Diiron(III) complexes of tridentate 3N ligands as functional models for methane monooxygenases: Effect of the capping ligand on hydroxylation of alkanes

Article abstract of DOI:10.1016/j.poly.2013.08.067

A series of non-heme (μ-oxo)bis(μ-benzoato)-bridged diiron(III) complexes of the type [Fe₂(O)(OBz)₂(L)₂] ²⁺ (1-6), where OBz = benzoate, L = bis(pyridin-2-ylmethyl)amine (L1), N-((6-methylpyridin-2-yl-)methyl)(pyridin-2-yl)methanamine (L2), N,N-dimethyl-N′-(pyrid-2-ylmethyl)ethylenediamine (L3), (1-methyl-1H-imidazol-2-yl)-N-(pyridin-2-ylmethyl)-methanamine (L4), (1H-benzo[o]imidazol-2-yl)-N-(pyridin-2-ylmethyl)methanamine (L5) and bis((1H-benzo[o]imidazol-2-yl)methyl)amine (L6), have been isolated and characterized by means of elemental analysis, spectral and electrochemical methods. They have been studied as catalysts for selective hydroxylation of alkanes using m-choloroperbenzoic acid (m-CPBA) as the oxidant. In acetonitrile/dichloromethane mixed solvent all the complexes display a d-d band characteristic of a triply bridged diiron(III) core, revealing that they retain their identity in solution. Upon replacing a donor atom on the capping ligand by a stronger donor, the E_1/2 value of the one-electron Fe ^IIIFe^III → Fe^IIIFe^II reduction becomes more negative. All the complexes function as efficient catalysts for hydroxylation of cyclohexane, with 390-410 total turnover numbers and good alcohol selectivity (A/K, 9.3-12.8). Adamantane is selectively oxidized (3 /2, 15.7-28.1) to 1-adamantanol and 2-adamantanol, along with a small amount of 2-adamantanone (Total TON, 336-437), and interestingly the 3N capping ligands with the pyridyl donor around the diiron(III) center lead to high 3 /2 bond selectivity.

Full text of DOI:10.1016/j.poly.2013.08.067

Polyhedron 67 (2014) 171–180
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Diiron(III) complexes of tridentate 3N ligands as functional models
for methane monooxygenases: Effect of the capping ligand
on hydroxylation of alkanes
Muniyandi Sankaralingam a, Mallayan Palaniandavar b,
⇑
a School of Chemistry, Bharathidasan University, Tiruchirappalli 620024, Tamil Nadu, India
b Department of Chemistry, Central University of Tamil Nadu, Thiruvarur 610004, Tamil Nadu, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 6 July 2013
Accepted 15 August 2013
Available online 12 September 2013
A series of non-heme (l-oxo)bis(l-benzoato)-bridged diiron(III) complexes of the type [Fe2(O)(OBz)2
(L)2]2+ (1–6), where OBz = benzoate, L = bis(pyridin-2-ylmethyl)amine (L1), N-((6-methylpyridin-2-yl-)
methyl)(pyridin-2-yl)methanamine (L2), N,N-dimethyl-N0-(pyrid-2-ylmethyl)ethylenediamine (L3),
(1-methyl-1H-imidazol-2-yl)-N-(pyridin-2-ylmethyl)-methanamine (L4), (1H-benzo[o]imidazol-2-yl)-
N-(pyridin-2-ylmethyl)methanamine (L5) and bis((1H-benzo[o]imidazol-2-yl)methyl)amine (L6), have
been isolated and characterized by means of elemental analysis, spectral and electrochemical methods.
They have been studied as catalysts for selective hydroxylation of alkanes using m-choloroperbenzoic
acid (m-CPBA) as the oxidant. In acetonitrile/dichloromethane mixed solvent all the complexes display
a d–d band characteristic of a triply bridged diiron(III) core, revealing that they retain their identity in
solution. Upon replacing a donor atom on the capping ligand by a stronger donor, the E1/2 value of the
one-electron FeIIIFeIII ? FeIIIFeII reduction becomes more negative. All the complexes function as efficient
catalysts for hydroxylation of cyclohexane, with 390–410 total turnover numbers and good alcohol selec-
tivity (A/K, 9.3–12.8). Adamantane is selectively oxidized (3°/2°, 15.7–28.1) to 1-adamantanol and 2-ada-
mantanol, along with a small amount of 2-adamantanone (Total TON, 336–437), and interestingly the 3N
capping ligands with the pyridyl donor around the diiron(III) center lead to high 3°/2° bond selectivity.
Ó 2013 Elsevier Ltd. All rights reserved.
Keywords:
Diiron(III) complexes
Tridentate 3N ligands
Alkane hydroxylation
Iron-oxo intermediate
1. Introduction
hand, the reduced form of sMMO possesses a diiron(II) center con-
taining four glutamate and two histidine residues, and the iron
The selective hydroxylation of hydrocarbons under mild condi-
tions is an exciting and challenging scientific goal in bioinorganic
and synthetic organic chemistry [1] because conventional hydrox-
ylation processes usually require high pressures and high temper-
atures, while enzymes catalyze these reactions very efficiently
with high selectivity under mild reaction conditions. This indicates
that enzymatic reactions follow methodologies different from
those of traditional synthetic processes. In nature, iron containing
enzymes, like methane monooxygenases, bleomycin and cyto-
chrome P450, play a vital role in catalyzing many biologically nec-
essary organic transformations. Particularly, soluble methane
monooxygenases (sMMO), which catalyze the oxidation of meth-
ane to methanol using dioxygen, are widely investigated metallo-
enzymes [2–11]. Thus the active site of the oxidized form of
sMMO possesses a diiron(III) center consisting of only one carbox-
ylate and two hydroxo bridges (sMMOox, Scheme 1). On the other
atoms are bridged by two carboxylate ligands from glutamate res-
idues (sMMOred, Scheme 1) [12–14]. It is interesting that a num-
ber of non-heme diiron proteins, such as hemerythrin (Hr) [15]
and ribonucleotide reductases (RNR) [16–18], also contain at least
two carboxylate bridges in the active site of their reduced forms,
but they exhibit differences in their functional aspects. Interest-
ingly, the bridging units also play a vital role in dictating the func-
tions of the enzyme. So, the isolation and study of synthetic models
containing the structural motif [Fe2(O)(O2CR)2]2+ have received
considerable interest amongst chemical and biochemical commu-
nities [19–23]. Also, a wide variety of coordination environments
and bridging modes around the iron center in non-heme enzymes
generate distinct oxidizing intermediates, which are supposed to
result in their native catalytic transformations [11,24–29]. Encour-
aged by these enzymes, efforts have been made during the last two
decades to develop synthetic models for sMMO in order to study
the mechanism of O2 activation involved in the oxygenation of
alkanes [30,31]. The study of synthetic diiron complexes with
different carboxylate motifs has mainly served as benchmarks for
⇑
Corresponding author. Tel.: +91 9489079402; fax: +91 4366225312.
E-mail address: palanim51@yahoo.com (M. Palaniandavar).
0277-5387/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2013.08.067

172
M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
Glu243
Glu243
O
H
H
O
O
O
O
O
H
O
H
O
O
O
O
O
H
O
Glu114
His147
Glu209
His246
FeII
FeII
O
O
Glu114
His147
Glu209
His246
O
FeIII
FeIII
H2
O
H
N
N
O
O
N
N
O
O
His246
N
N
Glu144
sMMOred
H
H
N
N
H
Glu144
sMMOox
H
Scheme 1. Active site structures of soluble methane monooxygenases.
illustrating the oxidizing intermediates and catalytic pathway of
the enzymes, and efforts to make use of them as functional mimics
are fewer.
the bridging carboxylate group. Encouraged by the above reactions,
we have now isolated a few more ( -oxo)bis( -benzoato)diiron(III)
l
l
complexes, [Fe2(O)(OBz)2(L)2](ClO4)2 (1–6) where L is a tridentate
3N ligand, namely bis(pyridin-2-ylmethyl)-amine (L1), N-((6-
methylpyridin-2-yl-)methyl)(pyridin-2-yl)methanamine (L2), (1-
methyl-1H-imidazol-2-yl)-N-(pyridin-2-ylmethyl)-methanamine
(L3), (1H-benzo[o]imidazol-2-yl)-N-(pyridin-2-ylmethyl)methan-
amine (L4), N,N-dimethyl-N0-(pyrid-2-ylmethyl)-ethylenediamine
(L5) and bis((1H-benzo[o]imidazol-2-yl)methyl)amine (L6)
(Scheme 2), as new synthetic analogues for sMMO and studied the
steric and electronic effects of the 3N ligands on the highly selective
hydroxylation of alkanes.
Several synthetic models containing the structural motif
[Fe2(O)(O2CR)2(L)2]2+, where L is a tridentate 3N ligand, have been
isolated and studied as structural and functional models for metHr
and sMMO [32–35]. Christou et al. [36,37] synthesized triply-
bridged diiron(III) complexes derived from bipyridine for alkane
hydroxylation, using tert-butylhydroperoxide (t-BuOOH) as the
oxygen source, while Kitajima et al. [38] have isolated triply-
bridged diiron(III) complexes resulting from the trispyrozolylb-
orate ligand for alkane hydroxylation using molecular oxygen (in
the presence of an electron source), but low selectivity and poor
yields were observed for these reactions.
The single-crystal X-ray structure of 5 (Fig. 1) has been reported
[51] by our laboratory and that of 6 has been reported earlier [52].
They show that each iron atom in the complexes is coordinated to
Later, Kodera et al. [39–41] found that the complex [Fe2(O)(O2-
CCH3)2(hexpy)](ClO4)2, where hexpy is 1,2-bis [2-(bis(pyrid-2-yl-
one oxygen atom of the
l-oxo bridge and two oxygen atoms of
methyl))-6-pyridyl]ethane, exhibits
a
high activity with
a
both -carboxylato bridges, in addition to three nitrogen atoms
l
remarkable alcohol/ketone ratio (A/K, 2.4) for the hydroxylation of
cyclohexane and tertiary/secondary carbon ratio (3°/2°, 4.2) for
monooxygenation of adamantane using m-chloroperbenzoic acid
(m-CPBA). Recently, Itoh et al. [42] isolated the triply-bridged diiro-
n(III) complex [Fe2(O)(O2CCH3)2(L)](ClO4)2, where L is the dinucleat-
of the 3N ligand. We have observed that all six complexes show
efficient hydroxylation of cyclohexane (290–410 TON) and ada-
mantane with approximately 335–440 turnover numbers and good
selectivity (A/K ratio, 9.3–12.8; 3°/2°, 15.7–28.1), as tuned, inter-
estingly, by varying one or both terminal nitrogen atoms of the
capping ligand from the weakly coordinating –NMe2 donor to the
strongly coordinating pyridyl, imidazolyl and benzimidazolyl
nitrogen donors.
ing
carboxylate
ligand
1,2-bis(N-benzyl-2-aminomethyl-6-
pyridyl)ethane-N’,N’-diacetic acid, and observed higher selectivity
for the hydroxylation of cyclohexane (A/K, 10.0) and monooxygen-
ationofadamantane (3°/2°, 13.6)using hydrogenperoxideasthe oxy-
gen source. However, the reasons for the high selectivity, efficiency
andthemodeofactionofthesecomplexesinalkanehydroxylationre-
main unclear. Also, many oxo-bridged diiron(III) complexes of tetra-
dentate ligands have been isolated [43–46,37,47–49] as synthetic
models fordiiron(III)biosites, but the catalytic efficiency and selectiv-
ity of these systems are lower than those of the enzymes.
2. Experimental
2.1. Materials
Pyridine-2-carboxaldehyde, 2-aminomethylpyridine, N,N-dimeth-
ylethylenediamine, 6-methylpyridine-2-carboxaldehyde, N-methyl-
imidazole-2-carboxaldehyde, iminodiaceticacid iron(III) perchlorate
hydrate, adamantane, m-chloroperbenzoic acid, sodium borohydride,
tetra-N-butylammonium bromide (Aldrich), o-phenylenediamine, tri-
ethylamine, benzoic acid, glacial acetic acid, hydrochloric acid (Merck,
India), cyclohexane (Ranbaxy) and ethanol (Hayman Limited, England)
were used as received. Dichloromethane, diethylether, tetrahydrofu-
All of the above observations prompted us to isolate diiron(III)
complexes of tridentate (3N) ligands with various carboxylate
bridging ligands and to study their use as functional models for
alkane hydroxylation reactions catalyzed by sMMO. Very recently,
we have reported [50] the triply-bridged diiron(III) complexes
[Fe2O(iBu-bpa/Bz-bpa)2(RCOO)2]2+
(pyrid-2-ylmethyl)-iso-butylamine,
,
where
Bz-bpa = N,N-bis(pyrid-2-
iBu-bpa = N,N-bis
ylmethyl)benzylamine, R = –CH3 and –C6H5, which act as efficient
catalysts (A/K, 10.2-13.8) towards alkane hydroxylation. This
H
N
H
N
H
N
N
encouraged us to isolate very recently [51] a few more (
l-oxo)-
bis( -carboxylato)diiron(III) complexes, [Fe2(O)(OOCR)2(L)2]
l
N
N
N
N
N
N
(ClO4)2 where L = N,N-dimethyl-N0-(pyrid-2-ylmethyl)-ethylenedi-
amine and R = HCOOꢀ, CH3COOꢀ, (CH3)3C-COOꢀ, PhCOOꢀ, (Ph)2-
CHCOOꢀ and (Ph)3CCOOꢀ, which exhibit high activity with a
notable alcohol/ketone ratio (A/K, 6.0-10.1) for the hydroxylation
of cyclohexane and tertiary/secondary carbon ratio (3°/2°, 12.9–
17.1) for monooxygenation of adamantane using m-chloroperben-
zoic acid (m-CPBA), and interestingly, the activity depends upon
L3
L2
L1
H
H
N
H
N
H
N
H
N
H
N
N
N
N
N
N
N
N
L4
L5
L6
Scheme 2. Tridentate 3N ligands used in the study.

M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
173
2.2.4. (1H-benzo[o]imidazol-2-yl)-N-(pyridin-2-ylmethyl)-
methanamine (L4)
The linear tridentate ligand L4 was synthesized by neutralizing
2-(aminomethyl)benzimidazole dihydrochloride [57,58] with
K2CO3 and then condensing it with the corresponding pyridine-2-
carboxaldehyde to form a Schiff base, followed by the reduction
of the latter with sodium borohydride [53,54]. Yield: 1.32 g
(55.4%); 1H NMR (400 MHz, CDCl3) d (ppm): 8.52-7.32 (m, 8H),
+
4.04 (s, 2H), 3.82 (s, 2H). EI-MS m/z: 238 C14H14N4
.
2.2.5. N,N-dimethyl-N0-(pyrid-2-ylmethyl)ethylenediamine (L5)
The ligand L5 was synthesized using the same procedure
[53,54] as followed for L1. The amine N,N-dimethylethylenedi-
amine (0.927 g, 10 mmol) was used to obtain L5 as a yellow oil,
which was used for complex preparation without further purifica-
tion. Yield: 1.12 g (82%). 1H NMR (400 MHz, CDCl3) d (ppm): 2.17
(s, 6H), 2.41 (t, 2H), 2.62 (t, 2H), 3.92 (s, 2H), 7.14 (t, H), 7.53 (d,
ꢁ+
Fig. 1. ORTEP view of [Fe2O(L5)2(OBz)2]2+ 5 taken from [51] (40% probability factor
H), 7.64 (t, H), 8.56 (d, H). EI-MS m/z: 179.14 C10H17N3
.
for the thermal ellipsoids). Hydrogen atoms have been omitted for clarity.
2.3. Isolation of the diiron(III) complexes
ran, acetonitrile (Merck, India) and methanol (Sisco Research Labo-
ratory, Mumbai) were distilled before use. The supporting electro-
lyte tetra-N-butylammonium perchlorate (TBAP) was prepared in
water and recrystallized twice from aqueous ethanol.
The general procedure involves the addition of carboxylic acids
(2.0 mmol), previously neutralized with 2.0 mmol of triethylamine,
to a mixture of Fe(ClO4)3ꢁ6H2O (0.924 g, 2.0 mmol) in methanol
(5 mL) containing H2O (0.1 mL) and then stirring for 30 min at
room temperature. To this, a methanol (8 mL) solution of the li-
gand (2.0 mmol) was added and the resulting dark red solution
turned reddish brown or green upon stirring for an hour. Reddish
brown or green precipitate or microcrystals of the diiron(III) com-
plexes were formed within 24 h upon standing. The precipitate
was collected by suction filtration, washed with small quantities
of cold methanol and then dried in vacuo.
2.2. Synthesis of the ligands
The ligands L1-L5 were prepared by using the procedures re-
ported already [53–58]. The ligand L6 was prepared by the Phillips
condensation reaction [59].
2.3.1. [Fe2(O)(OBz)2(L1)2](ClO4)2 (1)
2.2.1. Bis(pyridin-2-ylmethyl)amine (L1) [53,54]
Complex 1 was prepared using the above general procedure
using L1 (0.199 g, 1.0 mmol). Yield: 0.76 g (79%). ESI-MS, m/z:
866.87 [(Mꢀ2HꢀClO4)+], 744.87 [(Mꢀ2HꢀBzO–ClO4)+], 383.93
[(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc. for C38H38Fe2N6O13Cl2: C, 47.08; H,
3.95; N, 8.67. Found: C, 47.12; H, 3.91 N, 8.62%.
Pyridine-2-carboxaldehyde (1.07 g, 10 mmol) in methanol
(20 mL) was added dropwise to aminomethylpyridine (1.08 g,
10 mmol) in methanol (20 mL). The mixture was stirred overnight
and then NaBH4 (0.57 g, 15 mmol) was added. The solution was
stirred for another day and then rotary evaporated to dryness.
The resulting solid was dissolved in water and then extracted with
dichloromethane. The organic layer was dried with anhydrous so-
dium sulfate and then rotary evaporated to get the ligand as an oil.
2.3.2. [Fe2(O)(OBz)2(L2)2](ClO4)2 (2)
Complex 2 was prepared using the above general procedure
using L2 (0.213 g, 1.0 mmol). Yield: 0.73 g (73%). ESI-MS, m/z:
895.12 [(Mꢀ2HꢀClO4)+], 398.07 [(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc.
for C40H42Fe2N6O13Cl2: C, 48.17; H, 4.24; N, 8.43. Found: C,
48.12; H, 4.19; N, 8.39%.
Yield: 1.72 g (86.4%); 1H NMR (400 MHz, CDCl3) d (ppm): 8.48–
7.29 (m, 8H), 4.12 (s, 4H). EI-MS m/z: 199 C12H13N3
+
.
2.2.2. N-((6-methylpyridin-2-yl-)methyl)(pyridin-2-yl)methanamine
(L2)
2.3.3. [Fe2(O)(OBz)2(L3)2](ClO4)2 (3)
Complex 3 was prepared using the above general procedure
using L3 (0.202 g, 1.0 mmol). Yield: 0.76 g (78%). ESI-MS, m/z:
872.53 [(Mꢀ2HꢀClO4)+], 751.00 [(Mꢀ2HꢀBzOꢀClO4)+], 387.00
[(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc. for C36H40Fe2N8O13Cl2: C, 44.33; H,
4.13; N, 11.49. Found: C, 44.39; H, 4.15; N, 11.43%.
The ligand L2 was synthesized using a modified procedure
[53,54]. The aldehyde 6-methylpyridine-2-carboxaldehyde
(1.23 g, 10 mmol) was used to obtain the L2 as a yellow oil, which
was used for complex preparation without further purification.
Yield: 1.48 g (69.4%); 1H NMR (400 MHz, CDCl3) d (ppm): 8.48–
+
7.29 (m, 7H), 4.12 (s, 4H), 2.61 (s, 3H). EI-MS m/z: 213 C13H15N3
.
2.3.4. [Fe2(O)(OBz)2(L4)2](ClO4)2 (4)
Complex 4 was prepared using the above general procedure
using L4 (0.238 g, 1.0 mmol). Yield: 0.65 g (62%). ESI-MS, m/z:
945.07 [(Mꢀ2HꢀClO4)+], 423.07 [(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc.
for C42H40Fe2N8O13Cl2: C, 48.16; H, 3.85; N, 10.70. Found: C,
48.19; H, 3.81; N, 10.74%.
2.2.3. (1-Methyl-1H-imidazol-2-ylmethyl)pyridin-2-ylmethylamine
(L3)
The ligand L3 was synthesized using the same procedure [53–
56] as followed for preparing L1. The aldehyde N-methyl-2-imi-
dazolecarboxaldehyde (1.124 g, 10 mmol) was used to obtain L3
as a yellow oil, which was used for complex preparation without
further purification. Yield: 1.27 g (62.9%); 1H NMR (400 MHz,
2.3.5. [Fe2(O)(OBz)2(L5)2](ClO4)2 (5)
Complex 5 was prepared as reported [51] earlier. This complex
was already well characterized structurally and it possesses a dis-
torted bioctahedral geometry in which each iron atom is coordi-
CDCl3) d (ppm): 8.44-7.26 (m, 4H), 6.72 (s, 1H), 6.62 (s, 1H), 4.14
+
(s, 2H), 3.81 (s, 2H), 3.63 (s, 3H). EI-MS m/z: 202 C11H14N4
.
nated to the oxygen atom of the l-oxo bridge, two oxygen atoms

174
M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
of
l
-carboxylate bridges and three nitrogen atoms of the 3N ligand,
chemical cell was maintained by a cryo-circulator (HAAKE D8-G).
The E1/2 value observed under identical conditions for the Fc/Fc+
couple in acetonitrile was 0.102 V with respect to the Ag/Ag+ refer-
ence electrode. The experimental solutions were deoxygenated by
bubbling research grade nitrogen and an atmosphere of nitrogen
was maintained over the solution during measurements. The prod-
ucts were analyzed using a Hewlett Packard (HP) 6890 GC series
Gas Chromatograph equipped with a FID detector and a HP-5 cap-
capping the diiron(III) cluster. Yield: 0.74 g (80%). ESI-MS, m/z:
827.07 [(Mꢀ2HꢀClO4)+], 364.07 [(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc.
for C34H46Fe2N6O13Cl2: C, 43.94; H, 4.99; N, 9.04. Found: C,
43.89; H, 5.03; N, 9.09%.
2.3.6. [Fe2(O)(OBz)2(L6)2](ClO4)2 (6)
Complex 6 was prepared as reported [52] earlier. This was al-
ready well characterized by X-ray crystallographically and it pos-
sesses a distorted bioctahedral geometry in which each iron
atom is coordinated to the oxygen atom of the
oxygen atoms of -carboxylate bridges and three nitrogen atoms
illary column (30 m ꢂ 0.32 mm ꢂ 2.5
lm). GC-MS analysis was
performed on an Agilent GC-MS equipped with 7890A GC series
(HP-5 capillary column) and 5975C inert MSD under conditions
that were identical to those used for GC analysis.
l-oxo bridge, two
l
of the 3N ligand, capping the diiron(III) cluster. Yield: 0.79 g
[(Mꢀ2HꢀBzO)2+],
462.45
3. Results and discussion
(70%).
ESI-MS,
m/z:
401.38
[(Mꢀ2Hꢀ2ClO4)2+]. Anal. Calc. for C46H42Fe2N10O13Cl2: C, 49.09;
H, 3.76; N, 12.45. Found: C, 49.12; H, 3.72; N, 12.49%.
Caution! The perchlorate salts of the compounds are potentially
explosive! Only small quantities of these compounds should be
prepared and suitable precautions should be taken when they
are handled.
3.1. Synthesis and characterization of the 3N ligands and diiron(III)
complexes
The tridentate 3N ligands L1–L5 were synthesized according to
known procedures [53–58], which involve Schiff base condensa-
tion of the amines with the corresponding carboxaldehyde fol-
lowed by reduction with NaBH4. They were characterized by 1H
NMR and mass spectrometry. The ligand L6 was prepared by a
method involving a Philips condensation [59]. The diiron(III)
complexes 1–6 were prepared by adding one equivalent of the
3N ligand to an aqueous methanol solution containing a mixture
of one equivalent each of Fe(ClO4)3ꢁ6H2O, benzoic acid and
triethylamine. All of the complexes are formulated as
[Fe2(O)(OBz)2(L)2](ClO4)2 on the basis of elemental analysis,
UV-Vis spectroscopy, ESI-MS and X-ray crystal structures of 5
and 6. Attempts to grow single crystals of complexes 1–4 suitable
for X-ray diffraction analysis were unsuccessful. The pyridyl, imid-
azolyl and sterically hindering 6-methylpyridyl/benzimidazolyl
and weakly binding –NMe2 groups [pKa (BH+): pyridine, 5.2; imid-
azole, 7.01; benzimidazole, 5.4] in the ligands are expected to
influence the coordination as well as the electronic properties and
reactivities of the diiron(III) complexes. All the complexes, with dif-
ferent linear 3N donor ligands and benzoic acid as a bridging ligand,
are expected to mimic the active site environment of the substrate-
bound enzyme sMMO and so they have been chosen to model the
alkane hydroxylation reactions of the enzyme. The ligands with
different nitrogen donors and the benzoate bridging ligand are
expected to play an important role in determining the reactivity.
2.4. Catalytic oxidations
In a typical reaction, the oxidant m-CPBA (0.8 mol dmꢀ3) was
added to a mixture of the diiron(III) complex (1 ꢂ 10ꢀ3 mmol dmꢀ3
)
and the alkanes (3 mol dmꢀ3) in a CH2Cl2:CH3CN mixture (4:1 v/v).
After 30 min the reaction mixture was quenched with triphenylphos-
phine, the reaction mixture was filtered over a silica column and then
eluted with diethylether. An internal standard (bromobenzene) was
added at this point and the solution was subjected to GC analysis.
The mixture of organic products was identified by GC-MS and quan-
titatively analyzed by HP 6890 series GC equipped with an HP-5 cap-
illary column (30 m ꢂ 0.32 mm ꢂ 2.5
lm) using a calibration curve
obtained with authentic compounds. All of the products were quanti-
fied using GC (FID) with the following temperature program: injector
temperature 130 °C; initial temperature 60 °C, heating rate
10 °C minꢀ1 to 130 °C, increasing the temperature to 160 °C at a rate
of 2 °C minꢀ1, and then increasing the temperature to 260 °C at a rate
of 5 °C minꢀ1; FID temperature 280 °C. GC-MS analysis was per-
formed under conditions identical to those used for GC analysis.
The averages of three measurements are reported.
2.5. Physical measurements
3.2. Electronic spectral studies
Elemental analyses were performed on a Perkin Elmer Series II
CHNS/O Analyzer 2400. 1H NMR spectra were recorded on a Bruker
400 MHz NMR spectrometer. Electronic spectra were recorded on
an Agilent 8453 Diode Array spectrophotometer. Low temperature
spectra were obtained on an Agilent 8453 Diode Array spectropho-
tometer equipped with an UNISOKU USP-203 cryostat. Electro-
spray-ionization mass-spectrum (ESI-MS) analyses were recorded
on a Micromass Quattro II triple quadrupole mass spectrometer.
Cyclic voltammetry (CV) and differential pulse voltammetry
(DPV) were performed at 25.0 0.2 °C using a three-electrode cell
configuration. A platinum sphere, a platinum plate and Ag(s)/
AgNO3 were used as the working, auxiliary and reference elec-
trodes, respectively. The platinum sphere electrode was sonicated
for two minutes in dilute nitric acid, dilute hydrazine hydrate
and in double distilled water to remove any impurities. The refer-
ence electrode for a non-aqueous solution was Ag(s)/Ag+, which
consists of a Ag wire immersed in a solution of AgNO3 (0.01 M)
and tetra-N-butylammonium perchlorate (0.1 M) in acetonitrile
placed in a tube fitted with a Vycor plug. The instruments utilized
included an EG & G PAR 273 Potentiostat/Galvanostat and P-IV
computer along with EG & G M270 software to carry out the exper-
iments and to acquire the data. The temperature of the electro-
Inacetonitrilesolutionallthe diiron(III) complexes showtwo elec-
tronic spectral bands located in the range 450–520 nm (
emax, 930–
1450 Mꢀ1 cmꢀ1
,
Table 1, Fig. 2), which are characteristic
[39,50,51,60,61] of O2ꢀ ? iron(III) LMCT bands in oxo-bridged
diiron(III) complexes containing two carboxylate bridges. Four weak
spin-forbidden d–d transitions are expected for high-spin octahedral
iron(III) complexes; however, onlyone band around700 nmisreadily
observed for all the complexes and the remaining bands are obscured
by the more intense LMCT bands. The molar absorptivity of the band
around 700 nm is much higher than that observed for mononuclear
high-spin octahedral iron(III) complexes [62–69] due to lower sym-
metry of the dimeric complex with the oxo-bridge and also the
spin–spin interaction between the iron(III) centers [62–69]. The
oxo-bridge of the complexes remains intact in dichloromethane/ace-
tonitrile solvent mixture as revealed by ESI-MS analysis.
3.3. Electrochemical studies
The electrochemical properties of the diiron(III) complexes 1–6
were investigated in an acetonitrile/dichloromethane solvent

M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
175
Table 1
UV–Vis spectral data and electrochemical data of the diiron(III) complexes in DCM:ACN mixture at 25.0 °C.
Complex
kmax (nm) (
e
/Mꢀ1 cmꢀ1
)
Ep,c (CV) (V)
E1/2 (DPV) (V)
ꢀ0.685
Redox process
454 (1230)
499 (925)
698 (136)
459 (1410)
503(1148)
692 (233)
449 (1450)
498 (1110)
662 (175)
458 (1094)
498 (938)
672 (148)
466 (1445)
507 (1108)
720 (171)
489 (49)
[Fe2(O)(OBz)2(L1)2]2+
ꢀ0.781
FeIIIFeIII ? FeIIFeIII
FeIIIFeIII ? FeIIFeIII
FeIIIFeIII ? FeIIFeIII
FeIIIFeIII ? FeIIFeIII
FeIIIFeIII ? FeIIFeIII
FeIIIFeIII ? FeIIFeIII
[Fe2(O)(OBz)2(L2)2]2+
[Fe2(O)(OBz)2(L3)2]2+
[Fe2(O)(OBz)2(L4)2]2+
[Fe2(O)(OBz)2(L5)2]2+ [51]
[Fe2(O)(OBz)2(L6)2]2+
ꢀ0.795
ꢀ0.877
ꢀ0.807
ꢀ0.743
ꢀ0.855
ꢀ0.729
ꢀ0.805
ꢀ0.743
ꢀ0.679
ꢀ0.775
520 (20)
623 (17)
Potential measured vs. Ag/AgNO3 (0.001 M, 0.1 M TBAP); add 0.544 V to convert to NHE.
50
40
30
20
10
0
Wavelength (nm)
Fig. 2. Electronic spectrum of 4 ꢂ 10ꢀ4
acetonitrile mixture at 25 °C.
M a DCM/
[Fe2(O)(OBz)2(L1)2]2+ (1) in
-400 -500 -600 -700 -800 -900 -1000 -1100
E (mV)
mixture by employing cyclic (CV) and differential pulse voltamme-
try (DPV) on a stationary platinum electrode. All the complexes
show a cathodic peak in the range ꢀ0.743 to ꢀ0.877 V in the CV,
which corresponds to a one electron reduction of the diiron(III)
unit, and no coupled oxidation wave is discernible (Fig. 3). The
E1/2 values of the FeIII/FeII redox couple (ꢀ0.679 to ꢀ0.805 V, Ta-
ble 1) fall in the range observed for a similar type of triply-bridged
diiron(III) complexes [50,51,60,70]. The complexes are highly neg-
ative, mainly due to the strong coordination of the bridging carbox-
ylate and oxo groups, and follow the trend 5 < 1 < 2 < 4 < 3; 1 < 6,
revealing the importance of the capping 3N ligands in stabilizing
the iron(III) oxidation state in the above complexes with the same
benzoate bridges.
Fig. 3. Cyclic voltammogram (CV)/differential pulse voltammogram (DPV) of 1 mM
complex 4 in a DCM/acetonitrile mixture at 25 °C. Supporting electrolyte: 0.1 M
TBAP. scan rate: 50 mV sꢀ1 for CV and 5 mV/sꢀ1 for DPV.
(3); this reveals that the stronger coordination of the imidazolyl
nitrogen [(pKa (BH+): pyridine, 5.2; imidazole, 7.01] to the diiro-
n(III) center leads to a decrease in the Lewis acidity of the iron(III)
center and hence the more negative value of the redox potential.
Similarly, upon changing one of the pyridylmethyl arms of L1 in
1 to a benzimidazolyl methyl arm to get 4, the FeIIIFeIII/FeIIIFeII re-
dox potential of 1 moves towards a more negative value of
ꢀ0.743 V; this is due to the stronger coordination of the benzimi-
dazolyl nitrogen [(pKa (BH+): pyridine, 5.2; benzimidazole, 5.4)].
Also, upon replacing both the pyridylmethyl arms of L1 in 1, by
FeðIIIÞ ꢀ O ꢀ FeðIIIÞ þ eꢀ ! FeðIIIÞ ꢀ O ꢀ FeðIIIÞ
Upon replacing one of the pyridylmethyl arms in 1 by a 6-meth-
ylpyridyl arm to get 2, the Fe(III)/Fe(II) redox potential is shifted
from ꢀ0.685 V (1) to a more negative value of ꢀ0.729 V (2). The
6-methylpyridyl arm, with a sterically hindered nitrogen donor,
is expected to coordinate weakly to iron(III) resulting in an in-
crease in the positive charge on iron(III), which is compensated
by the stronger coordination of other pyridyl nitrogen donors of
the 3N ligand leading to render the reduction of iron(III) more dif-
ficult, and hence the more negative redox potential of 2. Upon
replacing one of the pyridylmethyl arms of L1 in 1 by an imi-
dazolylmethyl arm to get 3, the Fe(III)/Fe(II) redox potential is
shifted from ꢀ0.685 V (1) to a more negative value of ꢀ0.805 V
two strongly
r-bonding benzimidazolyl methyl arms to get 6,
[(pKa (BH+): pyridine, 5.2; benzimidazole, 5.4)] a similar decrease
in redox potential (ꢀ0.743 V) is observed. The replacement of
one of the pyridylmethyl arms of L1 in 1 by the –NMe2 group to
get 5, the redox potential is shifted to a less negative value of
ꢀ0.679 V, revealing that the weaker coordination of the sterically
hindering –NMe2 group leads to an increase in the Lewis acidity
of the iron(III) centers and hence the higher FeIIIFeIII/FeIIIFeII redox
potential of the diiron(III) center. Thus the above order of increas-
ing FeIIIFeIII/FeIIIFeII redox potentials represents the increasing or-
der of Lewis acidity of the diiron(III) center and also the stability

176
M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
of the diiron(III) core towards reduction. This is relevant to the
reduction of the diiron(III) core in the oxidized form of the sMMO
enzyme to obtain the reduced form, which is the active form of the
enzyme.
mixture at room temperature no spectral change is observed, even
after adding triethylamine. Also, the addition of the strong oxidant
m-CPBA to an acetonitrile solution of the diiron(III) complexes at
room temperature produces no significant spectral changes. How-
ever, the addition of one equivalent of triethylamine to a reaction
mixture containing m-CPBA leads to the appearance of a new band
in the region of 620–720 nm (Fig. 4), which is ascribed to the ad-
duct species [Fe2O(L)2(BzO)(OOCOC6H4Cl)]2+, formed by replacing
one of the bridging carboxylates by m-CPBA. During the decay of
this adduct in the absence of a substrate no new band characteris-
tic of high-valent intermediate species is observed. Similar obser-
vations have been made for the analogous diiron(III) complexes
[Fe2O(iBubpa/Bzbpa)2(OBz)2]2+ and [Fe2O(L)2(RCOO)2]2+ upon
addition of m-CPBA followed by triethylamine. We have already
characterized the adduct species [50,51] formed as [Fe2O(iBubpa/
Bzbpa)2(OBz)(OOCOC6H4Cl)]2+ and [Fe2O(L)2(RCO2)(OOCOC6H4-
Cl)]2+ by using UV-Vis spectroscopy and ESI-MS techniques. Also
for these adducts was no intermediate discerned during their de-
cay in the absence of a substrate. Such an intermediate species
has been proposed to be formed initially in alkane hydroxylation
reactions catalyzed by the complexes [Fe2(O)(OBz)2(iBubpa/
Bzbpa)2](ClO4)2 [50,51] and [Fe2(O)(OAc)2(hexpy)](ClO4)2 [39]
using m-CPBA as oxidant. So, we now propose that the acyloxo ad-
duct species [Fe2O(L)2(BzO)(OOCOC6H4Cl)]2+ (Scheme 3) under-
goes either O–O bond homolysis or O-O bond heterolysis leading
to the formation respectively of high-valent FeIV@O or FeV@O
intermediate species, which are involved in the selective hydroxyl-
ation of alkanes. We have already proposed [50,51] a similar mech-
anistic pathway to successfully illustrate the selective
hydroxylation of cyclohexane by [Fe2O(iBubpa/Bzbpa)2(OBz)2]2+
3.4. Functionalization of alkanes
The experimental conditions and results of the catalytic oxida-
tion of alkanes for the diiron(III) complexes 1–6 are summarized in
Tables 2–4. The conversion of alkanes into hydroxylated products
was quantified based on gas chromatographic analysis by using
authentic samples and an internal standard (bromobenzene). The
catalytic ability of the diiron(III) complexes towards the oxidation
of alkanes like cyclohexane, adamantane and cumene was investi-
gated using m-CPBA as the oxidant in a dichloromethane/acetoni-
trile solvent mixture (4:1 v/v) at room temperature. Control
reactions performed in the absence of the diiron(III) complexes
with m-CPBA as the oxidant yielded only very small amounts of
the oxidized products (cyclohexane, 3 TON; adamantane, 5 TON;
cumene 4 TON), revealing that all the diiron(III) complexes act as
catalysts towards the oxidation of alkanes to alcohols. In the pres-
ence of the diiron(III) complexes the oxidation of cyclohexane pro-
ceeds to give cyclohexanol (A) as the major product, along with
cyclohexanone (K) and e-caprolactone as the minor products. The
latter is the over-oxidized product of cyclohexanone in the pres-
ence of excess or unreacted m-CPBA. All the complexes display effi-
cient alkane hydroxylation with turnover numbers (TON) of 293–
401, with good selectivity for the hydroxylation of cyclohexane
(A/K, 9.3–12.8; Table 2). The catalytic activity of the diiron(III)
complexes towards hydroxylation of cyclohexane follows the trend
5 (Total TON, 411; A/K, 10.1) > 1 (388; 12.8) > 2 (376; 11.5) > 3
(362; 10.3) > 4 (319; 9.3); 1 > 6 (293; 9.5), illustrating the impor-
tance of the 3N capping ligands. Very interestingly, the same order
of TON is also observed for adamantane oxidation: 5 (Total TON,
437; 3°/2°, 15.7) > 1 (415; 28.1) > 2 (402; 22.1) > 3 (390; 21.3) >
4 (362; 17.8); 1 > 6 (336; 18.9) (Table 3), revealing the involvement
of a mechanism the same as that for the cyclohexane oxidation.
Thus, all the present diiron(III) complexes show high selectivity
in the hydroxylation of cyclohexane (A/K, 9.3–12.8) and adaman-
tane (3°/2°, 15.7–28.1) as well, signifying the involvement of me-
tal-based oxidants rather than non-selective freely diffusing
radical species in the alkane hydroxylation [43,37,71,72]. Also, un-
der a nitrogen atmosphere, almost the same type of reactivity pat-
tern is observed supporting the involvement of metal-based
oxidants rather than cyclohexylperoxide species in the catalytic
reaction.
and [Fe2O(L)2(RCO2)(OOCOC6H4Cl)]2+
.
This is similar to the reaction of a mononuclear iron(III) species
with m-CPBA to give a benzoylperoxoiron(III) species, which
undergoes O–O bond heterolysis to afford an FeV@O intermediate
and m-chlorobenzoic acid as a byproduct, and/or O–O bond
homolysis to yield an FeIV@O intermediate and chlorobenzene as
a byproduct. It is expected that the distinct nature of the ligand
donor atom would determine which one of the two mechanistic
pathways operates predominantly. Also, very recently, Que et al.
have reported that the complex cation [FeII(N4Py)(CH3CN)]2+
where N4Py is N,N-bis(2-pyridylmethyl)-N-(bis-2-pyridyl-
,
methyl)amine, promotes O–O bond heterolysis, while the cation
[FeIII(N4Py)(CH3CN)]3+ favors O–O bond homolysis, and they con-
cluded that the nature of the O–O bond cleavage is dependent on
the oxidation state of iron [72]. Also, high-valent FeIV@O species
have been invoked as the key intermediates in C–H bond func-
tionalization by enzymes, as well as their model complexes [7–
11,73]. Further, the involvement of a high-valent FeV@O species
has been proposed [74] in the hydroxylation of alkanes using
hydrogen peroxide in the presence of the iron(II) complex [Fe(Me2-
PyTACN)(OTf)2], where Me2PyTACN is N,N-dimethyl-N’-(pyrid-2-
ylmethyl)triazacyclononane, as a catalyst. For the present com-
plexes, we have observed the formation of chlorobenzene up to
60% based on the total TON, which reveals that up to 60% of the
oxidized products are formed due to the involvement of high-
valent FeIV@O species. This observation supports the proposed
When the oxidant H2O2 or t-BuOOH is added to the diiron(III)
complexes dissolved in a dichloromethane/acetonitrile solvent
Table 2
Products of oxidation of cyclohexane catalyzeda by the diiron(III) complexes.
Complex
Cyclohexane (TON)
Total TONc
A/Kd
Yielde (%)
-olb
-oneb
e-caprolactone
1
2
3
4
360
346
330
280
374
265
20
20
18
16
23
16
08
10
14
14
14
12
388
376
362
310
411
293
12.8
11.5
10.3
9.3
10.1
9.5
48.5
47.0
45.3
38.8
51.3
36.6
formation of the adduct species [Fe2O(L)2(BzO)(OOCOC6H4Cl)]2+
,
which undergoes O–O bond homolysis leading to generate the
FeIV@O species involved in the catalytic reaction, and chloroben-
zene is formed by the decarboxylation of the chlorobenzoate rad-
ical. As chlorobenzene is produced only up to 60% compared with
the total TON, it is clear that some other intermediate species is
also involved in the hydroxylation reaction to produce the
remaining hydroxylated products. Also, the formation of 3-chloro-
benzoic acid supports the formation of high-valent FeV@O species
in the selective hydroxylation of alkanes through heterolysis of
5 [51]
6
a Reaction conditions: Catalyst (1 ꢂ 10ꢀ3 mmol dmꢀ3), Substrate (3 mol dmꢀ3),
Oxidant (0.8 mol dmꢀ3) in DCM:ACN solvent mixture (9:1 v/v).
b -ol = cyclohexanol and -one = cyclohexanone.
c Total TON = no. of mmol of product/no. of mmol of catalyst.
d A/K = TON of -ol/(TON of -one + TON of
e Yield based on the oxidant.
e-caprolactone).

M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
177
Table 3
Products of oxidation of adamantane catalyzeda by the diiron(III) complexes.
Complex
Adamantane (TON)
1-adolb
Total TONc
Selectivityd
Yielde (%)
2-adolb
2-adoneb
3°/2°
1
2
3
4
375
354
342
310
367
290
40
48
38
44
58
35
–
–
10
08
12
11
415
402
390
362
437
336
28.1
22.1
21.3
17.8
15.7
18.9
69.1
67.0
65.0
60.3
72.8
56.0
5 [51]
6
a Reaction conditions: Catalyst (1 ꢂ 10ꢀ3 mmol dmꢀ3), Substrate (1 mol dmꢀ3), Oxidant (0.6 mol dmꢀ3) in DCM:ACN solvent mixture (4:1 v/v).
b 1-adol = 1-adamantanol, 2-adol = 2-adamantanol and 2-adone = 2-adamantanone.
c TON = no. of mmol of product/no. of mmol of catalyst.
d 3°/2° = (TON of 1-adol ꢂ 3)/(TON of 2-adol + TON of 2-adone).
e Yield based on the oxidant.
[Fe2O(iBubpa/Bzbpa)2(OBz)2]2+ after approximately 50 turnovers.
Kodera et al. observed [39] only a total of 280 turnovers for the
hydroxylation of cyclohexane using the diiron(III) complex [Fe2-
O(hexpy)(OAc)](ClO4)2, with the dinucleating ligand as a catalyst
and m-CPBA as the oxidant. So our observation of a high total
TON and better A/K ratio for the present complexes reveal that
the mononuclear [Fe(L)(BzO)(Sol)2]2+ species formed by breaking
the dimeric core also act as catalysts.
Table 4
Products of oxidation of cumene catalyzeda by the diiron(III) complexes.
Complex
2-Phenyl-2-propanol
Total TONb
Yieldc (%)
1
2
3
4
5
6
171
136
138
110
162
102
171
136
138
110
162
102
21.4
17.0
17.3
13.8
20.3
12.8
a Reaction conditions: Catalyst (1 ꢂ 10ꢀ3 mmol dmꢀ3), Substrate (3 mol dm-3),
Oxidant (0.8 mol dmꢀ3) in DCM:ACN solvent mixture (9:1 v/v).
b Total TON = no. of mmol of product/no. of mmol of catalyst.
c Yield based on the oxidant.
3.5. Cyclohexane oxidation
The ability of the complexes to catalyze the oxidation of cyclo-
hexane varies in the order 5 > 1 > 2 > 3 > 4; 1 > 6, as mentioned
above. Complex 1 catalyses the oxidation of cyclohexane to give a
total TON of 388 (cyclohexanol, 360 TON; cyclohexanone, 20
0.25
0.2
0.15
0.1
0.05
0
TON; e-caprolactone, 8 TON) with good alcohol selectivity (A/K,
12.8). Upon replacing one of the pyridylmethyl arms of L1 in 1 by
an imidazolylmethyl arm to get 3, the oxidation of cyclohexane
takes place with a decrease in the total TON from 388 to 362 (cyclo-
hexanol, 330 TON; cyclohexanone, 18 TON;
e-caprolactone, 14
TON) and the selectivity of the alcohol product (A/K, 10.3) as well;
also, the decreased Lewis acidity of the metal center due to the
stronger coordination of the imidazolyl nitrogen [pKa (BH+): pyri-
dine, 5.2; imidazole, 7.01] does not facilitate ligand exchange with
the m-CPBA anion to form the adduct and the proposed reactive
intermediate (cf. above). Similarly, upon replacing one of the pyri-
dylmethyl arms of L1 in 1 by the strongly coordinating benzimidaz-
olylmethyl arm [pKa (BH+): pyridine, 5.2; benzimidazole, 5.4] to get
4, the total TON (310; cyclohexanol, 280 TON; cyclohexanone, 16
500
600
700
800
900
1000
Wavelength (nm)
TON; e-caprolactone, 14 TON) as well as the selectivity (A/K, 9.3)
Fig. 4. Reaction of complex 1 with m-CPBA (5 equiv.) and triethylamine (1 equiv.)
followed by UV–Vis spectroscopy at room temperature.
decrease. Also, when both the pyridylmethyl arms of L1 in 1 are re-
placed by two benzimidazolylmethyl arms to get 6, the total TON
decreases from 388 to 293 (cyclohexanol, 265 TON; cyclohexanone,
16 TON; e-caprolactone, 12 TON) for cyclohexane oxidation and the
the O–O bond in the acyloxo adduct species. A similar but mono-
nuclear intermediate FeV@O species [Fe(TAML)(O)]ꢀ, derived from
the reaction of the iron(III) precursor with m-CPBA in butrylonit-
rile solution at ꢀ60 °C, has been already characterized [75] and
the stability of the intermediate species is traced to the strong
selectivity of the alcohol also decreases (A/K; 9.5). The steric bulk of
the benzimidazolyl groups may also hinder the binding of m-CPBA
to the diiron(III) center and decrease the concentration of the reac-
tive intermediate. In contrast, upon replacing one of the pyridyl-
methyl arms of L1 in 1 by the -CH2CH2NMe2 arm to give 5, the
oxidation of cyclohexane takes place with an increase in the total
TON from 388 to 411 (cyclohexanol, 374 TON; cyclohexanone, 23
r
-coordination of the tetra-anionic macrocyclic tetra-amide
ligand TAML.
For complex 1 the spectral bands characteristic of triply-
bridged diiron(III) complexes disappear after approximately 50
turnovers, revealing that the diiron(III) core degrades slowly dur-
ing catalysis. This suggests that after approximately 50 turnovers
a new active metal-based mononuclear intermediate species,
[Fe(L)(PhCO2)(Sol)2]2+, is formed, which is responsible for the
highly selective hydroxylation of alkanes to alcohols. We have
observed earlier [50] the breaking of the dimeric core of
TON; e-caprolactone, 14 TON), but with a decrease in selectivity
of the alcohol product (A/K, 10.1). The higher Lewis acidity of the
diiron(III) center (cf. above) in 5 enhances the ligand exchange with
m-CPBA and the formation of the proposed reactive intermediate
and hence the catalytic activity. A similar enhancement in Lewis
acidity of the diiron(III) center and hence higher catalytic activity
is expected [41,51] upon replacing one of the pyridylmethyl arms
of L1 in 1 by the 6-methylpyridylmethyl arm to get 2; however,

178
M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
Ph
O
H
N
Ph
O
H
N
N
N
O
FeIII
O
O
O
O
O
FeIII
N
III
m-CPBA
N
N
FeIII
N
Fe
-RCOO
O
O
N
N
N
N
O
Ph
H
H
O-O bond
heterolysis
O-O bond
homolysis
6
,
4
Cl
-
5
1
Ph
Ph
O
H
N
H
N
N
O
Fe III
N
O
O
O
N
N
FeIII
N
FeV
IV
O
O
N
N
Fe
O
O
N
N
N
O
O
m-CPBA
H
O
H
Cl
Cl
Ph
O
H
N
Ph
O
H
N
N
N
O
FeIII
O
O
FeIII
O
FeIII
N
N
FeIII
N
O
N
Sol Sol
Sol Sol
N
N
N
H
N
H
HO
O
OH
O
Cl
OH
CO2
Cl
Scheme 3. Proposed mechanism for alkane hydroxylation [51].
the oxidation of cyclohexane takes place with a decrease in both the
catalytic activity from 388 TON to 376 (cyclohexanol, 346 TON;
cyclohexanone, 20 TON; e-caprolactone, 10 TON) and the selectiv-
the same kind of intermediates and ligand stereoelectronic factors
are involved in both the oxidation reactions. Complex 1 catalyzes
the oxidation of adamantane with a total TON of 415 (1-adamant-
anol, 375 TON; 2-adamantanol, 40 TON) with a very good selectiv-
ity (3°/2°, 28.1), revealing that the reaction involves a metal based
oxidant. The catalytic activity of the complex is higher than those
previously reported for benzoate-bridged diiron(III) complexes
[Fe2O(OBz)2(iBubpa/Bzbpa)2](ClO4)2 (Total TON: iBubpa, 412;
Bzbpa, 329) [50]. The bond selectivity (3°/2°, 28.1) is lower than
ity (A/K, 11.5). So, it is evident that the incorporation of a methyl
group, which sterically hinders the coordination of the pyridyl
nitrogen, enhances the Lewis acidity of the diiron(III) center, but
is compensated by the stronger coordination of the other pyridyl
nitrogen leading to an effective decrease in the Lewis acidity of
the diiron(III) center, and hence the lower catalytic activity. Thus
the Lewis acidity of the diiron(III) center, as modified by the differ-
ent donor atoms of the capping ligand, dictates the ligand exchange
with the oxidant and the formation of the intermediate high-valent
oxo species and hence the catalytic activity. Also, a homolytic O–O
bond cleavage pathway [72] is suggested to occur predominantly
for 5, with the highest Lewis acidity for the diiron(III) center, while
heterolytic O–O bond cleavage is favored for the remaining com-
plexes with lower Lewis acidity.
i
that for the Bubpa (3°/2°, 30.3) complex, but higher than that for
the Bzbpa (3°/2°, 26.1) complex [50], suggesting that the capping
ligand plays an important role in determining the bond selectivity.
Upon introducing the imidazolylmethyl arm into L1 of 1 to get 3,
the oxidation of adamantane takes place with a decrease in both
the total TON 390 (1-adamantanol, 342 TON; 2-adamantanol, 38
TON; 2-adamantanone 10 TON) and the 3°/2° (21.3) bond selectiv-
ity. As discussed above for the cyclohexane oxidation, the strongly
coordinating imidazolyl nitrogen [pKa (BH+): pyridine, 5.2; imidaz-
ole, 7.01] decreases the ligand exchange of the bridging corboxy-
late with m-CPBA, leading to lower catalytic efficiency (cf.
above). A similar decrease in the catalytic activity is observed
when one pyridylmethyl arm in L1 of 1 is replaced with one benz-
imidazolyl arm to get 4, both the total TON (362; 1-adamantanol,
310 TON; 2-adamantanol, 44 TON; 2-adamantanone 8 TON) and
3°/2° bond selectivity (17.8) decrease. Similarly, on replacement
of both the pyridylmethyl arms in 1 by benzimidazolylmethyl arms
3.6. Adamantane and cumene oxidation
All the diiron(III) complexes catalyze the oxidation of adaman-
tane efficiently to give 1-adamantanol and 2-adamantanol as the
major products, along with 2-adamantanone as a minor product.
The ability of the complexes to catalyze the oxidation of adaman-
tane varies in the order 5 (437) > 1 (415) > 2 (402) > 3 (390) > 4
(362); 1 > 6 (336), as for the cyclohexane oxidation, revealing that

M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
179
[6] A.L. Feig, S.J. Lippard, Chem. Rev. 94 (1994) 759.
[7] B.J. Waller, J.D. Lipscomb, Chem. Rev. 96 (1996) 2625.
[8] M. Baik, M. Newcomb, R.A. Friesner, S.J. Lippard, Chem. Rev. 103 (2003)
2385.
[9] M. Costas, M.P. Mehn, M.P. Jensen, L. Que Jr., Chem. Rev. 104 (2004) 939.
[10] E.Y. Tshuva, S.J. Lippard, Chem. Rev. 104 (2004) 987.
[11] S.V. Kryatov, E.V. Rybak-Akimova, S. Schindler, Chem. Rev. 105 (2005) 2175.
[12] A.C. Rosenzweig, C.A. Frederick, S.J. Lippard, P. Nordlund, Nature 366 (1993)
537.
[13] L. Westerheide, M. Pascaly, B. Krebs, Curr. Opin. Chem. Biol. 4 (2000) 235.
[14] M. Merkx, D.A. Kopp, M.H. Sazinsky, J.L. Blazyk, J. Muller, S.J. Lippard, Angew.
Chem., Int. Ed. 40 (2001) 4000.
[15] R.E. Stenkamp, Chem. Rev. 94 (1994) 715.
[16] D.T. Logan, X.D. Su, A. Aberg, K. Regnstrom, J. Hajdu, H. Eklund, P. Nordlund,
Structure 4 (1996) 1053.
[17] A. Jordan, P. Reichard, Annu. Rev. Biochem. 67 (1998) 71.
[18] J. Stubbe, Curr. Opin. Chem. Biol. 7 (2003) 183.
[19] P.C. Wilkins, R.C. Wilkins, Coord. Chem. Rev. 79 (1987) 195.
[20] S.J. Lippard, Angew. Chem., Int. Ed. Engl. 27 (1988) 344.
[21] D.M. Kurtz Jr., Chem. Rev. 90 (1990) 585.
to get 6, the catalytic activity decreases, as illustrated above for the
cyclohexane oxidation (Total TON 336; 1-adamantanol, 290 TON;
2-adamantanol, 35 TON; 2-adamantanone 11 TON; bond selectiv-
ity 3°/2°, 18.9). Also, the steric bulk of the benzimidazolyl groups
may hinder the binding of m-CPBA to the diiron(III) center. In con-
trast, the replacement of the pyridylmethyl arms in L1 of 1 by the
bulky –NMe2 group to get 5 increases the total TON to 437 (1-ada-
mantanol, 367 TON; 2-adamantanol, 58 TON; 2-adamantanone 12
TON) and also the bond selectivity (15.7). Interestingly, the cata-
lytic efficiency is higher than those of the previously reported ben-
zoate-bridged complexes (total TON: iBubpa, 412; Bzbpa, 329).
However, the 3°/2° selectivity for 5 (15.7) is again lower than those
i
of the previously reported [50] complexes (3°/2°: Bubpa, 30.3;
Bzbpa, 26.1). Similarly, for complex 2 the catalytic oxidation of
adamantane takes place with a small decrease in the catalytic
activity from 415 TON to 402 with a good selectivity (3°/2°,
22.1), as illustrated above for cyclohexane oxidation (cf. above).
Interestingly, complex 2 is very selective in catalyzing adamantane
to the alcohol product. All the above observations suggest that the
capping ligand plays an important role in determining the selectiv-
ity of the alcohol product.
[22] J.B. Vincent, G.L. Oliver-Lilley, B.A. Averill, Chem. Rev. 90 (1990) 1447.
[23] E.I. Solomon, F. Tuczek, D.E. Root, C.A. Brown, Chem. Rev. 94 (1994) 827.
[24] W. Nam, R. Ho, J.S. Valentine, J. Am. Chem. Soc. 113 (1991) 7052.
[25] K. Chen, L. Que Jr., Chem. Commun. (1999) 1375.
[26] K. Chen, L. Que Jr., Angew. Chem., Int. Ed. 38 (1999) 2227.
[27] K. Chen, L. Que Jr., J. Am. Chem. Soc. 123 (2001) 6327.
[28] Y. Mekmouche, S. Ménage, C. Toia-Duboc, M. Fontecave, J.-B. Galey, C. Lebrun,
J. Pécaut, Angew. Chem., Int. Ed. 40 (2001) 949.
The catalytic activity of the diiron(III) complexes towards the
oxidation of cumene has also been investigated and the results
are summarized in Table 4. Interestingly, all the complexes cata-
lyze the oxidation of cumene selectively to form 2-phenyl-2-pro-
panol without any side product formation. The oxidation of
cumene follows the same order as that for cyclohexane oxidation.
[29] K. Chen, M. Costas, L. Que Jr., J. Chem. Soc., Dalton Trans. (2002) 672.
[30] M. Fontecave, S. Ménage, C. Duboc-Toia, Coord. Chem. Rev. 178 (1998)
1555.
[31] W.H. Armstrong, S.J. Lippard, J. Am. Chem. Soc. 105 (1983) 4837.
[32] K. Wieghardt, K. Pohl, W. Gebert, Angew. Chem., Int. Ed. Engl. 22 (1983) 727.
[33] K. Wieghardt, I. Tolksdorf, W. Herrmann, Inorg. Chem. 24 (1985) 1230.
[34] K.B. Jensen, C.J. McKenzie, O. Simonsen, H. Toftlund, A. Hazell, Inorg. Chim.
Acta 257 (1997) 163.
[35] S. Nishino, H. Hosomi, S. Ohba, H. Matsushima, T. Tokii, Y. Nishida, J. Chem.
Soc., Dalton Trans. (1999) 1509.
4. Conclusions
[36] J.B. Vincent, J.C. Huffmann, G. Christou, Q. Li, M.A. Nanny, D.N. Hendrickson,
R.H. Fong, R.H. Fish, J. Am. Chem. Soc. 110 (1988) 6898.
[37] R.H. Fish, M.S. Konings, K.J. Oberhausen, R.H. Fong, W.M. Yu, G. Christou, J.B.
Vincent, D.K. Coggin, R.M. Buchanan, Inorg. Chem. 30 (1991) 3002.
[38] N. Kitajima, H. Fukui, Y. Moro-oka, J. Chem. Soc., Chem. Commun. (1988) 485.
[39] M. Kodera, H. Shimakoshi, K. Kano, Chem. Commun. (1996) 1737.
[40] M. Kodera, H. Shimakoshi, M. Nishimura, H. Okawa, S. Iijima, K. Kano, Inorg.
Chem. 35 (1996) 4967.
[41] M. Kodera, Y. Taniike, M. Itoh, Y. Tanahashi, H. Shimakoshi, K. Kano, S. Hirota,
S. Iijima, M. Ohba, H. Okawa, Inorg. Chem. Commun. 40 (2001) 4821.
[42] M. Itoh, M. Kodera, K. Kano, J. Inorg. Biochem. 96 (2003) 158.
[43] R.M. Buchanan, S. Chen, J.F. Richardson, M. Bressan, L. Forti, A. Morvillo, R.H.
Fish, Inorg. Chem. 33 (1994) 3208.
[44] D. Tetard, J.-B. Verlhac, J. Mol. Catal. A: Chem. 113 (1996) 223.
[45] R.A. Leising, J. Kim, M.A. Pérez, L. Que Jr., J. Am. Chem. Soc. 115 (1993) 9524.
[46] I.W.C.E. Arends, K.U. Ingold, D.D.M. Wayner, J. Am. Chem. Soc. 117 (1995)
4710.
[47] T. Okuno, S. Itoh, S. Ohba, Y. Nishida, J. Chem. Soc., Dalton Trans. (1997) 3547.
[48] S. Ménage, J.-B. Galey, J. Dumats, G. Hussler, M. Seité, I.G. Luneau, G. Chottard,
M. Fontecave, J. Am. Chem. Soc. 120 (1998) 13370.
We have isolated a few triply-bridged diiron(III) complexes de-
rived from linear 3N ligands and studied their ability to carry out
alkane functionalization using m-CPBA as an oxidant. All the com-
plexes show spectral and electrochemical properties characteristic
of l-oxo-bridged diiron(III) complexes also containing two carbox-
ylate bridges. The A/K and 3°/2° bond selectivity for hydrocarbon
oxidation catalyzed by the present complexes fall in the ranges
9.3–12.8 and 15.7–28.1 respectively. This suggests that the capping
ligand plays an important role in determining the selectivity of the
product. The observation of high selectivity towards alkane
hydroxylation suggests the involvement of an intermediate me-
tal-based oxidant rather than freely diffusing radicals in the oxy-
genation. The 3°/2° bond selectivity observed for the
hydroxylation of adamantane for the diiron(III) complex of the
3N capping ligand with two pyridyl donors (L1) is the highest
among the present complexes.
[49] E. Pardo, F. Lloret, R. Carrasco, M.C. Muñoz, T.T. Sánchez, R.R. García, Inorg.
Chim. Acta 357 (2004) 2713 (and references therein).
[50] K. Visvaganesan, E. Suresh, M. Palaniandavar, Dalton Trans. (2009) 3814.
[51] M. Balamurugan, E. Suresh, M. Palaniandavar, Inorg. Chem. (2013) Submitted.
[52] P.G. Romero, N.C. Pastor, A.B. Hussein, G.B. Jameson, J. Am. Chem. Soc. 110
(1988) 1988.
Acknowledgments
[53] M. Velusamy, R. Mayilmurugan, M. Palaniandavar, J. Inorg. Biochem. 99 (2005)
1032.
We sincerely thank the Council of Scientific and Industrial Re-
search, New Delhi, for a Senior Research Fellowship to M.S. This
work was also supported by the Indo-French Centre (IFCPAR)
[Scheme No. IFC/A/4109-1/2009/993]. Professor M. Palaniandavar
is a recipient of a DST Ramanna Fellowship [Scheme No. SR/S1/
RFIC-01/2007 and SR/S1/RFIC-01/2010]. We thank Dr. Balachan-
dran Unni Nair, Central Leather Research Institute, Chennai for pro-
viding the ESI-MS facility.
[54] N. Anitha, M. Palaniandavar, Dalton Trans. 40 (2011) 1188.
[55] T. Dhanalakshmi, E. Suresh, M. Palaniandavar, Dalton Trans. (2009) 8317.
[56] K. Sundaravel, E. Suresh, M. Palaniandavar, Inorg. Chim. Acta 363 (2010) 2768.
[57] D.C. Crans, A.D. Keramidas, S.S. Amin, O.P. Anderson, S.M. Miller, J. Chem. Soc.,
Dalton Trans. (1997) 2799.
[58] L.A. Cescon, A.R. Day, J. Org. Chem. 27 (1962) 581.
[59] M.A. Philips, J. Chem. Soc. (1928) 2393.
[60] W.H. Armstrong, A. Spool, G.C. Papaefthymiou, R.B. Frankel, S.J. Lippard, J. Am.
Chem. Soc. 106 (1984) 3653.
[61] P. Gomez-Romero, N. Casan-Pastor, A. Ben-Hussein, G.B. Jameson, J. Am. Chem.
Soc. 110 (1988) 1988.
[62] R. Viswanathan, M. Palaniandavar, T. Balasubramanian, P.T. Muthiah, Inorg.
Chem. 37 (12) (1998) 2943.
References
[63] M. Velusamy, R. Mayilmurugan, M. Palaniandavar, Inorg. Chem. 43 (2004)
6284.
[64] K. Visvaganesan, R. Mayilmurugan, E. Suresh, M. Palaniandavar, Inorg. Chem.
46 (2007) 10294.
[65] R. Mayilmurugan, E. Suresh, M. Palaniandavar, Inorg. Chem. 46 (2007) 6038.
[66] K. Sundaravel, T. Dhanalakshmi, E. Suresh, M. Palaniandavar, Dalton Trans.
(2008) 7012.
[1] D. Riley, M. Stern, J. Ebner, in: D.H.R. Barton, A.E. Martell, D.T. Sawyer (Eds.),
The Activation of Dioxygen and Homogeneous Catalytic Oxidation, Plenum,
New York, 1993, p. 31.
[2] J. Colby, H. Dalton, R. Whittenbury, Annu. Rev. Microbiol. 33 (1979) 481.
[3] H. Dalton, Adv. Appl. Microbiol. 26 (1980) 71.
[4] J.B. Vincent, G.L. Olivier-Lilley, B.A. Averill, Chem. Rev. 90 (1990) 1447.
[5] J.D. Lipscomb, Annu. Rev. Microbiol. 48 (1994) 317.

180
M. Sankaralingam, M. Palaniandavar / Polyhedron 67 (2014) 171–180
[67] R. Mayilmurugan, K. Visvaganesan, E. Suresh, M. Palaniandavar, Inorg. Chem.
[71] J. Kaizer, E.J. Klinker, N.Y. Oh, J.-U. Rohde, W.J. Song, A. Stubna, J. Kim, E. Münck,
W. Nam, L. Que Jr., J. Am. Chem. Soc. 126 (2004) 472.
48 (2009) 8771.
[68] K. Sundaravel, E. Suresh, K. Saminathan, M. Palaniandavar, Dalton Trans. 40
[72] K. Ray, S.M. Lee, L. Que Jr., Inorg. Chim. Acta 361 (2008) 1066.
[73] X. Shan, L. Que Jr., J. Inorg. Biochem. 100 (2006) 421.
(2011) 8092.
[69] K. Sundaravel, M. Sankaralingam, E. Suresh, M. Palaniandavar, Dalton Trans. 40
(2011) 8444.
[74] A. Company, L. Gómez, M. Güell, X. Ribas, J.M. Luis, L. Que Jr., M. Costas, J. Am.
Chem. Soc. 129 (2007) 15766.
[70] J.-A.R. Hartman, R.L. Rardin, P. Chaudhuri, K. Pohl, K. Wieghardt, B. Nuber, J.
Weiss, G.C. Papaefthymiou, R.B. Frankel, S.J. Lippard, J. Am. Chem. Soc. 109
(1987) 7387.
[75] F.T. Oliveira, A. Chanda, D. Banerjee, X. Shan, S. Mondal, L. Que Jr., E.L.
Bominaar, E. Münck, T.J. Collins, Science 315 (2007) 835.