mondal,
S.; Rajak, K. K.; Falvello, L. R.; Goswami, S. Organometallics 1999,
18, 3772. (b) Saha, A.; Majumder, P.; Goswami, S. J. Chem. Soc., Dalton
Trans. 2000, 11, 1703. (c) Saha, A.; Majumder, P.; Peng, S.-M.;
Goswami, S. Eur. J. Inorg. Chem. 2000, 12, 2631.
(11) (a) Cotton, F. A.; Duraj, S.; Roth, W. J. Am. Chem. Soc. 1984,
106, 4749. (b) Hill, J. E.; Profilet, R. D.; Fanwick, P. E.; Rothwell, I. P.
Angew. Chem., Int. Ed. Engl. 1990, 29, 664. (c) Hill, J. E.; Fanwick, P. E.;
Rothwell, I. P. Inorg. Chem. 1991, 30, 1143. (d) Zambrano, C. H.;
Fanwick, P. E.; Rothwell, I. P. Organometallics 1994, 13, 1174.
(12) Sengupta, S.; Panda, B. K. Transition Met. Chem. 2005, 30, 426.
(13) Saha, A.; Das., C.; Goswami, S.; Peng, S.-M. Indian J. Chem.,
Sect. A 2001, 40A, 198.
(14) Paul, N. D.; Kr€amer, T.; Mcgrady, J. E.; Goswami, S. Chem.
Commun 2010, 46, 7124.
(15) (a) Chun, H.; Bill, E.; Weyherm€uller, T.; Wieghardt, K. Inorg.
Chem. 2003, 42, 5612 and reference therein. (b) Sproules, S.; Benedito,
F. L.; Bill, E.; Weyherm€uller, T.; George, S. D.; Wieghardt, K. Inorg.
Chem. 2009, 48, 10926. (c) Bhattacharya, S.; Gupta, P.; Basuli, F.;
Pierpont, C. G. Inorg. Chem. 2002, 41, 5810.
(16) (a) Das, C.; Ghosh, A. K.; Hung, C.-H.; Lee, G.-H.; Peng, S.-M.;
Goswami, S. Inorg. Chem. 2002, 41, 7125. (b) Das, C.; Saha, A.; Hung,
C.-H.; Lee, G.-H.; Peng, S.-M.; Goswami, S. Inorg. Chem. 2003, 42, 198.
(c) Kamar, K. K.; Saha, A.; Casti~neiras, A.; Hung, C.-H.; Goswami, S.
Inorg. Chem. 2002, 41, 4531.
(17) Campbell, N.; Henderson, A. W.; Taylor, D. J. Chem. Soc.
1953, 1281. (b) Anderson, R. G.; Nickless, G. Anal. Chim. Acta 1967,
39, 469.
(18) Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467.
(19) Sheldrick, G. M. SHELXL 97; University of G€ottingen;
G€ottingen, Germany, 1997.
(20) Frisch, M. J. et al. Gaussian 03; Gaussian, Inc.: Wallingford, CT,
2004.
(21) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.
Chem. 2008, 29, 839.
7893
dx.doi.org/10.1021/ic201200r |Inorg. Chem. 2011, 50, 7886–7893