Novel square pyramidal iron(iii) complexes of linear tetradentate bis(phenolate) ligands as structural and reactive models for intradiol-cleaving 3,4-PCD enzymes: Quinone formation vs. intradiol cleavage

Article abstract of DOI:10.1039/c0dt00171f

The iron(iii) complexes of the bis(phenolate) ligands 1,4-bis(2-hydroxy-4- methyl-benzyl)-1,4-diazepane H₂(L1), 1,4-bis(2-hydroxy-4-nitrobenzyl) -1,4-diazepane H₂(L2), 1,4-bis(2-hydroxy-3,5-dimethylbenzyl)-1,4- diazepane H₂(L3) and 1,4-bis(2-hydroxy-3,5-di-tert-butylbenzyl)-1,4- diazepane H₂(L4) have been isolated and studied as structural and functional models for 3,4-PCD enzymes. The complexes [Fe(L1)Cl] 1, [Fe(L2)(H₂O)Cl] 2, [Fe(L3)Cl] 3 and [Fe(L4)Cl] 4 have been characterized using ESI-MS, elemental analysis, and absorption spectral and electrochemical methods. The single crystal X-ray structure of 3 contains the FeN₂O₂Cl chromophore with a novel square pyramidal (τ, 0.20) coordination geometry. The Fe-O-C bond angle (135.5°) and Fe-O bond length (1.855 A) are very close to the Fe-O-C bond angles (133, 148°) and Fe-O(tyrosinate) bond distances (1.81, 1.91 A) in 3,4-PCD enzyme. All the complexes exhibit two intense absorption bands in the ranges 335-383 and 493-541 nm, which are assigned respectively to phenolate (pπ) → Fe(iii) (dσ*) and phenolate (pπ) → Fe(iii) (dπ*) LMCT transitions. The Fe(iii)/Fe(ii) redox potentials of 1, 3 and 4 (E_1/2, -0.882- -1.010 V) are more negative than that of 2 (E_1/2, -0.577 V) due to the presence of two electron-withdrawing p-nitrophenolate moieties in the latter enhancing the Lewis acidity of the iron(iii) center. Upon adding H ₂DBC pretreated with two equivalents of Et₃N to the iron(iii) complexes, two catecholate-to-iron(iii) LMCT bands (656, ε, 1030; 515 nm, ε, 1330 M^-1 cm^-1) are observed for 2; however, interestingly, an intense catecholate-to-iron(iii) LMCT band (530-541 nm) is observed for 1, 3 and 4 apart from a high intensity band in the range 451-462 nm. The adducts [Fe(L)(DBC)]^- generated from 1-4in situ in DMF/Et₃N solution react with dioxygen to afford almost exclusively the simple two-electron oxidation product 3,5-di-tert-butylbenzoquinone (DBQ), which is discerned from the appearance and increase in intensity of the electronic spectral band around 400 nm, and smaller amounts of cleavage products. Interestingly, in DMF/piperidine the amount of quinone product decreases and those of the cleavage products increase illustrating that the stronger base piperidine enhances the concentration of the catecholate adduct. The rates of both dioxygenation and quinone formation observed in DMF/Et ₃N solution vary in the order 1 > 3 > 4 < 2 suggesting that the ligand steric hindrance to molecular oxygen attack, the Lewis acidity of the iron(iii) center and the ability of the complexes to rearrange the Fe-O phenolate bonds to accommodate the catecholate substrate dictate the extent of interaction of the complexes with substrate and hence determine the rates of reactions. This is in line with the observation of DBSQ/H₂DBC reduction wave for the adduct [Fe(L2)(DBC)]^- at a potential (E _1/2: -0.285 V) more positive than those for the adducts of 1, 3 and 4 (E_1/2: -0.522 to -0.645 V).

Full text of DOI:10.1039/c0dt00171f

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/dalton | Dalton Transactions
Novel square pyramidal iron(III) complexes of linear tetradentate
bis(phenolate) ligands as structural and reactive models for intradiol-cleaving
3,4-PCD enzymes: Quinone formation vs. intradiol cleavage†
Ramasamy Mayilmurugan,a Muniyandi Sankaralingam,a Eringathodi Sureshb and Mallayan Palaniandavar*a
Received 18th March 2010, Accepted 14th July 2010
DOI: 10.1039/c0dt00171f
The iron(III) complexes of the bis(phenolate) ligands 1,4-bis(2-hydroxy-4-methyl-benzyl)-1,4-diazepane
H2(L1), 1,4-bis(2-hydroxy-4-nitrobenzyl)-1,4-diazepane H2(L2), 1,4-bis(2-hydroxy-3,5-
dimethylbenzyl)-1,4-diazepane H2(L3) and 1,4-bis(2-hydroxy-3,5-di-tert-butylbenzyl)-1,4-diazepane
H2(L4) have been isolated and studied as structural and functional models for 3,4-PCD enzymes. The
complexes [Fe(L1)Cl] 1, [Fe(L2)(H2O)Cl] 2, [Fe(L3)Cl] 3 and [Fe(L4)Cl] 4 have been characterized
using ESI-MS, elemental analysis, and absorption spectral and electrochemical methods. The single
crystal X-ray structure of 3 contains the FeN2O2Cl chromophore with a novel square pyramidal (t,
◦
˚
0.20) coordination geometry. The Fe–O–C bond angle (135.5 ) and Fe–O bond length (1.855 A) are
◦
˚
very close to the Fe–O–C bond angles (133, 148 ) and Fe–O(tyrosinate) bond distances (1.81, 1.91 A) in
3,4-PCD enzyme. All the complexes exhibit two intense absorption bands in the ranges 335–383 and
493–541 nm, which are assigned respectively to phenolate (pp) → Fe(III) (ds*) and phenolate (pp) →
Fe(III) (dp*) LMCT transitions. The Fe(III)/Fe(II) redox potentials of 1, 3 and 4 (E1/2, -0.882– -1.010 V)
are more negative than that of 2 (E1/2, -0.577 V) due to the presence of two electron-withdrawing
p-nitrophenolate moieties in the latter enhancing the Lewis acidity of the iron(III) center. Upon adding
H2DBC pretreated with two equivalents of Et3N to the iron(III) complexes, two catecholate-to-iron(III)
LMCT bands (656, e, 1030; 515 nm, e, 1330 M-1 cm-1) are observed for 2; however, interestingly, an
intense catecholate-to-iron(III) LMCT band (530–541 nm) is observed for 1, 3 and 4 apart from a high
intensity band in the range 451–462 nm. The adducts [Fe(L)(DBC)]- generated from 1–4 in situ in
DMF/Et3N solution react with dioxygen to afford almost exclusively the simple two-electron oxidation
product 3,5-di-tert-butylbenzoquinone (DBQ), which is discerned from the appearance and increase in
intensity of the electronic spectral band around 400 nm, and smaller amounts of cleavage products.
Interestingly, in DMF/piperidine the amount of quinone product decreases and those of the cleavage
products increase illustrating that the stronger base piperidine enhances the concentration of the
catecholate adduct. The rates of both dioxygenation and quinone formation observed in DMF/Et3N
solution vary in the order 1 > 3 > 4 < 2 suggesting that the ligand steric hindrance to molecular oxygen
attack, the Lewis acidity of the iron(III) center and the ability of the complexes to rearrange the Fe–O
phenolate bonds to accommodate the catecholate substrate dictate the extent of interaction of the
complexes with substrate and hence determine the rates of reactions. This is in line with the observation
of DBSQ/H2DBC reduction wave for the adduct [Fe(L2)(DBC)]- at a potential (E1/2: -0.285 V) more
positive than those for the adducts of 1, 3 and 4 (E1/2: -0.522 to -0.645 V).
are divided into two subclasses: the intradiol dioxygenases,
Introduction
which utilize a non-heme iron(III) cofactor in catalyzing the
cleavage of the carbon–carbon bond between the two catechol
oxygens; and the extradiol dioxygenases, which utilize a non-
heme iron(II) cofactor in catalyzing the cleavage of the carbon–
carbon bond adjacent to the catechol oxygens.8–17 Even though
intradiol cleavage is the less common route in the biodegradation
of aromatic molecules as compared to extradiol cleavage, extensive
investigation of intradiol dioxygenases has been made due to the
rich spectroscopy of the iron(III) active site in the enzymes.14–16
The two most investigated enzymes of the intradiol dioxygenases
are catechol 1,2-dioxygenase (1,2-CTD) and protocatechuate 3,4-
dioxygenase (3,4-PCD). The iron(III) center in these enzymes
adopts a trigonal bipyramidal geometry with two inequivalent
tyrosine ligands (axial, Tyr447 and equatorial, Tyr408), two
histidines (His460, His462), and a hydroxide ion.10–18 On binding
A number of important oxidative transformations are carried out
by iron enzymes utilizing dioxygen as an oxygen source.1–5 The
oxidative cleavage of catechol and other dihydroxy aromatics is
a key step in the biodegradation of naturally occurring aromatic
molecules and many aromatic environmental pollutants by soil
bacteria.6,7 Catechol dioxygenases are a class of non-heme iron
enzymes that catalyse the oxidative cleavage of catechols. They
aCentre for Bioinorganic Chemistry, School of Chemistry, Bharathidasan
University, Tiruchirapalli, 620 024, India. E-mail: palanim51@yahoo.com,
palaniandavarm@gmail.com
bAnalytical Science Discipline, Central Salt and Marine Chemicals Research
Institute, Bhavnagar, –364 002, India
† CCDC reference number 770712. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/c0dt00171f
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9611
©

View Article Online
Scheme 1 Structures of active site of 3,4-PCD (a) and the 3,4-PCD-substrate adduct (b).
to iron(III) the catechol substrate (PCA) donates its hydroxyl
protons to the hydroxide ligand and the axial tyrosine (Tyr447)
to afford a bidentate iron(III) catecholate complex.16,17 The active
site geometry is converted into a square-pyramidal geometry and
the axial Tyr447 and equatorial OH- are displaced by the doubly
deprotonated substrate and retains the endogenous histidine
and equatorial tyrosine (Tyr408) ligands (Scheme 1).10–18 The
open coordination position is trans to His460, and the substrate
chelates asymmetrically to the iron(III) center, with the longer
bond trans to the equatorial Tyr408.19 The asymmetric chelation
of the substrate moiety is determined19 by the Fe–O–C bond
angle of equatorially bound tyrosinate in the 3,4-PCD enzyme–
substrate complex, which in turn influences the extent of iron(III)-
catecholate covalency and hence substrate activation for reaction
with O2.
The study of iron(III) complexes18,20–30 of ligands with phenolate
oxygen and pyridine nitrogen donors have provided valuable
information to elucidate the structure–function correlation for
the active site geometries of the intradiol-cleaving enzymes. The
ligand donor functions in the complexes display different steric
and electronic effects in determining the Lewis acidity of the
iron(III) center and hence their intradiol cleavage activity. Recently
we have reported22,31 a few mononuclear iron(III) complexes of
sterically hindering tetradentate tripodal monophenolate ligands,
which closely mimic the enzyme active site structure and function.
The complex [Fe(L)Cl2], where H(L) is N,N-dimethyl-N¢-(pyrid-2-
ylmethyl)-N¢-(2-hydroxy-4-nitrobenzyl)ethylenediamine, exhibits
a relatively high Fe–O–C bond angle of 136.1◦ and shows an
enhanced rate of dioxygenation. Interestingly, upon replacement22
ligands. The structure of the novel complex [Fe(Mes6-SALEN)-
(OH2)]ClO4, where Mes6-SALEN is bis(3,5-dimesitylsalicylidene)-
1,2-dimesitylethylene-diamine, has been reported33 to adopt dis-
torted trigonal bipyramidal coordination geometry (t, 0.48) as in
the 3,4-PCD active site (t, 0.44), but no reactivity studies on the
complex were carried out. Krebs et al.26,27 isolated several iron(III)
complexes of substituted bis-phenolate ligands and studied them
as models for the inhibitor-substrate adducts of intradiol-cleaving
1,2-CTD enzyme. This report describes novel iron(III) complexes
of new sterically hindering diazepane-based linear bis-phenolate
ligands [H2(L1)–H2(L4), Scheme 2] which not only closely mimic
the coordination of two tyrosine phenolate donors in the active
site of 3,4-PCD enzymes but also mimic their spectral properties
and function. Such ligands with a 1,4-diazepane back bone are
interesting because incorporation of this 1,4-diazepane moiety in a
4N ligand system confers a rare cis-b configuration on the iron(III)
complex-catecholate adduct and elicits regioselective extradiol
cleavage of catechol.34 Also, we have very recently found21 that
incorporation of two 3,5-dimethylphenolate arms as in [Fe(L5)Cl]
(Scheme 2) confers a distorted trigonal bipyramidal coordination
geometry on iron(III), which is closely related to the trigonal
bipyramidal iron(III) core in the substrate-free 3,4-PCD enzyme,
but the complex fails to exhibit dioxygenase activity. In contrast,
the analogous complex [Fe(L6)(Cl)(H2O)] exhibits an octahedral
coordination geometry and shows, interestingly, intradiol cleavage
activity. Similarly, the five-coordinate complex [Fe(L7)Cl] fails to
cleave catechol while the octahedral complex [Fe(L8)(Cl)(H2O)]
exhibits a very fast intradiol cleavage reaction.21 Also, we have
of the p-nitrophenolate arm in this complex by
a 3,5-
dimethylphenolate arm to obtain the complex [Fe(L)Cl2],
where H(L) is N,N-dimethyl-N¢-(2-hydroxy-3,5-dimethylbenzyl)-
N¢-(pyrid-2-ylmethyl)ethylenediamine, the higher Fe–O–C bond
angle (134.0◦) is retained but a regioselective extradiol cleavage of
3,5-di-tert-butylcatechol (H2DBC) is observed.31
Though the active site structures of intradiol-cleaving enzymes
contain two tyrosine phenolate residues, only a few iron(III)
complexes of bis(phenolate) ligands have been so far isolated and
studied as functional models for the enzymes. Earlier Que et al.
reported32 the catecholate adducts of iron(III)-SALEN complexes
and observed that the rates of oxygenation of the adducts are
poor, and the main product obtained in the reaction is typically
the two-electron oxidation product benzoquinone. Previously we
observed25b that iron(III) complexes of bis(phenolate) ligands
also elicit intradiol cleavage products with yields higher than
those of iron(III) complexes of the corresponding monophenolate
Scheme 2 Structures of linear and tripodal bis(phenolate) ligands.
9612 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
now shown that the parent ligand 1,4-bis(2-hydroxybenzyl)-1,4-
diazepane (H2(L9)) forms the dimeric complex [Fe2(m-O)(L9)2]
in which each iron atom has a distorted square pyramidal
coordination geometry (t, 0.45).35 So, it is expected that incorpo-
ration of different substituents with varying electronic and steric
effects on the phenolate ring as in the present ligands would
confer unusual coordination geometries on iron(III) and varying
dioxygenase activities as well. Thus, the X-ray crystal structure of
the present bis(phenolato)iron(III) complex [Fe(L3)Cl] with two
3,5-dimethylphenolate arms exhibit a novel square pyramidal co-
ordination geometry with a Fe–O–C bond angle of 135.5◦, which is
close to that in the enzyme active site. Interestingly, the catecholate
adducts of all the complexes in DMF–acetonitrile/triethylamine
solution yield the two-electron oxidized product 3,5-di-tert-butyl-
benzoquinone almost exclusively with small amounts of cate-
chol cleavage products. However, in DMF/piperidine solution
decreased amounts of benzoquinone and enhanced amounts of
intradiol cleavage products are observed.
2.5 mm) and GC-MS analysis was performed on a Perkin Elmer
Clarus 500 GC-MS instrument using a PE-5 with the previously31
reported temperature program.
Synthesis of ligands
1,4-Bis(2-hydroxy-4-methylbenzyl)-1,4-diazepane H2(L1). To
a solution of 2-hydroxy-5-methylbenzaldehyde (1.36 g, 10.0 mmol)
in methanol (50 mL) were added homopiperazine (0.50 g,
5.0 mmol) and a small amount of acetic acid. Sodium cyanotri-
hydroborate (0.63 g, 10.0 mmol) in methanol (5 mL) was added
dropwise to the resulting solution with stirring. After the solution
◦
was stirred for 3 days at 25 C, it was acidified by adding conc.
HCl and then evaporated almost to dryness under a reduced
pressure. The residue is dissolved in saturated aqueous Na2CO3
solution (50 mL) and extracted with CHCl3 (3 ¥ 50 mL). The
combined extracts were dried over anhydrous Na2SO4 and filtered.
Slow evaporation of the filtrate gives a white crystalline ligand
H2(L1). Yield: 0.84 g (49%). The ligand was further purified
by recrystallization from CH2Cl2. mp: 110–112 ◦C. 1H NMR
(200 MHz, CDCl3): d 1.27 (s, 6H), 1.42 (s, 4 H), 1.95 (pentet, 2 H),
2.77 (t, 4 H), 3.77 (s, 4H), 6.82 (d, 2H), 7.21 (d, 2H), 7.21 (s, 2H).13C
NMR (360 MHz, C6D6): d 20.91 (CH3), 27.06, 53.50, 54.81, 62.28
(CH2), 116.86, 122.29, 128.20, 129.74, 130.07, 156.97 (ArC).
Experimental section
Materials
Homopiperazine, 2-hydroxy-5-methylbenzaldehyde, 2,4-di-tert-
butylphenol, 3,5-di-tert-butylcatechol (H2DBC), 4-tert-butyl-
catechol (H2TBC), 2,4-dimethylphenol, sodium cyanotrihydrobo-
rate, catechol (H2CAT) and 4-nitrocatechol (H2NCAT) (Aldrich)
and iron(III) chloride (anhydrous) (Merck, India), 3,4,5,6-
tetrachlorocatechol (H2TCC, Lancaster) were used as received,
unless noted otherwise and H2DBC was recrystallized from hexane
before use. The supporting electrolyte tetra-N-butylammonium
perchlorate (NBu4ClO4, G. F. Smith, USA) was recrystallized
twice from aqueous ethanol.
1,4-Bis(2-hydroxy-4-nitrobenzyl)-1,4-diazepane H2(L2). The
synthesis of H2(L2) was carried out in three steps. The first
step in the synthesis of this compound involves preparation of
methylal,21,22 which was reacted with p-nitrophenol in the second
step to obtain 2-hydroxy-5-nitrobenzyl chloride. The latter was
then reacted with homopiperazine to yield H2(L2).
To a solution of 2-hydroxy-5-nitrobenzyl chloride (1.50 g,
8 mmol) in tetrahydrofuran (30 mL) were added homopiperazine
(0.40 g, 4 mmol) and triethylamine (0.81 g, 1.11 mL, 8 mmol). The
mixture was refluxed for 2 h to give a yellow suspension which
was cooled and filtered. The filtrate was rotaevaporated to give a
yellow solid that was washed thoroughly with methanol and then
used as such for synthesis of its complex. Yield: 0.87 g (54%). mp
Physical measurements
Elemental analyses were performed on a Perkin Elmer Series II
CHNS/O Analyzer 2400. 1H NMR spectra were recorded on
a Bruker 200 MHz NMR spectrometer. The electronic spectra
were recorded on an Agilent diode array-8453 spectrophotometer.
The EPR spectra were recorded on a JEOL JES-TE 100 X-
band spectrometer. ESI-Mass spectrometry was performed on a
Thermo Finnigan LCQ 6000 Advantage Max instrument. Cyclic
voltammetry (CV) and differential pulse voltammetry (DPV) were
performed using a three electrode cell configuration. A platinum
sphere, a platinum plate and Ag(s)/Ag+ were used as working,
auxiliary and reference electrodes respectively. The supporting
electrolyte used was NBu4ClO4 (TBAP). The temperature of
◦
1
148–150 C. H NMR (200 MHz, CDCl3): d 1.94 (pentet, 2 H),
2.81 (s, 4 H), 2.85 (t, 4 H), 3.85 (s, 4H), 6.84 (d, 2H), 8.03 (s, 2H),
8.07 (d, 2H). 13C NMR (400 MHz, CDCl3): d 26.57, 53.71, 54.10,
61.28 (CH2), 116.80, 121.51, 124.87, 125.74, 140.33, 164.59 (ArC).
1,4-Bis(2-hydroxy-3,5-dimethylbenzyl)-1,4-diazepane H2(L3).
A
solution of 2,4-dimethylphenol (3.66 g, 30.0 mmol),
homopiperazine (1.50 g, 15.0 mmol) and 37% aqueous
formaldehyde (3.5 mL, 42.0 mmol) in methanol (10 mL) was
stirred and refluxed for 24 h. The mixture was cooled and the
product obtained was filtered off and washed with ice-cold
methanol to give a colorless product. Yield: 4.20 g (76%). The
product◦ was purified by recrystallization from CH2Cl2. mp:
the electrochemical cell was maintained at 25.0
0.2 ◦C by
a cryocirculator (HAAKE D8 G). By bubbling research grade
nitrogen the solutions were deoxygenated and an atmosphere of
nitrogen was maintained over the solutions during measurements.
The E1/2 values were observed under identical conditions for
various scan rates. The instruments utilized included an EG &
G PAR 273 Potentiostat/Galvanostat and Pentium-IV computer
along with EG & G M270 software to carry out the experiments
and to acquire the data. The product analyses were performed
using a HP 6890 GC series Gas Chromatograph equipped with
a FID detector and a HP-5 capillary column (30 m ¥ 0.32 mm ¥
1
98–100 C. H NMR (200 MHz, CDCl3): d 1.89 (pentet, 2 H),
2.17 (s, 12H), 2.74 (s, 4 H), 2.79 (t, 4H), 3.70 (s, 4H), 6.83 (s, 2H),
6.58 (s, 2H), 7.23 (s, 2H, OH). 13C NMR (360 MHz, CDCl3):
d 15.56, 20.34 (CH3), 26.46, 53.37, 54.36, 61.72 (CH2), 120.63,
124.60, 126.61, 127.61, 130.62, 153.51 (ArC).
1,4-Bis(2-hydroxy-3,5-di-tert-butylbenzyl)-1,4-diazepane H2(L4).
A solution of 2,4-di-tert-butylphenol (5.0 g, 24.2 mmol), ho-
mopiperazine (1.21 g, 12.1 mmol) and 37% aqueous formaldehyde
(2.5 mL, 33.6 mmol) in methanol (10 mL) was stirred and refluxed
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9613
©

View Article Online
Table 1 Crystal data and structure refinement for [Fe(L3)Cl] 3
for 24 h. The mixture was cooled and the product was filtered off
and washed with ice-cold methanol to give a colorless product.
Yield: 4.71 g (72%). The pro◦duct was purified by recrystallization
Empirical formula
C23H29ClFeN2O2
1
from CH2Cl2. mp: 173–175 C. H NMR (200 MHz, CDCl3): d
FW
456.78
Orthorhombic
Cmc21
20.257(2)
15.3291(17)
7.3121(8)
90
2270.5(4)
273(2)
0.71073
1.336
1.27 (s, 18 H), 1.42 (s, 18 H), 1.94 (pentet, 2H), 2.77 (s, 4 H),
2.82 (t, 4 H), 3.77 (s, 4H), 6.82 (s), 7.21 (s, 2H), 7.26 (s, 2H, OH).
13C NMR (400 MHz, CDCl3): d 26.76, 29.61, 31.70 (CH3), 34.15,
34.86, 53.07, 54.45 (CH2), 62.45 (tert-C), 121.22, 123.01, 123.44,
135.66, 140.59, 154.29 (Ar–C).
Crystal system
Space group
˚
a/A
˚
b/A
˚
c/A
a = b = g (◦)
3
˚
V/A
T/K
˚
Synthesis of complexes
Mo-Ka l, (A)
density/Mg m-3
Z
[Fe(L1)Cl] 1. The complex [Fe(L1)Cl] 1 was prepared by
adding of a solution of FeCl3 (0.16 g, 1.0 mmol) in methanol
(5 mL) to an acetonitrile solution (10 mL) of an equivalent amount
of the ligand H2(L1) (0.34 g, 1.0 mmol). The solution was stirred
for an hour. The blue colored precipitate obtained was filtered off,
washed with small amounts of cold methanol and dried in vacuo
over P4O10. Yield: 0.18 g (42%). Anal. Calcd. for C21H26N2O2FeCl:
C, 58.69; H, 6.40; N, 6.52. Found C, 58.72; H, 6.39; N, 6.53.
4
m/mm-1
0.802
960
6558
1.144
0.0522
0.1350
F(000)
no. of reflections collected
Goodness-of-fit on F2
a
R1
b
wR2
ꢀ
ꢀ
a R1 = ꢀFo| -|Fcꢀ/R|Fo|, b wR2 = w{(Fo - Fc2)2/Rw [(Fo2)2]}
2
1/2
[Fe(L2)Cl(H2O)] 2. This complex was prepared as a dark
brown colored product by using the procedure employed for
preparing complex 1. Yield: 0.26 g (52%). Anal. Calcd. for
C19H22N4O7FeCl: C, 44.77; H, 4.35; N, 10.99. Found C, 44.75;
H, 4.38; N, 10.72. ESI-MS, m/z, 508.
The dioxygenase activities of the present complexes were
determined using a known22,31,34 procedure with modifications.
The complex (0.1 mmol), H2DBC (0.022 g, 0.1 mmol), and Et3N
(0.20 g, 28 mL, 0.2 mmol) were dissolved in DMF (5 mL) and
exposed to dioxygen and stirred for 48 h. The oxygenation reaction
was quenched by the addition of 6 M HCl solution (5 mL).
The products were extracted from the aqueous solution with
diethyl ether (3 ¥ 30 mL). The clear yellow organic layer was
separated, washed twice with 2 M HCl (20 mL) and then dried
over anhydrous Na2SO4 at room temperature and then filtered off
and the combined filtrate was evaporated in vacuo, which yields
the products. The products were analyzed by GC-MS (EI) and
quantified by GC (FID). The cleavage products were quantified
by comparing the GC retention times of the reaction products with
those of authentic samples prepared by using iron(III) complexes
reported by us previously.31,34
[Fe(L3)Cl] 3. A methanolic slurry (4 mL) of the ligand H2(L3)
(0.37 g, 1.0 mmol) was treated with FeCl3 (0.16 g, 1.0 mmol) in
methanol (4 mL) and then triethylamine (0.20 g, 280 mL, 2.0 mmol)
in methanol (2 mL) was added. The mixture was stirred for
30 min. The dark blue coloured precipitate obtained was filtered
off, washed with small amounts of cold methanol and dried over
P4O10. Yield: 0.33 g (72%). Anal. Calcd. for C23H30N2O2FeCl: C,
60.34; H, 6.60; N, 6.12. Found C, 60.30; H, 6.82; N, 6.3. The
complex was recrystallized from hot acetonitrile to give a dark
blue crystalline solid. The solid was refluxed in CH3OH–CH3CN
(1 : 1) mixture for 10 min and upon cooling the solution at room
temperature bright blue crystals suitable for X-ray diffraction were
obtained.
Single-crystal X-ray data collection and structure solution
[Fe(L4)Cl] 4. The complex 4 was prepared by the reaction of a
methanolic (4 mL) slurry of the ligand H2(L4) (0.54 g, 1.0 mmol)
with FeCl3 (0.16 g, 1.0 mmol) in methanol (4 mL) in the presence
of Et3N (0.20 g, 280 mL, 2.0 mmol) in methanol (2 mL). The
mixture was stirred for 30 min. The dark blue crystalline precipitate
obtained was filtered off, washed with small amounts of cold
methanol and dried over P4O10. Yield: 0.49 g (78.72%). Anal.
Calcd. for C35H54N2O2FeCl: C, 67.14; H, 8.69; N, 4.47. Found
C, 67.16; H, 8.62; N, 4.50.
A single crystal of 3 of suitable size was selected from the mother
liquor and immersed in paraffin oil, then mounted on the tip of
a glass fiber and cemented using epoxy resin. Intensity data for
˚
the crystal were collected using Mo-Ka (l = 0.71073 A) radiation
on a Bruker SMART APEX diffractometer equipped with a CCD
area detector at 273 K. The crystallographic data are collected
in Table 1. The SMART38 program was used for collecting
frames of data, indexing reflection, and determination of lattice
parameters; SAINT38 program for integration of the intensity
of reflections and scaling; SADABS39 program for absorption
correction, and the SHELXTL40 program for space group and
structure determination, and least-squares refinements on F2.
The structure was solved by heavy atom method. Other non-
hydrogen atoms were located in successive difference Fourier
syntheses. The final refinement was performed by full-matrix
least-squares analysis. Hydrogen atoms attached to the ligand
moiety were located from the difference Fourier map and refined
isotropically.
Kinetics and reactivity studies
Kinetic analyses21–29,31 of the catechol cleavage reactions were
carried out by time-dependent measurement of the disappearance
of the lower energy DBC2--to-iron(III) LMCT band at ambient
temperature (25 ◦C) by exposing to molecular oxygen. The solvents
were equilibrated at the atmospheric pressure of O2 at 25 ◦C and
the solubility of O2 at 25 ◦C is: acetonitrile, 8.1 ¥ 10-3 M;28,36 DMF
(dimethylformamide), 4.86 ¥ 10-3 M.36,37
9614 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
a
a
˚
Table 2 Selected bond lengths [A] and bond angles [deg] for [Fe(L3)Cl]
Results and discussion
3
Synthesis and characterization of ligands and complexes
Fe(1)–N(1)
2.202(4)
Fe(1)–O(1)
Fe(1)–Cl(1)
1.855(3)
2.2470(18)
98.2(2)
87.26(14)
147.62(16)
72.36(18)
107.45(12)
101.13(11)
135.5(3)
The present ligands were synthesized according to known
procedures,21,35,41 which involve Mannich condensation, reduc-
tive amination and simple substitution reactions, with suitable
modifications. The amino hydrogens in homopiperazine have
been replaced with differently substituted phenolate moieties
to generate the ligands (Scheme 2) for the present study. The
tetradentate ligand H2(L1) was synthesized by the reductive am-
ination of 2-hydroxy-5-methylbenzaldehyde using sodium cyan-
otrihydroborate as the reducing agent.35 The ligand H2(L2) was
synthesized21 by substitution reaction of homopiperazine with 2-
hydroxy-5-nitrobenzyl chloride. The ligands H2(L3) and H2(L4)
were synthesized41 in good yields by the reaction of the Mannich
base formed in situ by ortho and para substituted phenols using
excess of 37% aqueous formaldehyde. The N2O2 donor set of the
present ligands mimics the metal coordinating side chains of amino
acids (histidine and tyrosinate residues) in 3,4-PCD enzymes. The
two phenolate [pKa (BH+): phenol, 9.95] arms with electron-
releasing methyl and tert-butyl and electron-withdrawing -NO2
groups in the ligands are expected not only to duplicate tyrosinate
residues in the active site but also mimic the function of 3,4-
PCD. The methyl and tert-butyl substituents have been introduced
not only to tune the spectral and electrochemical properties but
also to effect a variation in reactivity of the model complexes
by providing steric hindrance to O2 attack. The 2,4-disubstituted
phenolate moiety has been shown21,33 to favor the formation of
monomeric five-coordinate complexes but with varying structural,
spectral and electrochemical properties. The reaction of FeCl3
and the bis-phenolate ligands in the presence of two equivalents
of Et3N results in the immediate formation of the complexes.
The complexes 1, 3 and 4 are formulated as [Fe(L)Cl], which is
supported by the X-ray crystal structure of one of them (3, cf.
below), while 2 is formulated as [Fe(L2)(Cl)(H2O)] based on the
ESI-MS and elemental analysis. Attempts to grow single crystals
of the latter were unsuccessful. The conductivities of all the bis-
phenolate complexes (KM, 6–13 X-1 cm2 mol-1) suggest that the
chloride ion is certainly coordinated in acetonitrile solution.22–25
The five-coordinate complexes 1, 3 and 4 containing only one
coordinated chloride ion may expand the coordination sphere to
accommodate the bidentate catecholate substrates. All the iron(III)
complexes have magnetic moments in the range 5.6–5.8 BM at
room temperature, which is characteristic of a high-spin ferric
center with five unpaired electrons.22-29,31 The catecholate adducts
of the iron(III) complexes were generated in solution for spectral
and reactivity studies.
O(1a)–Fe(1)–O(1)
O(1)–Fe(1)–N(1)
O(1)–Fe(1)–N(1a)
N(1a)–Fe(1)–N(1)
O(1)–Fe(1)–Cl(1)
N(1)–Fe(1)–Cl(1)
C(1)–O(1)–Fe(1)
a Standard deviations in parenthesis.
Fig. 1 Molecular structure of complex [Fe(L3)(Cl)] 3 (40% probability
factor for the thermal ellipsoid). Hydrogen atoms have been omitted for
clarity.
bisphenolate ligand are bonded to iron(III) at the four corners
of the basal plane of the square pyramid and the chloride ion
occupies the apical site. The observed Fe–Ophenolate bond distances
are equal (Fe–O, 1.855 A) and are nearly identical to the Fe–Otyr
bond distances in 3,4-PCD enzymes (Fe–Otyr447 (axial), 1.91 A,
˚
˚
15,16,19
˚
Fe–Otyr408 (equatorial), 1.81 A).
However, the Fe–Ophenolate
bond length in 3 is shorter than the average octahedral Fe–O
bond distance21–23,31,33 of ~1.92 A, which is consistent with the
˚
high molar absorptivity of the PhO- → Fe(III) LMCT band
around 541 nm (cf. below). This illustrates that the pp orbital
of phenolate oxygen atom interacts43 strongly with the half-filled
dp* orbital of iron(III) in the complex. The Fe–Cl bond distance
(2.247 A) falls within the range21,33 for the five-coordinate iron(III)
˚
complexes suggesting that the chloride ion is less likely to be labile
˚
in solution (cf. above). The Fe–Namine bond distance (2.202 A) is
almost close to the Fe–Nhis bond distance in 3,4-PCD enzyme15,19
˚
(Fe–NHis462, 2.26, Fe–NHis460, 2.33 A) but is shorter than that in
five-coordinate bis-phenolate iron(III) complexes21 and the average
III
˚
Fe –N bond distance (~2.15 A) observed for octahedral iron(III)
complexes.21–26,31 The Fe–O–C bond angle (Fe–O1–C1, 135.5◦)
is also close to that in 3,4-PCD enzyme15,19 (Fe–O–Ctyr408, 133
(equatorial); Fe–O–Ctyr447 148◦ (axial)). It is closer to that in
octahedral iron(III) complexes of monophenolate ligands22,31 with
sterically demanding -NMe2 group but it is much higher than
those in octahedral iron(III) complexes of tripodal monophenolate
ligands (~128.5◦).22,25b It has been argued that the magnitude of
this angle is a key factor to elicit catechol dioxygenase activity in
intradiol-cleaving enzymes19 and their model complexes.22,31
The square pyramidal coordination geometry of 3 is similar
to that of the complex [Fe(Mes6-SALEN)(Cl)] 10 reported by
Description of the X-ray crystal structure of [Fe(L3)Cl] 3
The molecular structure of complex 3, which looks like a butterfly,
is depicted in Fig. 1 together with the atom numbering scheme.
The selected bond lengths and bond angles are collected in Table 2.
The coordination environment around the iron atom is described
as distorted square pyramidal with the trigonality index42 t of 0.20
[t = (b-a)/60, where b = O1–Fe–N1a = 147.62◦ and a = C1–O1–
Fe = 135.5◦; for perfect square pyramidal and trigonal bipyramidal
geometries the t values are zero and unity respectively]. The
two phenolate oxygens and two tertiary amine nitrogens of the
Fujii et al.33 The Fe–Namine bond distance in 3 (2.202 A) is
longer than the Fe–Nimine bond distance in 10 (2.087, 2.102 A),
˚
˚
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9615
©

View Article Online
Table 3 Electronic spectral data lmax, nm (e, M-1 cm-1) for iron(III)
which is expected of sp3 and sp2 hybridizations44 respectively of
the tertiary amine and imine nitrogen atoms. However, the Fe–
complexesa and their adducts in acetonitrile–DMF solution
˚
Ophenolate bond distance in 3 (1.855 A) is shorter than that in 10
[Fe(L2)-
˚
(1.881 A) suggesting that the iron-oxygen overlap in the former
Solvent Added
catecholsb
Fe(L1)Cl]
(H2O)Cl]
[Fe(L3)Cl] [Fe(L4)Cl]
is stronger. Interestingly, on replacement of the chloride ion in 10
by a water molecule to obtain [Fe(Mes6-SALEN)(OH2)](ClO4) a
structural change from preferred square pyramidal to distorted
trigonal bipyramidal geometry (t, 0.48) is observed and the latter
resembles that found in the active site (t, 0.44) of 3,4-PCD
enzyme in the resting state. So, it is expected that substitution
of chloride ion in 3 by a water molecule would lead to a
structural change from square pyramidal to a distorted trigonal
bipyramidal geometry as in 3,4-PCD enzyme. It is to be noted
that the five-coordinate iron(III) complex [Fe(L5)Cl] possesses a
trigonal bipyramidal coordination geometry (t, 0.79)21 and closely
resembles the iron(III) coordination environment of the 3,4-PCD
enzyme active site. Further, the coordination geometry around
each iron atom in [Fe2(m-O)(L9)2] with a Fe–O–Fe core is distorted
square pyramidal (t, 0.45) with the two amine nitrogen and two
phenolate oxygen atoms of the ligand constituting the corners
of the square plane and the m-oxo oxygen atom (O3) occupying
the axial position.35 A square-pyramidal coordination geometry
similar to 3 is suggested for the analogous complexes 1 and 4
based on stoichiometry and spectral properties (cf. below). On the
other hand, the complex [Fe(L2)(Cl)(H2O)] 2 is proposed to have
an octahedral coordination geometry in which the tetradentate
ligand L2 occupies the four equatorial sites around iron(III) and
the chloride ion and water molecule occupy the two labile sites
trans to each other. A similar coordination structure has been
proposed34 for the complex [Fe(L)Cl2]Cl, where L is the linear
ligand 1,4-bis(2-pyridylmethyl)-1,4-diazepane.
CH3CN None
517 (1 960)
335 (2 700)
283 (5 100)
238 (5 630)
200 (14 530)
—
541 (3 180) 540 (7 670)
335 (3 880) 339 (9 650)
284 (6 610) 282 (15 180)
239 (7 140) 241 (16 740)
DBC2-
TBC2-
548 (2 050)
468 (2 340)
375 (2 410)
213 (sh)
—
—
—
540 (5 380) 645 (7 320)
412 (4 910) 341 (3 020)
333 (sh)
286 (sh)
309(4300)
277 (8 450)
533 (1 810)
331 (3 450)
284 (6 700)
545 (19 980)
505 (5 640) 636 (2 680)
332 (8 840) 430 (18 780)
282 (16 570) 374 (sh)
304 (10 280)
TCC2-
516 (2 880)
307 (sh)
304 (10 070)
502 (5 600) 575 (2 730)
332 (sh)
491 (sh)
282 (17 030) 425 (19 990)
DMF None
493 (3 000) 499 (3 530) 536 (3 010) 551(sh)
327 (6 000) 383 (7 640) 329 (6 760) 501(3 820)
288 (11 320) 345 (9 650) 284 (11 520) 326 (6 730)
238 (16 980)
288 (11 100)
DBC2-
530 (2 950) 656 (1 030) 541 (2 980) 539 (3 130)
462 (3 140) 515 (1 330) 451 (3 330) 451 (3 950)
332 (sh)
300 (8 780) 331 (7 190) 375 (4 630)
287(15 850)
292 (14 160)
TBC2-
TCC2-
—
513 (2 160) 523 (3220) 504 (2870)
337 (sh)
287 (9 110)
330(sh)
327(5330)
—
499 (2 850) 508 (4490) 489 (3660)
305 (7 490) 329(sh)
264 (sh)
Electronic absorption and EPR spectra
335(sh)
305(sh)
All the bis(phenolate) complexes exhibit two intense LMCT bands
(493–551, 327–501 nm, Table 3) in DMF solution. While the low
energy band is assigned to the charge transfer transition from the
pp orbital (HOMO) of the phenolate ion to the half-filled dp*
orbital of iron(III), the high energy band is assigned to the charge
transfer transition from the pp orbital (HOMO) of the phenolate
oxygen to the half-filled d-orbital of iron(III) (ds*).19,21,22,45 The
more intense band observed in the range 200–284 nm is assignable
to ligand-based transitions. Interestingly, the bands observed for
1, 3 and 4 in DMF solution are higher in energy than those (517–
541, 335–339 nm, Fig. 2, Table 3) in acetonitrile solution. It is
well known that addition of a base increases the negative charge
built on iron(III) leading to a raising in energy of the iron(III)
d-orbitals and thus shifts the phenolate-to-iron(III) LMCT band
to higher energies with or without change in absorptivity.25 So
it is suggested that DMF is coordinated to the iron(III) center
of the complexes. The energy of the low energy LMCT band in
DMF varies in the order 1 > 3 ª 4 > 2 reflecting the effect of
phenolate ring substituents on the Lewis acidity of the iron(III)
center. The incorporation of an electron-releasing methyl group
at the ortho position of the phenolate ring in 1 to obtain 3
would be expected to raise25a the energy of the POMO orbital
of the phenolate moiety and hence lead to a decrease in energy
of the lowest energy phenolate (pp) to iron(III) (dp*) LMCT
band.19 A similar decrease in LMCT band energy is observed
a Concentration of iron(III) complexes, 2 ¥ 10-4 M. b The ratio of added
ligand to iron complexes was 1 : 1; the anions were generated by adding
2 equivalents of triethylamine.
on replacing the methyl groups in 3 by tert-butyl groups to give 4.
Also, replacement of the electron-releasing methyl group on the
phenolate moiety in 1 by the electron-withdrawing nitro group
to give 2 would be expected to lower25a the energy of the POMO
orbital of the phenolate moiety and hence lead to a lower energy for
the lowest energy phenolate(pp)-to-iron(III) (dp*) LMCT band.
The sensitivity of the band positions as well as their intensities to
the iron(III) environment, that is the nature of the ligand donor
functionalities, is reminiscent of the spectral features observed for
the native CTD and PCD enzymes.19
When H2DBC pretreated with two equivalents of Et3N is
added to [Fe(L2)(Cl)(H2O)] 2 in DMF two well-defined bands
(656, 515 nm) are observed, which are assignable to DBC2--to-
Fe(III) LMCT transitions21–29,31 involving two different catecholate
orbitals on the chelated DBC2-. A similar observation of two
catecholate-to-iron(III) LMCT bands has been made for the
catecholate adducts of iron(III) complexes of tridentate 3N,23,44
tetradentate 4N28,34,37 and monophenolate ligands.22,31 The high
energy DBC2--to-Fe(III) LMCT band would have merged with
9616 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
added to [Fe(L)Cl3] (L = tridentate 3 N ligand) or [Fe(L)Cl2]
(L = tridentate monophenolate ligand) two DBC2- bands appear
even in the absence of a base suggesting spontaneous deproto-
nation of H2DBC upon interaction with the highly Lewis acidic
iron(III) center. Upon adding increasing amounts of the strong
base piperidine the intensity of the shoulder at 515 nm decreases
and a new sharp band around 656 nm appears and starts growing
in intensity. When the absorbance of the 515 nm feature is plotted
as a function of number of equivalents of piperidine, a sharp
inflexion is observed at one equivalent of base (Fig. 4). However,
interestingly, when the absorbance of the 656 nm band is plotted,
two inflexions are noted around one and two equivalents of
piperidine. Similar results are obtained when Et3N is used as
the base; however, more than one equivalent but less than two
equivalents of the weaker base is needed to obtain the inflexion
Fig. 2 Electronic absorption spectra of iron(III) complexes (2.0 ¥
10-4 M) in acetonitrile [Fe(L1)Cl] (a); [Fe(L3)Cl] (b); [Fe(L4)Cl] (c) and
[Fe(L2)Cl(H2O)] (d) in DMF solution, 1.0 ¥ 10-4 M.
the intense phenolate-to-iron(III) LMCT band observed for 2
around 499 nm, which is now blue-shifted21 to lower energy
upon adduct formation. It is evident that upon adding the cat-
echolate dianion to 2 the sterically constrained linear tetradentate
ligand H2(L2) rearranges21,34 itself to provide cis-coordination
positions for bidentate coordination of the catecholate dianion
(Fig. 3). A similar rearrangement of the related linear 4N
ligand 1,4-bis(2-pyridinylmethyl)-1,4-diazepane (L) coordinated
to iron(III) to give a cis-b configuration has been established in the
X-ray crystal structure of the adduct [Fe(L)(TCC)]+, where TCC2-
is tetrachlorocatecholate.34 In order to understand the adduct
formation in more detail 2 was treated with one equivalent
of H2DBC, an increasing amount of piperidine or Et3N was
added and the changes in spectra are observed. Only very
small spectral changes were observed upon adding the catechol
to the complex suggesting negligible interaction of catechol
with the complex. This is expected because when H2DBC is
Fig. 4 Electronic absorption spectral changes observed during addition
of piperidine to [Fe(L2)(H2O)Cl] 2 (2 ¥ 10-4 M) and H2DBC (2 ¥ 10-4 M)
mixture in DMF. Inset: Plot of absorbance vs number of equivalents of
added base at 656 nm.
Fig. 3 Schematic illustration of mono- (a) and bidentate coordination of DBC2- to iron(III) center.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9617
©

View Article Online
for the changes in 515 nm band and only one well-defined
inflexion but at two equivalents of base is observed for the
disappearance of the 656 nm band (Fig. 4). So it is clear that the
adduct species [Fe(L2)(HDBC)] in which HDBC- is coordinated
in bidentate fashion to iron(III) and/or [Fe(L2)(HDBC)(H2O)] in
which HDBC- is monodentatively coordinated, is formed in DMF
solution and that this species reacts with one more equivalent of
piperidine to form the species [Fe(L2)(DBC)]- in which DBC2- is
coordinated in bidentate fashion.
Further, the energies of both the DBC2--to-Fe(III) LMCT
bands of the adducts [Fe(L)(DBC)]- decrease along the series
[Fe(L1)(DBC)]- > [Fe(L3)(DBC)]- > [Fe(L2)(DBC)]- revealing
a remarkable dependence of the bands on the nature of the
primary ligands21–29 and reflecting the decrease in Lewis acidity
of the iron(III) center of the adducts along this series, as modi-
fied by ligand substituents. Thus incorporation of the electron-
withdrawing nitro group would tend to stabilize the dp* orbital
and hence decrease the energy gap between the metal orbital and
the DBC2- orbital leading to a lower energy for the catecholate-
to-iron(III) LMCT band.21,25a In contrast, the electron-releasing
methyl groups would destabilize the dp* orbital leading to a
higher energy for the LMCT band. Also, the position of the
low energy catecholate → Fe(III) LMCT band in the complex-
substrate adducts [Fe(L)(catecholate)]- generated from [Fe(L)Cl]
is found to be shifted to lower energies as the substituents21–25,31,46
on the catecholate ring are varied from electron-releasing to
In contrast to 2, when H2DBC pretreated with two equivalents
of Et3N is added to 1 both in DMF and acetonitrile solutions,
a less intense broad band around 462 (DMF)/468 nm (CH3CN)
appears along with a low energy broad band, which corresponds
to the DBC2--to-iron(III) LMCT bands21,31 (Table 3). The intense
phenolate-to-iron(III) LMCT band observed for 1 around 530
(DMF)/548 nm (CH3CN) appears to have merged21 with the high
energy DBC2--to-Fe(III) LMCT band. Similar observations are
made for the addition of DBC2- to 3 with the appearance of a
shoulder around 412 nm and a low energy band with low (DMF)
or high (CH3CN) intensity. The adduct formation of [Fe(L3)Cl] is
followed by monitoring the decrease in intensity of the phenolate-
to-iron(III) LMCT band around 535 nm and the growth of the
high energy catecholate to iron(III) LMCT band around 451 nm
as well, two inflexions at one and two equivalents of piperidine
but only one but less well defined inflexion at two equivalents of
Et3N is observed. Upon adding DBC2- to 4 in both CH3CN and
DMF a band around 451 nm with a low energy shoulder around
539 nm are observed revealing the catecholate adduct formation.
When 4 is treated with one equivalent of H2DBC, and then an
incremental amount of the bases are added, the changes observed
for the disappearance of the 499 nm band are found to be small,
and only one but less well-defined inflexion at two equivalents of
base is observed when the growth of the shoulder around 375 nm
is monitored. It is obvious that bidentate coordination of DBC2-
to iron(III) complexes of the linear bis(phenolate) ligands H2(L1)–
H2(L4) to give six-coordinate adducts with cis-b configuration (cf.
above) would involve breaking of a Fe–Ophenolate bond, followed
by ligand rearrangement, and formation of a Fe–Ophenolate bond.31
Such a geometrical rearrangement in 2 would be more facile than
in 1, 3 and 4 as the electron-withdrawing p-nitro substituent in 2
would render the Fe–Ophenolate bond weaker. Upon incorporation
of one more methyl group in 1 to obtain 3 the Fe–Ophenolate bonds
become stronger (cf. below), illustrating the lower ability of the
latter complex to form catecholate adducts. And upon replacing
the methyl substituents in 3 by the more electron-releasing tert-
butyl substituents on the phenolate rings to obtain 4 the Fe–
Ophenolate bond becomes much stronger illustrating the enhanced
reluctance of the complex to form catecholate adducts. Thus
structural changes from square pyramidal to octahedral geometry
to accommodate the bidentatively coordinated DBC2- would be
more difficult than simple substitution reactions in octahedral
complexes.31 Molecular model building studies reveal that the
re-coordination of the displaced phenolate oxygen in the vacant
axial position is not facile, particularly for 4. So, it is evident
that monodentate coordination of the substrate (DBC2-) by
replacing the axially coordinated chloride ion followed by weaker
coordination of the other catecholate oxygen would occur (Fig. 3)
and hence an equilibrium between mono- and bidentatively
coordinated adducts may be present in solution (cf. below).
electron-withdrawing as observed previously: TCC2- > TBC2-
>
DBC2-. This is expected as the electron-releasing substituents on
the catecholate ring would decrease while electron-withdrawing
substituents enhance the position of the low energy band,21–25,31,46
thus reflecting the importance of electronic effects provided by the
substituents on catechols.
It may be noted that the structure of the mono(m-O)diiron(III,III)
complex of the parent ligand of L1 has been determined.35 In order
to show that the complexes do not dimerize under the conditions of
catecholate adduct formation the DMF solutions of the complexes
were treated with increasing amounts of the bases and the spectral
changes monitored. The color of solution changed from violet
to red–orange. Interestingly, inflexions are observed at one and
two equivalents of added base both for the disappearance of
the phenolate-to-LMCT band and for the growth of the band
around 427 nm for [Fe(L3)Cl] and that around 501 nm (Fig. 5) for
[Fe(L4)Cl] suggesting the formation of a dimer with the Fe2O core
in two steps as follows:
[FeIII(L)Cl] + H2O → [FeIII(L)(H2O)] + Cl-
Fig. 5 Electronic absorption spectral changes observed during addition
of piperidine to [Fe(L3)Cl] 3 in DMF. Inset: Plot of absorbance vs number
of equivalents of added base for the formation of the absorption band
around 427 nm (a) and disappearance of the band around 540 nm (b).
9618 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 4 EPR spectral data for iron(III) complexes in methanol/acetone
to an increase in g value and rhombicity. Also, on substituting
the methyl groups in the ligand in 3 by tert-butyl groups to
obtain 4 an appreciable decrease in g value (8.50) and rhombicity
(E/D: 3, 0.100; 4, 0.092) is observed. This suggests that stronger
coordination of the more basic bis-phenolate ligand H2(L4) (cf.
below) offsets the steric constraint at the iron(III) center and
confers a less rhombic coordination environment on iron(III).
solution at 77 K
Compound
[Fe(L1)Cl]
g values
E/D
0.095
0.087
0.10
8.70, 4.10, 2.10
4.80a, 2.01, 1.99,1.89
8.22, 4.01,
[Fe(L2)Cl(H2O)]b
[Fe(L3)Cl]
4.83a
9.03, 4.09, 2.0
4.90, 1.95
Electrochemical properties
[Fe(L4)Cl]
8.50, 4.10, 2.10
0.092
The redox behavior of the iron(III) complexes and their
catecholate adducts generated in situ was studied by employing
Cyclic Voltammetry (CV) and Differential Pulse Voltammetry
(DPV) on a stationary platinum sphere as working electrode.
The electrochemical responses of the complexes are well-behaved
in DMF but not in acetonitrile. All the complexes exhibit a
cathodic (-0.624 to -1.164 V) as well as an anodic wave (-0.540
to -0.996 V, Table 5, Fig. 7, 8) in DMF. The cathodic current
functions (D, 1.0–3.2 ¥ 10-6 cm2 s-1) calculated by substituting
5.0a, 2.01, 1.99, 1.98,1.88
a Minor component. b DMF/acetone solution.
[(L)FeIII-OH2] + Pip ꢀ [(L)FeIII-OH] + PipH+
[(L)FeIII --OH + HO--FeIII (L)] ⎯P⎯iP→ [(L)FeIII --O--FeIII (L)]+ H2O
The frozen solution EPR spectra of all the complexes display
high-spin (S = 5/2) rhombic ferric signals31,33,46 at g = 8.22–
9.03, 4.01–4.10 (Fig. 6, Table 4) and 2.0–2.1 (E/D, 0.087–0.100)
associated with the |5/2, -1/2 > → |5/2, +1/2 > transition
and also signals corresponding to rhombic low-spin47,48 (S =
1/2) iron(III) species (g, 4.8, 2.01, 1.99, 1.89). Upon replacing
the two electron-releasing p-methylphenolate arms in 1 (g, 8.70;
E/D, 0.095) by the electron-withdrawing p-nitrophenolate arms
to obtain 2 a remarkable decrease in both the g value (8.22) and
rhombicity (E/D, 0.087) is observed, which is consistent with
the change in coordination geometry from square pyramidal to
octahedral. In contrast, on incorporation of the methyl groups at
ortho positions of the ligand in 1 to obtain 3 an appreciable increase
in g value (9.03) and rhombicity (E/D, 0.10) is observed. This
suggests that stronger Fe–O bond formation (cf. below) followed
by an increase in steric constraint at the iron(III) center leads
1/2
the slope obtained from the linear ipc vs. n (n < 0.5 Vs-1) plot
in the Randles–Sevcik’s equation49 are of the same order as those
observed for other iron(III) complexes undergoing a one-electron
reduction process.22–25,31 The E1/2 values of the Fe(III)/Fe(II) redox
couple of the complexes follow the trend 2 > 1 > 3 > 4. On
replacing the electron-releasing methyl groups on the phenolate
arms in 1 by the electron-withdrawing nitro groups as in 2 the
Fig. 7 Differential pulse voltammograms of 1 mM [Fe(L3)Cl] before (a)
and after addition of 1 mM H2DBC (b) and 2 mM Et3N (c) in DMF at
25 ◦C. Supporting electrolyte: 0.1 M TBAP; Scan rate: 5 mV s-1; Reference
electrode: Ag/Ag+; working electrode: Pt sphere.
Fig. 8 Differential pulse voltammograms of 1 mM [Fe(L2)(H2O)Cl]
before (a) and after addition of 1 mM H2DBC (b) and 2 mM Et3N (c)
in DMF at 25 ◦C. Supporting electrolyte: 0.1 M TBAP; Scan rate: 5 mV
s-1; Reference electrode: Ag/Ag+; working electrode: Pt sphere.
Fig. 6 EPR spectra of the iron(III) complexes [Fe(L1)Cl] 1 (a) and
[Fe(L2)(H2O)Cl] 2 (b) in frozen acetonitrile/acetone and DMF/acetone
solutions respectively at 77 K.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9619
©

View Article Online
Table 5 Electrochemical dataa for the FeIII/FeII redox process of iron(III) complexes and their DBC2- adductsb in DMF at 25 0.2 ◦C obtained using
scan rates of 50 mV s-1 (CV) and 5 mV s-1 (DPV)
E
1/2/V
Compound
Epc/V
E
pa/V
DEp/mV
CV
DPV
Redox process
[Fe(L1)Cl]
+ DBC2-
-0.931
-0.977
-0.573
-0.624
-0.642
-0.326
-0.874
-0.330
-0.909
-0.877
-0.615
-0.901
-0.601
-1.164
-1.416
-0.737
-0.680
-0.843
-0.465
-0.540
-0.534
-0.246
—
-0.218
-0.759
-0.781
-0.533
-0.861
-0.513
-0.996
-1.152
-0.573
251
134
108
84
108
80
-0.805
-0.910
-0.519
-0.582
-0.588
-0.286
—
-0.274
-0.834
-0.829
-0.574
-0.881
-0.557
-1.080
-1.284
-0.655
-0.808
-0.932
-0.522
-0.577
-0.579
-0.279
-1.020
-0.285
-0.836
-0.850
-0.560
-0.850
-0.568
-1.010
-1.308
-0.645
FeIII →FeII
FeIII →FeII + LR
DBSQ→H2DBC
FeIII → FeII
[Fe(L2)Cl(H2O)]c
+ H2DBC
FeIII → FeII
DBSQ → H2DBC
FeIII → FeII + LR
DBSQ → H2DBC
FeIII → FeII
+ DBC2-
—
112
150
96
82
40
[Fe(L3)Cl]
+ H2DBC
FeIII → FeII + LR
DBSQ → H2DBC
FeIII → FeII + LR
DBSQ → H2DBC
FeIII → FeII
+ DBC2-
96
[Fe(L4)Cl]
+ DBC2-
168
264
164
FeIII → FeII + LR
DBSQ → H2DBC
a Potential measured vs. Ag/AgNO3 (0.01 M, 0.1 M TBAP); add 0.544 V to convert to NHE. b Generated by adding two equivalents Et3N to H2DBC.
LR = ligand reduction.
Fe(III)/Fe(II) redox potential (E1/2, -0.577 V) becomes more
positive suggesting the enhancement in Lewis acidity of the
iron(III) center as expected. Upon incorporation of one more
methyl group at the ortho position of the phenolate arm in 1
to obtain 3 and upon replacement of the methyl groups in 3
by the more electron-releasing tert-butyl groups to form 4 the
Fe(III)/Fe(II) redox potential becomes more negative reflecting
the decrease in Lewis acidity of the iron(III) center.
Catechol dioxygenase activity of iron(III) complexes
When the 3,5-di-tert-butylcatecholate (DBC2-) adducts of all the
complexes were generated in situ in DMF solution by reacting
the complexes with H2DBC pretreated with two equivalents of
Et3N/piperidine and their reactivity towards O2 was investigated
by monitoring the decay of the catecholate-to-iron(III) LMCT
band (1, 530; 2, 656; 3, 541; 4, 539 nm). The adducts of all
the complexes are found to react as observed from the decrease
in absorbance of the catecholate-to-iron(III) LMCT band. The
pseudo-first order rate constants (kobs: pip: 0.32–3.96 ¥ 10-4 s-1;
Et3N: 0.20–7.43 ¥ 10-4 s-1, Table 6) were calculated using equation
(1) by fitting the decrease in intensity of the LMCT band into the
following equation31,51 applicable for relatively slow reactions,
On adding one equivalent of H2DBC to 1, 3 and 4 in DMF
the Fe(III)/Fe(II) redox wave becomes intense suggesting the
superposition of the redox wave corresponding to DBSQ/H2DBC
(DBSQ = di-tert-butylsemiquinone) couple21–25,31 on it. And only
for 2 and 3 a less intense new redox wave at potentials (E1/2: 2,
-0.279; 3, -0.560 V, Fig. 7, 8) more positive than the Fe(III)/Fe(II)
redox wave appears. On adding two equivalents of Et3N the
Fe(III)/Fe(II) redox wave disappears,31 which is shifted to a more
negative potential (-0.850 to -1.308 V) for all the complexes
illustrating formation of higher amounts of the adducts and the
redox potential of DBSQ/DBC2- couple (E1/2: 1, -0.522; 2, -0.285;
3, -0.568; 4, -0.645 V, Fig. 7, 8) becomes considerably more
positive than that of free DBSQ/H2DBC couple (-1.434 V)50
reflecting the significant stabilization of DBC2- towards oxidation
upon coordination to iron(III). The DBSQ/H2DBC reduction
wave for 2 appears even in the absence of Et3N suggesting the
spontaneous deprotonation of H2DBC due to the strong affinity
of catecholate anions to iron(III) with a higher Lewis acidity. Also,
interestingly, the DBSQ/H2DBC redox couple for 2 appears at a
potential (E1/2, -0.285 V, Fig. 8, Table 5) more positive than those
for 1, 3 and 4 (E1/2: -0.522 to -0.645 V, Table 5) and the Fe(III)
→ Fe(II) redox wave disappears. This shows that DBC2- bound
to 2 is stabilized towards oxidation much more strongly than that
bound to 1, 3 and 4 and that the iron(III)-bound substrate in 2
is activated much more than that in 1, 3 and 4 but much less
A = A• + (A0 - A•) ¥ exp(-kobs ¥ t)
(1)
(2)
kobs = 1 + log(Abs)
where, t is time, A, A0 and A• are the absorbances at time t, 0 and •
respectively and kobs is the pseudo-first order rate constant and the
t
1/2 of the catecholate adducts were calculated31 using the equation
t1/2 = 0.693/kobs. The second order rate constants of the reactions
were then derived (Pip: 0.66–9.60 ¥ 10-2 M-1 s-1; Et3N: 0.41–15.28 ¥
10-2 M-1 s-1) using the equation ko ( = kobs/[O2]. Interestingly, a
2
new band around 400 nm appears and grows in intensity during
oxygenation reaction of the DBC2- adducts of 1, 3 and 4; however,
no such band is discerned for the adduct of 2 as it is masked (cf.
below) possibly by the absorption bands of ligand –NO2 group.
The oxygenation products were identified by GC-MS (EI) and
quantified by GC (FID) techniques. It may be noted that catechols
are also fragmented into unstable species, which are difficult to
detect and characterized using GC-MS. When the DBC2- adducts
of 1–4 generated in situ using Et3N as base were reacted with
dioxygen in DMF solution over 48 h (t1/2: pip: 0.49–6.02 h;
Et3N: 0.32–9.63 h) the simple two-electron oxidation product di-
tert-butylbenzoquinone (DBQ) 11 is obtained in higher amounts
for all the complexes (Et3N: 6.4–26.1%; pip: 3.0–16.0%) and the
than that in the previously reported [Fe(N4)Cl2]+ complexes (E1/2
,
-0.065 to -0.079 V).34 The DBC2- anion coordinated strongly to
2 has a tendency to lose electrons to form benzoquinone more
readily than that coordinated weakly to 1, 3 and 4.
9620 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 6 Kinetic data for the oxidative cleavage of H2DBC catalyzed by iron(III) complexes with O2 in acetonitrile–DMF and the cleavage products
analyzed after 48 h (see Scheme 3)
Complexa
Solvent
CH3CN
CH3CN
Base
Et3N
Pip
Cleavage product (%)b
Product (%)b
kobs (10-4 s-1)
ko (10-2 M-1 S-1)c
t1/2 (h)d
0.29
2
[Fe(L1)(DBC)]-
11 (27.2)
12 (1.5)
6.7 0.5
2.3 0.3e
8.2 0.6
9.9 0.4
14 (1.4)
11 (5.8)
8.0 0.3
2.0 0.1e
0.24
12 (3.7)
14 (2.1)
DMF
DMF
Et3N
Pip
11 (25.0)
12,13 (1.5)
11 (9.0)
12,13 (1.5)
14 (1.4)
11 (6.4)
7.4 0.1
15.3 0.3e
4.7 0.1
9.6 0.1e
15.3 0.2
9.6 0.1
0.25
0.41
[Fe (L2)(DBC)]-
[Fe (L3)(DBC)]-
DMF
DMF
Et3N
Pip
0.2 0.1
0.4 0.1
0.4 0.1
0.9 0.1
9.63
5.06
12,13 (1.0)
11 (16.0)
12,13 (1.0)
14 (<1.0)
11 (25.8)
12 (1.5)
CH3CN
CH3CN
DMF
Et3N
Pip
2.7 0.1
0.4 0.1e
3.4 0.1
6.3 0.2
12.5 3.0
8.2 0.7
0.70
0.38
0.32
0.49
14 (1.2)
11 (6.2)
5.1 0.2
0.3 0.1e
12 (3.2)
14 (1.7)
Et3N
Pip
11 (26.1)
12,13 (<1.0)
14 (<1.0)
11 (3.0)
6.1 1.4
0.5 0.3e
DMF
4.0 0.3
0.5 0.1e
12,13 (10.8)
14 (1.0)
[Fe (L4)(DBC)]-
CH3CN
CH3CN
Et3N
Pip
11 (25.7)
12 (1.3)
11 (19.6)
12 (2.3)
14 (1.0)
0.2 0.1
0.1 0.0e
1.2 0.1
0.3 0.1
1.5 0.0
8.75
1.62
DMF
DMF
Et3N
Pip
11 (23.46)
12,13 (1.34)
14 (<1.0)
11 (16.0)
12,13 (10.8)
14 (1.1)
1.0 0.3
0.2 0.0e
2.0 0.6
0.7 0.1
2.06
6.02
0.3 0.0
0.2 0.0e
d
a Generated by adding two equivalents of Et3N and Piperidine, b Based on H2DBC, c kO = kobs/ [O2], t1/2 = 0.693/kobs
,
e Rate constant 3,5-di-tert-
2
butylbenzoquinone formation (k¢obs), Pip = Piperidine.
Scheme 3 Oxygenation products of H2DBC for iron(III) complexes.
intradiol cleavage products 3,5-di-tert-butyl-5-(carboxymethyl)-2-
furanone 12, C15H24O4, m/z, 254, and 3,5-di-tert-butyl-5-(N,N-
dimethylamidomethyl)-2-furanone 13 (Et3N: <1–1.5%; pip: 9.0–
10.8%, Scheme 3, Table 6) and the extradiol cleavage products
4,6-di-tert-butyl-2-pyrone 14 were also obtained but in very small
quantities (Et3N: 0.0 – <1.0%; pip: <1.0–1.4%, Scheme 3). So,
the band appearing around 400 nm and growing in intensity is
ascribed52 to the formation of 11 and the kobs¢ values calculated for
1 and 3 from the increase in intensity of the band using equation
(2) represent the rate of formation of DBQ (kobs¢: Pip: 0.02–0.48 ¥
10-4 s-1; Et3N; 0.02–1.77 ¥ 10-4 s-1, Table 6) and the changes
observed for 4 are not sufficiently good to calculate the rates.
The intradiol cleavage products 12 and 13 are derived53 from the
nucleophilic attack of methoxide ion on cis,cis-muconic anhydride,
which is the immediate product of oxidative cleavage.
The DBC2- adducts of 1, 3 and 4 were generated in situ in
acetonitrile solution also and their reactivity towards O2 was
investigated as described above by monitoring the decay of the
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9621
©

View Article Online
DBC2--to-iron(III) LMCT band (1, 548; 3, 540; 4, 538 nm, Fig. 9).
As complex 2 is insoluble in acetonitrile its reactivity was not
investigated. The rate constant for oxygenation reaction of the
have found55 that when FeCl3/BPY/Py was used for dioxygenation
of catechol in THF solution the oxidized product benzoquinone
was mainly obtained. Very recently, we have observed the for-
mation of benzoquinone product in major amounts for the
iron(III) complexes of tripodal tris(pyrazolyl)methane ligands.56
The formation of benzoquinone product for the present complexes
is illustrated by invoking molecular oxygen attack on the vacant
iron(III) site in the five-coordinate adduct (a, Scheme 4; cf. above)
in which DBC2- is coordinated in bidentate fashion, as evidenced
by spectral and electrochemical results (cf. above). As the bound
O2 molecule is located in the same plane as that of the catecholate
substrate, it cannot attack the latter; however, it accepts two
electrons from the bound substrate through iron(III) to for the
quinone.56 It is to be noted that one of the phenolate arms
in the DBC2- adduct remains uncoordinated because of steric
constraint from the diazapane back bone, as revealed by molecular
model building studies (cf. above). The re-coordination of the
bulky t-butylphenolate arm displaced upon substrate binding is
rendered difficult as it sterically clashes with the t-butyl group
on the catecholate substrate, as revealed by building molecular
models, leading to facilitate the attack of O2 on the vacant
iron(III) site and hence the amount of quinone product obtained
for 4 in DMF–acetonitrile solution using piperidine as base is
higher than those for 1 and 3. However, when Et3N is used as
base all the adducts, except that of 2, yield a higher amount of
the quinone product in the same range suggesting that strong
bidentate coordination effected by using piperidine encourages
cleavage products. In contrast to 1, 3 and 4, a major amount of
quinone product is observed for 2 in DMF/piperidine with smaller
amounts of catechol cleavage products, which is consistent with
the observation of its DBSQ/DBC2- redox couple at a potential
more positive than those for the other complexes (cf. above). As
the displaced nitrophenolate arm in the catecholate adduct of
2 remains mostly uncoordinated to iron(III) the probability of
molecular oxygen attacking the vacant iron(III) site is very high
leading to the formation of quinone in major amounts.
complexes 1, 3 and 4 (ko : pip: 1.47–9.87 ¥ 10-2 M-1 s-1; Et3N: 0.27–
2
8.23 ¥ 10-2 M-1 s-1; Table 6) and the t1/2 values of the catecholate
adducts were calculated31 as described above. The in situ generated
adducts of 1, 3 and 4 react with dioxygen in acetonitrile solution
over 48 h (t1/2: pip: 0.24–1.62 h; Et3N: 0.29–8.75 h) to afford DBQ
11 as the simple two-electron oxidation product (Et3N: 25.7–
27.2%, pip: 5.8–19.6%, Scheme 3) and small amounts of both
intradiol (Et3N: 1.3–1.5%; pip: 2.3–3.7%) and extradiol cleavage
products (Et3N: 0–1.4%; pip: 1.0–2.1%) also.
When dioxygen attacks the activated carbon atom of cate-
cholate substrate strongly bound to the iron(III) center in the
adduct (substrate-activation mechanism), the Criegee intermedi-
ate [(L)(DBSQ)Fe(III)O2]- gives intradiol cleavage products upon
acyl migration and extradiol cleavage products upon alkenyl
migration44,57,58 (b, Scheme 4). On the other hand, when dioxygen
attacks the iron(III) site of the DBC2- adduct, vacated by the
phenolate donor cis to both the catecholate oxygen atoms (oxygen-
activation mechanism18–26,29,34) the same substrate-alkylperoxo-
Fe3+ intermediate is formed leading to the yield of extradiol
cleavage products (c, Scheme 4). The displacement of the second
coordinated phenolate oxygen atom and hence the formation of
the intermediate [(L)(DBSQ)Fe(III)O2]- responsible for catechol
cleavage does not appear to be facile and hence the observed
formation of smaller amounts of the cleavage products. Also,
it may be noted that the catecholate substrate, as it is weakly
bound to the present bis(phenolate)iron(III) complexes with de-
creased Lewis acidity and hence insufficiently activated, is not
prone to attack by free or iron(III)-bound dioxygen leading to
smaller amounts of intradiol or extradiol cleavage products. The
amounts of cleavage products obtained are higher and that of
quinone product lower in DMF/piperidine than those respectively
in acetonitrile/piperidine. The catecholate substrate is strongly
Fig. 9 (a) Electronic absorption spectral changes observed during the
reaction of the adduct [Fe(L3)(DBC)]- generated in situ using Et3N with O2
at 25 ◦C in CH3CN (2 ¥ 10-4 M). (b) Electronic absorption spectral changes
observed during the reaction of the adduct [Fe(L3)(DBC)]- generated
in situ using piperidine with O2 at 25 ◦C in CH3CN (2 ¥ 10-4 M).
The benzoquinone product for 1–4 in DMF/Et3N is almost
exclusive (6.4–26.1%) with only smaller amounts of intradiol
(<1–1.5%) and extradiol cleavage products (0.0–1.0%) and a
higher amount of benzoquinone (25.7–27.2%) is observed for
1, 3 and 4 in acetonitrile/Et3N solution. Similar formation of
quinone as well as intradiol cleavage products has been observed
for [Fe(SSCTH)(Cat)], where SS-CTH = (SS)-5,5,7,12,12,14-
hexamethyl-l,4-,8,11-tetraazacyclo-tetradecane, in which the cat-
echolate anion is coordinated in a monodentate fashion but no
explanation was tendered for the observation.54 Funabiki et al.
9622 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 4 Proposed reaction mechanism for H2DBC oxidation and intradiol-cleavage catalyzed by 1–3.
bound and hence activated in the coordinating solvent DMF more
than in acetonitrile leading to the formation of higher amounts
of intradiol cleavage products in DMF. However, when Et3N is
used as base the amount of quinone product obtained in DMF
falls in the same range as that in acetonitrile. Also, in both
DMF and acetonitrile solvents the amounts of cleavage products
obtained are higher and that of quinone product lower when
piperidine is used as base than when Et3N is used as base. This
is because a higher amount of six-coordinate complex is formed
in the presence of the strong base piperidine. Also, interestingly,
the ratio of the amount of intradiol cleavage product to that of
quinone product observed for 1 and 3 is higher than that for 4 in
DMF/piperidine. Obviously, the catecholate substrate is bound
to the former complexes more strongly than to the latter and
so molecular oxygen binds to adducts of the former complexes
more strongly leading to a higher amount of intradiol cleavage
product (cf. below). This is consistent with the observation of the
DBSQ/DBC2- redox couple for 4 at a potential more negative
than for 1 and 3.
As a major amount of quinone is mainly obtained for 1–4 (6.4–
26.1%) in both DMF and acetonitrile solutions, it is evident that
the observed rates of oxygenation correspond to mostly formation
of quinone. The second order reaction rate constant (ko2 , 1.9–
15.3 ¥ 10-2 M-1 s-1, Table 6) for the adducts [Fe(L)(DBC)]- derived
from 1, 3 and 4 in DMF/Et3N/piperidine are higher than that (ko2 ,
0.41–0.78 ¥ 10-2 M-1 s-1) for the adduct of 2. As DBC2- is bound to
iron(III) in [Fe(L2)(DBC)]- more strongly than in the adducts of
the former complexes, as revealed by the more positive E1/2 value
of DBSQ/DBC2- couple (cf. above), the electron-transfer from
bound catecholate to dioxygen is rendered difficult and hence
the lower rates of quinone formation for 2. This is consistent
with the relatively low yields of the quinone product for 2 when
Et3N is used as base (cf. above). Interestingly, the rate of quinone
formation decreases in the order, 1 > 3 > 4 in both acetonitrile
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9623
©

View Article Online
and DMF when Et3N or piperidine is used as base. The increase
in molecular crowding in the complexes along this series would
discourage the attack of molecular oxygen on the vacant iron(III)
site. Also, as the Lewis acidity of the iron(III) center decreases along
this series (cf. electrochemical studies), the extent of interaction of
the catecholate substrate and hence the rate of oxygenation of
these complexes would decrease in the above order. Interestingly,
a higher rate of quinone formation is observed for complexes with
a higher LMCT band energy suggesting stronger interaction of
catecholate with iron(III) center confers higher reactivity. This is
similar to that observed for the catechol cleavage reactions.31,34
Further, the DBC2- adducts of the present iron(III)-bis(phenolate)
complexes undergo dioxygenation much more slowly than those
for the analogous iron(III) complexes of pyridyl-based iron(III)-
N4 ligands,34 which is due to the lower Lewis acidity and hence
the weaker interaction of 1–4 with the catecholate substrate.
However, they undergo oxygenation with rates faster than the
previously reported iron(III)-bis(phenolate) complexes,21 which is
consistent with the higher Fe–O–C bond angle (135.5◦) observed
for one of them (3). Also, it may be noted that the trigonal
bipyramidal and octahedral iron(III)-phenolate complexes having
the Fe–O–C bond angle of ~128.5◦ have been reported to show
slow or no dioxygenase activity.21,22 Further, interestingly, the rate
of dioxygenation for 1, 3 and 4 in DMF is higher than that in
acetonitrile. The enhanced bidentate coordination of catechol in
DMF leads to faster electron transfer from catecholate adduct
to iron(III)-bound O2. However, this trend is not observed when
piperidine is used as base because cleavage reactions are also
facilitated in the presence of the strong base used. Also, for all
the complexes the rate of dioxygenation is dependent on the base
used; thus they are higher in the presence of Et3N than in piperidine
because of the occurrence of cleavage reactions, which contribute
to the observed rates of reactions. In fact, when the weak base Et3N
is used, interestingly, the oxidized product is obtained in very large
quantities with the cleavage products in smaller quantities.
Thus, the subtituents on the present bis(phenolate) ligands
determine the Lewis acidity of the iron(III) center in the complexes,
and also the ability of the ligands to rearrange to accommodate the
chelating catecholate substrate, and dictate the extent of interac-
tion of the complexes with catechol and hence the type of reaction
of catechol—quinone formation vs. catechol cleavage—and also
the rate of formation and amount of cleavage/quinone product
formed. Thus the iron(III) complex of a weakly coordinating
primary bis(p-NO2-phenolate) ligand tends to yield the oxidized
product in major amounts while that of the sterically crowded
bis(di-t-butylphenolate) ligand yields the quinone as well as an
enhanced amount of intradiol cleavage product in the presence of
a strong base.
with methyl- and 3,5-di-tert-butyl substituents on the phenolate
rings also exhibit a similar coordination geometry. In contrast, the
analogous bis(p-nitrophenolate) complex possesses an octahedral
coordination geometry. In DMF/Et3N all the complexes elicit
the two-electron oxidation of the model substrate 3,5-di-tert-
butylcatechol (H2DBC) to give a major amount of benzoquinone
and smaller amounts of both intra- and extradiol cleavage
products. However, interestingly, when piperidine is used as base
a decrease in the amount of oxidation product with an increase in
catechol cleavage product is achieved. Thus the extent of interac-
tion of the bis(phenolate) complexes with catechol and the catechol
cleavage activities and/or the two-electron oxidation of the cate-
chol as well are dictated by the Lewis acidity of the iron(III) center
of the bis(phenolate) complexes, as fine-tuned by the nature of sub-
stituents on the phenolate rings, the solvent and base as well. This
study reveals that catechol cleavage can be achieved and catechol
oxidation prevented by tuning the ligand architecture by incorpo-
rating strongly coordinating ligand donors and by using a coor-
dinating solvent and a stronger base to encourage the bidentate
coordination of catechol. The displacement of one of the two co-
ordinated phenolate donors from the iron(III) coordination sphere
upon catecholate binding is followed by molecular oxygen attack at
the same site leading to the formation of quinone product. Such a
displacement of the phenolate donor is interestingly similar to that
observed for the 1,2-CTD dioxygenase upon binding to the cate-
chol substrate; however, only the quinone is formed as the site of at-
tack of dioxygen is located in the plane of the catecholate substrate.
Acknowledgements
We sincerely thank the Department of Science and Technology,
New Delhi for supporting this research [Scheme No. SR/S5/BC-
05/2006], and the Council of Scientific and Industrial Research,
New Delhi for a Senior Research Fellowship to R. M. Professor
M. Palaniandavar is a recipient of a Ramanna Fellowship of
Department of Science and Technology, New Delhi. We thank
Professor P. Sambasiva Rao, Pondicherry University for providing
the EPR facility.
Notes and references
1 A. L. Feig and S. J. Lippard, Chem. Rev., 1994, 94, 759–805.
2 E. L. Hegg and L. Que, Jr., Eur. J. Biochem., 1997, 250, 625–629.
3 J. D. Lipscomb and A. M. Orville, in Degradation of Environmental
Pollutants by Microorganisms and Their Metalloenzymes, H. Sigel, A.
Sigel, ed. Marcel Dekker: New York, 1992, Vol. 28, pp 243–298.
4 (a) L. Que, Jr. and R. Y. N. Ho, Chem. Rev., 1996, 96, 2607–2624;
(b) M. Abu-Omar, A. Loaiza and N. Hontzeas, Chem. Rev., 2005, 105,
2227–2252.
5 E. I. Solomon, T. Brunold, M. I. Davis, J. N. Kemsley, S. K. Lee, N.
Lehnert, F. Neese, A. J. Skulan, Y. S. Yang and J. Zhou, Chem. Rev.,
2000, 100, 235–349.
6 S. Dagley, In The Bacteria, J. R. Sokatch and L. N. Ornston, ed.;
Academic Press: Orlando, 1986 Vol. 10, Chapter 10, pp 527–555.
7 M. A. Levin and M. A. Gealt, in Biotreatment of Industrial and
Hazardous Waste, M. A. Levin and M. A. Gealt, ed.; McGraw-Hill,
New York, 1993, pp 5–7.
8 L. Que, Jr., J. D. Lipscomb, E. Mu¨nck and J. M. Wood, Biochim.
Biophys. Acta, 1977, 485, 60–74.
9 (a) L. Que, Jr., In Iron Carriers and Iron Proteins, T. M. Loehr, ed. VCH:
New York, 1989, pp 467–524; (b) L. Shu, Y.-M. Chiou, A. M. Miller, J.
D. Lipscomb and L. Que, Jr., Biochemistry, 1995, 34, 6649–6659.
10 D. H. Ohlendorf, J. D. Lipscomb and P. C. Weber, Nature, 1988, 336,
403–405.
Conclusions and relevance to catechol dioxygenases
A few novel iron(III) complexes of bis(phenolate) ligands with
diazepane back bone have been isolated and studied as structural
and functional models for intradiol-cleaving catechol dioxyge-
nases. One of these complexes with two 3,5-dimethylphenolate
donors possesses a novel square pyramidal coordination geometry
◦
˚
with structural parameters (Fe–Ophenolate, 1.855 A; Fe–O–C, 135.5 )
close to those for the intradiol-cleaving 3,4-PCD enzymes (Fe–
◦
˚
Otyr, 1.81, 1.91 A; Fe–O–C, 133, 148 ) and the iron(III) complexes
9624 | Dalton Trans., 2010, 39, 9611–9625
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
11 D. H. Ohlendorf, A. M. Orville and J. D. Lipscomb, J. Mol. Biol., 1994,
244, 586–608.
12 M. P. Valley, C. K. Brown, D. L. Burk, M. W. Vetting, D. H. Ohlendorf
and J. D. Lipscomb, Biochemistry, 2005, 44, 11024–11039.
13 R. W. Frazee, A. M. Orville, K. B. Dolbeare, H. Yu, D. H. Ohlendorf
and J. D. Lipscomb, Biochemistry, 1998, 37, 2131–2144.
14 (a) M. W. Vetting, D. A. D’ Argenio, L. N. Ornston and D. H.
Ohlendorf, Biochemistry, 2000, 39, 7943–7955; (b) T. E. Elgren, A.
M. Orville, K. Kelly, J. D. Lipscomb, D. H. Ohlendorf and L. Que, Jr.,
Biochemistry, 1997, 36, 11504–1513.
15 M. W. Vetting and D. H. Ohlendorf, Structure, 2000, 8, 429–440.
16 (a) J. W. Whittaker and J. D. Lipscomb, J. Biol. Chem., 1984, 259, 4487–
4495; (b) A. M. Orville and J. D. Lipscomb, J. Biol. Chem., 1989, 264,
8791–8801; (c) A. E. True, A. M. Orville, L. L. Pearce, J. D. Lipscomb
and L. Que, Jr., Biochemistry, 1990, 29, 10847–10854.
17 (a) A. M. Orville, J. D. Lipscomb and D. H. Ohlendorf, Biochemistry,
1997, 36, 10052–10066; (b) A. M. Orville, N. Elango, J. D. Lipscomb
and D. H. Ohlendorf, Biochemistry, 1997, 36, 10039–10051.
18 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Jr., Chem. Rev., 2004,
104, 939–986.
19 M. I. Davies, A. M. Orville, F. Neese, J. M. Zaleski, J. D. Lipscomb and
E. I. Solomon, J. Am. Chem. Soc., 2002, 124, 602–614.
20 R. Yamahara, S. Ogo, H. Masuda and Y. Watanabe, J. Inorg. Biochem.,
2002, 88, 284–294.
21 M. Velusamy, M. Palaniandavar, R. Srinivasa Gopalan and G. U.
Kulkarni, Inorg. Chem., 2003, 42, 8283–8293.
Delroisse, F. Banse and J.-J. Girerd, Eur. J. Inorg. Chem., 2001, 2249–
2254.
31 (a) R. Mayilmurugan, E. Suresh and M. Palaniandavar, Inorg. Chem.,
2007, 46, 6038–6049; (b) R. Mayilmurugan, K. Visvaganesan, E. Suresh
and M. Palaniandavar, Inorg. Chem., 2009, 48, 8771–8783.
32 (a) R. B. Lauffer, R. H. II. Heistand and L. Que, Jr., J. Am. Chem.
Soc., 1981, 103, 3947–3949; (b) R. H. II. Heistand, A. L. Roe and L.
Que, Jr., Inorg. Chem., 1982, 21, 676–681; (c) R. H. II. Heistand, R.
B. Lauffer, E. Fikrig and L. Que, Jr., J. Am. Chem. Soc., 1982, 104,
2789–2796; (d) R. B. Lauffer, R. H. II. Heistand and L. Que, Jr., Inorg.
Chem., 1983, 22, 50–55.
33 (a) H. Fujii and Y. Funabashi, Angew. Chem., Int. Ed., 2002, 41, 3638–
3641; (b) T. Kurahashi, K. Oda, M. Sugimoto, T. Ogura and H. Fujii,
Inorg. Chem., 2006, 45, 7709–7721.
34 R. Mayilmurugan, H. S. Evans and M. Palaniandavar, Inorg. Chem.,
2008, 47, 6645–6658.
35 R. Mayilmurugan, E. Suresh, H. S. Evans and M. Palaniandavar,
Dalton Trans., 2009, (26), 5101–5114.
36 (a) Japan Chemical Society, Kagaku-Binran Basic Part II, 2nd edition,
Maruzen, Tokyo, 1975; (b) D. T. Sawyer, Oxygen Chemistry, Oxford
University Press, New York, 1991.
37 P. Mialane, L. Tehertanov, F. Banse, J. Sainton and J. Girerd, Inorg.
Chem., 2000, 39, 2440–2444.
38 SMART & SAINT Software Reference manuals, version 5.0, Bruker
AXS Inc.: Madison, WI, 1998.
39 G. M. Sheldrick, SADABS software for empirical absorption correc-
22 M. Velusamy, R. Mayilmurugan and M. Palaniandavar, Inorg. Chem.,
2004, 43, 6284–6293.
23 M. Velusamy, R. Mayilmurugan and M. Palaniandavar, J. Inorg.
Biochem., 2005, 99, 1032–1042.
24 H. G. Jang, D. D. Cox and L. Que, Jr., J. Am. Chem. Soc., 1991, 113,
9200–9204.
tionUniversity of Go¨ttingen: Germany, 2000.
40 SHELXTL Reference Manual, version 5.1, Bruker AXS Inc., Madison,
WI, 1998.
41 E. Y. Tshuva, I. Goldberg, M. Kol and Z. Goldschmidt,
Organometallics, 2001, 20, 3017–3028.
42 A. W. Addison, T. N. Rao, J. Reedijk, J. Van Rijn and G. C. Vershcoor,
J. Chem. Soc., Dalton Trans., 1984, 1349–1356.
43 M. R. McDevitt, A. W. Addision, E. Sinn and L. K. Thompson, Inorg.
Chem., 1990, 29, 3425–3433.
44 K. Visvaganesan, R. Mayilmurugan, E. Suresh and M. Palaniandavar,
Inorg. Chem., 2007, 46, 10294–10306.
45 (a) L. Casella, M. Gullotti, A. Pintar, L. Messouri, A. Rockenbauer and
M. Gyor, Inorg. Chem., 1987, 26, 1031–1038; (b) S. Wang, L. Wang, X.
Wang and Q. Luo, Inorg. Chim. Acta, 1997, 254, 71–77; (c) B. Krebs,
K. Schepers, B. Bremer, G. Henkel, E. Althus, W. Muller-Warmuth,
K. Griesar and W. Haase, Inorg. Chem., 1996, 35, 2360–2368; (d) B.
P. Gaber, V. Miskowski and T. G. Spiro, J. Am. Chem. Soc., 1974, 96,
6868–6873; (e) M. S. Shongwe, C. H. Kaschula, M. S. Adsetts, E. W.
Ainscough, A. M. Brodie and M. J. Morris, Inorg. Chem., 2005, 44,
3070–3079; (f) C. Imbert, H. P. Hratchian, M. Lanznaster, M. J. Heeg,
L. M. Hryhorczuk, B. R. McGarvey, H. B. Schlegel and C. Verani,
Inorg. Chem., 2005, 44, 7414–7422.
46 D. D. Cox, S. J. Benkovic, L. M. Bloom, F. C. Bradley, M. J. Nelson, L.
Que, Jr. and D. E. Wallick, J. Am. Chem. Soc., 1988, 110, 2026–2032.
47 A. J. Simaan, M.-L. Biollot, E. Riviere, A. Boussac and J.-J. Girerd,
Angew. Chem., Int. Ed., 2000, 39, 196–198.
48 (a) M. R. Bukowshi, P. Comba, C. Limberg, M. Merz, L. Que, Jr. and
T. Wistuba, Angew. Chem., Int. Ed., 2004, 43, 1283–1287; (b) S. Yoon,
H.-J. Lee, K.-B. Lee and H. G. Jang, Bull. Korean. Chem. Soc., 2000,
21, 923–928.
49 A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamental
and Applications, John Wiley & Sons, New York, 1980pp 218.
50 E. J. Nanni, M. D. Stalling and D. T. Sawyer, J. Am. Chem. Soc., 1980,
102, 4481–4485.
51 Y. Hitomi, M. Yoshida, M. Higuchi, H. Minami, T. Tanaka and T.
Funabiki, J. Inorg. Biochem., 2005, 99, 755–763.
52 Y. Hitomi, A. Ando, H. Matsui, T. Ito, T. Tanaka, S. Ogo and T.
Funabiki, Inorg. Chem., 2005, 44, 3473–3478.
53 L. Que Jr., R. C. Kolanczyk and L. S. White, J. Am. Chem. Soc., 1987,
109, 5373.
54 A. Dei, D. Gatteschi and L. Pardi, Inorg. Chem., 1993, 32, 1389–1395.
55 T. Funabiki, A. Mizoguchi, T. Sugimoto, S. Tada, M. Tsugi, H.
Sakamoto and S. Yoshida, J. Am. Chem. Soc., 1986, 108, 2921.
56 T. Dhanalakshmi, E. Suresh and M. Palaniandavar, Dalton Trans.,
2009, 8317–8328.
25 (a) R. Viswanathan and M. Palaniandavar, J. Chem. Soc., Dalton
Trans., 1995, 1259–1266; (b) R. Viswanathan, M. Palaniandavar, T.
Balasubramanian and P. T. Muthiah, Inorg. Chem., 1998, 37, 2943–
2951; (c) T. Dhanalakshmi, M. Bhuvaneshwari and M. Palaniandavar,
J. Inorg. Biochem., 2006, 100, 1527–1534; (d) M. Palaniandavar, M.
Velusamy and R. Mayilmurugan, J. Chem. Sci., 2006, 118, 601–610;
(e) M. Palaniandavar and R. Mayilmurugan, C. R. Chimie., 2007,
10, 366–379; (f) K. Sundaravel, T. Dhanalakshmi, E. Suresh and M.
Palaniandavar, Dalton Trans., 2008, 7012–7025.
26 (a) M. Pascaly, M. Duda, A. Rompel, B. H. Sift, W. M. Klauckean
and B. Krebs, Inorg. Chim. Acta, 1999, 291, 289–299; (b) M. Pascaly,
C. Nazikkol, F. Scweppe, A. Wiedeman, K. Zurlinden Muller and B.
Krebs, Z. Anorg. Allg. Chem., 2000, 626, 50–55; (c) M. Duda, M.
Pascaly and B. Krebs, Chem. Commun., 1997, 835–836; (d) M. Merkel,
M. Pascaly, B. Krebs, J. Astner, S. P. Foxon and S. Schindler, Inorg.
Chem., 2005, 44, 7582–7589; (e) M. Merkel, D. Schindler, S. M. Baldeau
and B. Krebs, Eur. J. Inorg. Chem., 2004, (4), 783–790.
27 M. Merkel, F. K. Mu¨ller and B. Krebs, Inorg. Chim. Acta, 2002, 337,
308–316.
28 M. Pascaly, M. Duda, F. Schweppe, F. Zurlinden, K. Muller and B.
Krebs, J. Chem. Soc., Dalton Trans., 2001, 828–837.
29 W. O. Koch and H.-J. Kru¨ger, Angew. Chem., Int. Ed. Engl., 1996, 34,
2671–2674.
30 (a) L. S. White, P. V. Nilsson, L. H. Pignolet and L. Que, Jr., J. Am.
Chem. Soc., 1984, 106, 8312–8313; (b) J. W. Pyrz, A. L. Roe, L. J.
Stern and L. Que, Jr., J. Am. Chem. Soc., 1985, 107, 614–620; (c) T.
Funabiki, H. Sakamoto, S. Yoshida and K. Tarama, J. Chem. Soc.,
Chem. Commun., 1979, 754–755; (d) T. Funabiki, T. Yamazaki, A.
Fukui, T. Tanaka and S. Yoshida, Angew. Chem., Int. Ed., 1998, 37,
513–515; (e) S. Ogo, R. Yamahara, T. Funabiki, H. Masuda and Y.
Watanabe, Chem. Lett., 2001, 1062–1063; (f) T. Funabiki, A. Fukui, Y.
Hitomi, M. Higuchi, T. Yamamoto, T. Tanaka, F. Tani and Y. Naruta,
J. Inorg. Biochem., 2002, 91, 151–158; (g) T. Funabiki, D. Sugio, N.
Inui, M. Maeda and Y. Hitomi, Chem. Commun., 2002, 412–413; (h) Y.
Hitomi, M. Higuchi, T. Tanaka and T. Funabiki, Inorg. Chim. Acta,
2005, 358, 3465–3470; (i) M. Higuchi, Y. Hitomi, H. Minami, T. Tanaka
and T. Funabiki, Inorg. Chem., 2005, 44, 8810–8821; (j) J. H. Lim, H.-J.
Lee, K.-B. Lee and H. G. Jang, Bull. Korean. Chem. Soc., 1997, 18,
1166–1171; (k) P. C. A. Bruijnincx, M. Lutz, A. L. Spek, W. R. Hagen,
G. van Koten and R. J. M. Klein Gebbink, Inorg. Chem., 2007, 46,
8391–8402; (l) F. Li, M. Wang, P. Li, T. Zhang and L. Sun, Inorg.
Chem., 2007, 46, 9364–9371; (m) N. Raffard, R. Carina, A. J. Simaan,
J. Sainton, E. Riviere, L. Tchertanov, S. Bourcier, G. Bouchoux, M.
57 C. J. Winfield, Z. Al-Mahrizy, M. Gravestock and T. D. H. Bugg, J.
Chem. Soc., Perkin Trans. 1, 2000, 3277–3289.
58 S. Mendel, A. Arndt and T. D. H. Bugg, Biochemistry, 2004, 43, 13390–
13396.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 9611–9625 | 9625
©