Synthesis and characterization of the tetranuclear iron(iii) complex of a new asymmetric multidentate ligand. A structural model for purple acid phosphatases

Article abstract of DOI:10.1039/b709293h

The ligand, 2-((2-hydroxy-5-methyl-3-((pyridin-2-ylmethylamino)methyl) benzyl)(2-hydroxybenzyl)amino)acetic acid (H₃HPBA), which contains a donor atom set that mimics that of the active site of purple acid phosphatase is described. Reaction of H₃HPBA with iron(iii) or iron(ii) salts results in formation of the tetranuclear complex, [Fe₄(HPBA) ₂(OAc)₂(-O)(-OH)(OH₂)₂]ClO ₄·5H₂O. X-Ray structural analysis reveals the cation consists of four iron(iii) ions, two HPBA^3- ligands, two bridging acetate ligands, a bridging oxide ion and a bridging hydroxide ion. Each binucleating HPBA^3- ligand coordinates two structurally distinct hexacoordinate iron(iii) ions. The two metal ions coordinated to a HPBA ^3- ligand are linked to the two iron(iii) metal ions of a second, similar binuclear unit by intramolecular oxide and hydroxide bridging moieties to form a tetramer. The complex has been further characterised by elemental analysis, mass spectrometry, UV-vis and MCD spectroscopy, X-ray crystallography, magnetic susceptibility measurements and variable-temperature Moessbauer spectroscopy. This journal is The Royal Society of Chemistry.

Full text of DOI:10.1039/b709293h

View Article Online / Journal Homepage / Table of Contents for this issue
An international journal of inorganic chemistry
www.rsc.org/dalton
Number 44 | 28 November 2007 | Pages 5093–5224
ISSN 1477-9226
PAPER
PERSPECTIVE
Lawrence R. Gahan et al.
Synthesis and characterization of
the tetranuclear iron(III) complex of a
Emmanuelle Schulz et al.
Asymmetric hydroamination of
non-activated carbon–carbon
1477-9226(2007)44;1-0
new asymmetric multidentate ligand multiple bonds

View Article Online
PAPER
www.rsc.org/dalton | Dalton Transactions
Synthesis and characterization of the tetranuclear iron(III) complex of a new
asymmetric multidentate ligand. A structural model for purple acid
phosphatases†
Athanassios K. Boudalis,b Robyn E. Aston,a Sarah J. Smith,a Ruth E. Mirams,a Mark J. Riley,a
Gerhard Schenk,a Allan G. Blackman,c Lyall R. Hantonc and Lawrence R. Gahana
Received 19th June 2007, Accepted 21st August 2007
First published as an Advance Article on the web 6th September 2007
DOI: 10.1039/b709293h
The ligand, 2-((2-hydroxy-5-methyl-3-((pyridin-2-ylmethylamino)methyl)benzyl)(2-hydroxybenzyl)-
amino)acetic acid (H3HPBA), which contains a donor atom set that mimics that of the active site of
purple acid phosphatase is described. Reaction of H3HPBA with iron(III) or iron(II) salts results in
formation of the tetranuclear complex, [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]ClO4·5H2O. X-Ray
structural analysis reveals the cation consists of four iron(III) ions, two HPBA3− ligands, two bridging
acetate ligands, a bridging oxide ion and a bridging hydroxide ion. Each binucleating HPBA3− ligand
coordinates two structurally distinct hexacoordinate iron(III) ions. The two metal ions coordinated to a
HPBA3− ligand are linked to the two iron(III) metal ions of a second, similar binuclear unit by
intramolecular oxide and hydroxide bridging moieties to form a tetramer. The complex has been
further characterised by elemental analysis, mass spectrometry, UV-vis and MCD spectroscopy, X-ray
crystallography, magnetic susceptibility measurements and variable-temperature Mo¨ssbauer
spectroscopy.
The iron(III) is coordinated to the oxygen of a tyrosine, the nitrogen
atom of a histidine and the oxygen atoms of two aspartate residues,
one of which bridges the two metal sites, a l-hydroxo and a
Introduction
The synthesis of di- and tetra-iron species with oxo, hydroxo and
OHO3− bridges is of interest as such complexes can potentially
putative terminal hydroxo/aqua ligand. The divalent metal ion
serve as structural models of the active sites of iron-containing
is coordinated to the oxygen atom of the bridging aspartate, the
metallobiosites such as methane monooxygenase, ribonucleotide
reductase, purple acid phosphatases and ferritin, as well as the
oxygen evolving proteins.1–9
nitrogen atoms of two histidine residues, an asparagine oxygen,
the l-hydroxo and a putative aqua ligand. The enzyme thus
provides an asymmetric binuclear active site with a hard NO5
Our principal interest is in systems which model the active sites
Fe(III) site and a somewhat softer N2O4 divalent metal site (Fig. 1).
of metallohydrolases,10 of which the purple acid phosphatases
The number of terminal and bridging aqua/hydroxo ligands is
(PAPs) are prototypical.10–14 More generally, we are interested in
ligand systems with the ability to assemble Fe(III) clusters with
relevance to the build up of iron-oxo cores.15 Thus, we have
previously investigated the synthesis of ligand systems leading
to diiron(III), diiron(II)16 and tetrairon(III)15 complexes and the
potential relevance of these complexes to metallobiosites.
ambiguous; available crystal structures provide conflicting data20,21
but spectroscopic data for the pig enzyme indicate that in the
resting state the iron(III) is pentacoordinate with only a bridging
hydroxo/aqua and a terminal iron(II)-bound aqua ligand present
in the active site.22 The characteristic colour of PAPs is due to
PAPs are the only binuclear metallohydrolases where the neces-
sity for a heterovalent active site (Fe(III)–M(II), where M = Fe, Zn
or Mn) for catalysis has been established.10–14 PAPs catalyse the
hydrolysis of a broad range of phosphorylated substrates at acidic
to neutral pH. The enzyme isolated from mammalian organisms
(pig, bovine, mouse, rat and human) is a ∼35 kDa monomeric
protein with an Fe(III)–Fe(II) centre; the amino acid sequences of
animal PAPs are highly conserved with at least 85% identity.17–19
aSchool of Molecular and Microbial Sciences, The University of Queensland,
QLD, Australia 4072
bInstitute of Materials Science, NCSR “Demokritos”, 15310, Aghia
Paraskevi Attikis, Greece
cDepartment of Chemistry, Te Tari Hua-Ruanuku, University of Otago, PO
Box 56, Dunedin, New Zealand
† CCDC reference number 651214. For crystallographic data in CIF or
other electronic format see DOI: 10.1039/b709293h
Fig. 1 Schematic illustration of the FeIII–FeII uteroferrin active site.
Note that crystallographic data have only been obtained for the oxidized
FeIII–FeIII form.17
5132 | Dalton Trans., 2007, 5132–5139
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
a charge transfer transition (kmax = 510–560 nm; e = ∼3000–
4000 M−1 cm−1) in the active site from a conserved tyrosine ligand
to the ferric ion. Mammalian PAPs can easily (and reversibly)
be oxidized to the inactive di-ferric form due to the low redox
potential (∼340 mV) of the divalent iron.23,24
sequent reduction of the imine with sodium borohydride. Trans-
esterification at the borohydride step resulted in the isolation
of the pure methyl ester after column chromatography. The ligand
was characterised by NMR and MS. The ligand can be consid-
ered as an analogue of 2-bis[{(2-pyridyl-methyl)-aminomethyl}-6-
{(2-hydroxybenzyl)-(2-pyridyl-methyl)}-aminomethyl]-4-methyl-
phenol (H2BPBPMP; Fig. 2)25 which when complexed with two
metal ions in the presence of acetate, furnishes a (l-phenoxo)-
bis(l-carboxylato) core with a soft site (N3O3) and a harder site
(N2O4) mimicking the soft and hard sites of the pig PAP enzyme.25
In the case of H3HPBA the inclusion of a glycine moiety mimics
the asparagine residue.
Reaction of a 2 : 1 : 3 stoichiometric ratio of ferrous chloride,
H2HPBMA and sodium acetate in aqueous basic methanol
resulted in hydrolysis of the ester and, upon standing in the
presence of perchlorate anion, dark purple crystals were isolated.
These were subsequently identified as the tetrameric iron(III)
complex [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]ClO4·5H2O.
We report herein the synthesis and characterization of a new un-
symmetrical binucleating ligand methyl 2-((2-hydroxy-5-methyl-
3-((pyridin-2-ylmethylamino)methyl)benzyl)(2-hydroxybenzyl)-
amino)acetate (H2HPBMA), and the corresponding acid,
methyl
2-((2-hydroxy-5-methyl-3-((pyridin-2-ylmethylamino)-
methyl)benzyl)(2-hydroxybenzyl)amino)acetic acid (H3HPBA)
(Fig. 2). The formation of a tetrairon(III) complex and its
characterization is reported, as well as the relevance of its
dimeric units as a structural model of the oxidized form of
mammalian PAP.
Description of the structure
Crystals of the complex contain a discrete [Fe4(HPBA)2(OAc)2(l-
O)(l-OH)(OH2)2]+ cation and a disordered perchlorate anion. Al-
though the microanalytical data suggested that the bulk crystalline
sample crystallised with eight waters of crystallisation, the sample
employed for structural analysis, Mo¨ssbauer and magnetic studies
contained only five water molecules. An ORTEP diagram of the
complex cation is shown in Fig. 3; crystal data are listed in Table 1
with important bond distance and angles information listed in
Table 2.
Fig. 2 H3HPBA and H2BPBPMP.
Results and discussion
Synthesis of the ligand and complex
The methyl ester H2HPBMA was produced in a reaction se-
quence (Scheme 1) involving the stepwise addition of N-(2-
hydroxybenzyl)glycine ethyl ester and 2-aminomethylpyridine
to 2-hydroxy-3-chloromethyl-5-methylbenzaldehyde with sub-
Fig. 3 An ORTEP plot of the complex cation of [Fe4(HPBA)2(l-
O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O determined at 89 K. Thermal proba-
bility ellipsoids are drawn at the 50% probability level.
The cation consists of a (l-oxo)(l-hydroxo)-bridged dimer of
[Fe2(HPBA)(OAc)(OH2)]2+ dimers. Each HPBA3− ligand coordi-
nates two structurally distinct six-coordinate iron(III) sites bridged
1
1
by an acetate ligand, in the typical g :g :l2 arrangement, and a
l-phenoxo bridge. The NO5 coordination environments of Fe(1)
and Fe(3) are composed of oxygen donors from the monodentate
phenolate and l-phenolate, the carboxylate, the l-acetate and the
tertiary nitrogen donor; the sixth coordination position for Fe(1)
is the l-oxo ligand and for Fe(3) the l-hydroxo ligand. The N2O4
Scheme 1 Synthesis of H2HPBMA; (i) MnO2; (ii) conc. HCl; (iii) sali-
cylaldehyde–NaBH4; (iv) (2) + (3); (v) (4) + 2-aminomethylpyridine,
NaBH4, methanol.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5132–5139 | 5133
©

View Article Online
Table 1 Crystallographic data for [Fe4(HPBA)2(l-O)(l-OH)(OAc)2-
(OH2)2]ClO4·5H2O
donors; for Fe(2) the l-hydroxo ligand and for Fe(4) the l-oxo
ligand complete the octahedral coordination.
The assignment of the formal l-oxo and l-hydroxo bridges was
based on chemical and structural evidence. In the first instance, the
charge balance required either the presence of a terminal hydroxo
ligand on one of the Fe(III) sites or the presence of a l-hydroxo
Empirical formula
Formula weight
Temperature/K
Crystal system
Space group
C52H69ClFe4N6O25
1436.14
89(2)
Monoclinic
P21/c
22.667(5)
11.552(5)
24.952(5)
111.076(5)
6097
˚
ligand. The terminal Fe(III)–O distances (Fe(2)–O(7), 2.045(4) A;
˚
a/A
26
˚
Fe(4)–O(15), 2.056(4) A) appear typical of Fe–OH2 bond lengths,
˚
b/A
˚
rather than Fe–OH which may be as short as 1.804 A or as
˚
c/A
b/◦
Volume/A
27–29
˚
long as 1.93 A.
In addition, the H(105) atom was determined
3
˚
from the difference Fourier synthesis and appears localized on
Absorption coefficient/mm−1
Reflections collected/unique
Data/restraints/parameters
Final R indices [I > 2r(I)]
R indices (all data)
1.603
9
˚
O(8) (O(8)–H(105), 0.83 A) rather than being symmetrically
56356/10796 (Rint = 0.0500)
10796/0/845
R = 0.0585, Rw = 0.840
R = 0.1447, Rw = 0.1608
located.9,15,30 The l-oxo and l-hydroxo ligands are within H-bond
◦
˚
distance (O(16) · · · O(8); 2.50 A, 179.21 ); other intramolecular
H-bonds are apparent between O(3)–O(15), N(5)–O(1), N(2)–
O(9) and O(7)–O(11) (Fig. 4). Comparison of the relative bond
lengths also suggests that the interdimer bridges are different.
Thus, the l-oxo-diiron(III) distances are at the long end of those
◦
˚
Table 2 Selected bond lengths (A) and angles ( ) for [Fe4(HPBA)2(l-
O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O
˚
typically seen for this moiety (1.76–1.82 A) with Fe–Ooxo distances
˚
˚
(Fe(1)–O(16), 1.841(4) A; Fe(4)–O(16), 1.819(4) A, Fe(1)–O(16)–
Fe(1)–O(16)
Fe(1)–O(4)
Fe(1)–O(6)
Fe(3)–O(8)
Fe(3)–O(9)
Fe(3)–O(14)
Fe(4)–O(16)
Fe(4)–O(14)
Fe(4)–N(5)
Fe(2)–O(8)
Fe(2)–O(6)
Fe(2)–N(3)
1.841(4)
2.022(4)
2.117(4)
1.909(4)
2.008(4)
2.070(4)
1.819(4)
2.047(4)
2.177(4)
1.884(4)
2.008(4)
2.126(5)
Fe(1)–O(3)
Fe(1)–O(1)
Fe(1)–N(1)
Fe(3)–O(11)
Fe(3)–O(12)
Fe(3)–N(4)
Fe(4)–O(13)
Fe(4)–O(15)
Fe(4)–N(6)
Fe(2)–O(5)
Fe(2)–O(7)
Fe(2)–N(2)
1.957(4)
2.060(4)
2.209(5)
1.941(4)
2.017(4)
2.186(4)
1.994(4)
2.056(4)
2.202(4)
1.977(4)
2.045(4)
2.161(5)
Fe(4), 146.8(2)◦) shorter than the Fe–Ohydroxo distances (Fe(2)–O(8),
◦
˚
˚
1.884(4) A; Fe(3)–O(8), 1.909(4) A, Fe(2)–O(8)–Fe(3), 144.0(2) ).
For Fe(III)–O–Fe(III) complexes the angles range from 114–180◦,
whilst for Fe(III)–OH–Fe(III) complexes the angles vary between
103–123◦.2 For [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]+ the in-
terdimer Fe–O–Fe angles (Fe(2)–O(8)H–Fe(3), 144.0(2)◦; Fe(1)–
O(16)–Fe(4), 146.8(2)◦) are not significantly different. Whilst the
Fe–O–Fe angle is similar to those reported for the◦tetrairon(III)
complexes containing the (O–H–O)3− core (138.3(1) ),9,30 the Fe–
OH–Fe angle is very different. Spectroscopic (vide infra) and
crystallographic evidence suggest that at the temperatures at which
the structural data were collected the hydrogen is localized on O(8)
so that the l-oxo and l-hydroxo moieties can be considered as
separate structural entities rather than (O–H–O)3−,30 suggesting
that constraints imposed by the tetrameric structure are the cause
of the unusual Fe–OH–Fe angle.
O(16)–Fe(1)–O(3)
O(3)–Fe(1)–O(4)
O(3)–Fe(1)–O(1)
O(16)–Fe(1)–O(6)
O(4)–Fe(1)–O(6)
O(16)–Fe(1)–N(1)
O(4)–Fe(1)–N(1)
O(6)–Fe(1)–N(1)
O(8)–Fe(3)–O(9)
O(8)–Fe(3)–O(12)
O(9)–Fe(3)–O(12)
O(11)–Fe(3)–O(14)
O(12)–Fe(3)–O(14)
O(11)–Fe(3)–N(4)
O(12)–Fe(3)–N(4)
O(16)–Fe(4)–O(13)
O(13)–Fe(4)–O(14)
O(13)–Fe(4)–O(15)
O(16)–Fe(4)–N(5)
O(14)–Fe(4)–N(5)
O(16)–Fe(4)–N(6)
O(14)–Fe(4)–N(6)
N(5)–Fe(4)–N(6)
O(8)–Fe(2)–O(6)
O(8)–Fe(2)–O(7)
O(6)–Fe(2)–O(7)
O(5)–Fe(2)–N(3)
O(7)–Fe(2)–N(3)
O(5)–Fe(2)–N(2)
O(7)–Fe(2)–N(2)
Fe(2)–O(6)–Fe(1)
Fe(2)–O(8)–Fe(3)
94.67(16)
92.36(16)
91.71(15)
92.88(15)
84.82(15)
171.14(17)
88.71(16)
88.06(16)
91.22(16)
98.30(17)
170.39(16)
174.95(16)
87.19(15)
88.33(16)
90.55(16)
101.15(16)
91.53(15)
90.57(16)
94.31(17)
88.82(16)
171.06(16)
87.39(16)
76.80(16)
93.21(16)
87.01(19)
175.45(18)
89.58(17)
87.79(19)
167.93(17)
86.81(18)
121.00(17)
144.0(2)
O(16)–Fe(1)–O(4)
O(16)–Fe(1)–O(1)
O(4)–Fe(1)–O(1)
O(3)–Fe(1)–O(6)
O(1)–Fe(1)–O(6)
O(3)–Fe(1)–N(1)
O(1)–Fe(1)–N(1)
O(8)–Fe(3)–O(11)
O(11)–Fe(3)–O(9)
O(11)–Fe(3)–O(12)
O(8)–Fe(3)–O(14)
O(9)–Fe(3)–O(14)
O(8)–Fe(3)–N(4)
O(9)–Fe(3)–N(4)
O(14)–Fe(3)–N(4)
O(16)–Fe(4)–O(14)
O(16)–Fe(4)–O(15)
O(14)–Fe(4)–O(15)
O(13)–Fe(4)–N(5)
O(15)–Fe(4)–N(5)
O(13)–Fe(4)–N(6)
O(15)–Fe(4)–N(6)
O(8)–Fe(2)–O(5)
O(5)–Fe(2)–O(6)
O(5)–Fe(2)–O(7)
O(8)–Fe(2)–N(3)
O(6)–Fe(2)–N(3)
O(8)–Fe(2)–N(2)
O(6)–Fe(2)–N(2)
N(3)–Fe(2)–N(2)
Fe(4)–O(14)–Fe(3)
Fe(1)–O(16)–Fe(4)
100.15(16)
92.89(16)
165.97(15)
172.30(15)
89.41(15)
84.71(16)
78.30(16)
92.87(17)
92.25(16)
88.55(16)
90.42(16)
91.51(15)
171.08(17)
79.90(16)
89.01(15)
93.40(16)
90.08(17)
175.52(16)
164.49(16)
88.09(17)
87.72(16)
88.74(17)
97.95(17)
90.07(15)
94.40(18)
171.15(18)
91.42(16)
94.11(18)
88.64(16)
78.45(17)
120.52(18)
146.8(2)
Fig. 4 H-bonding in [Fe4(HPBA)2(l-O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O.
The l-oxo group has a trans effect such that the Fe–N bonds
˚
trans to the oxo ligand (Fe(1)–N(1) 2.209(5) A; Fe(4)–N(6)
˚
2.202(4) A) are longer than those trans to the l-hydroxo ligand
˚
˚
donor sets for Fe(2) and Fe(4) comprise an aqua ligand, the l-
phenolate, l-acetate, the secondary amine and pyridine nitrogen
(Fe(2)–N(3), 2.126(5) A; Fe(3)–N(4), 2.186(4) A) and longer than
other Fe–N bonds in the structure. The reduction in the trans
5134 | Dalton Trans., 2007, 5132–5139
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Fe–N bond lengths upon protonation of the l-oxo ligand has
temperatures (1.8 K) indicative of an S = 0 ground state. The
signal increased slightly with temperature before decreasing at
50 K indicating C term behaviour with an excited state >25 cm−1.
been observed previously.9 Finally, the Fe(1)–Fe(4) interdimer
˚
separation mediated by the l-oxo bridge is 3.507 A, shorter
˚
than the interdimer l-hydroxo separation, Fe(2)–Fe(3) (3.608 A).
Magnetic susceptibility
By comparison, the intradimer Fe–Fe distances (Fe(1)–Fe(2)
˚
˚
and Fe(3)–Fe(4)) are 3.591 A and 3.575 A, respectively. The
[Fe2(HPBA)(OAc)(OH2)]2+ dimers are thus linked asymmetrically
by the interdimer l-hydroxo O(8) and l-oxo moieties O(16)
resulting in the symmetry of all four Fe atoms being different.
The importance of the l4-OHO3− moiety has been noted
previously,15,31,32 in particular its relevance to the oxygen-evolving
Mn4 complex of photosystem II.9,33 In the previous examples9,15,30
the OHO3− is symmetrical, bridging the four metal ions, in contrast
to the present case where the H atom appears localized on O(8).
Comparison of the binuclear cation [Fe2(HPBA)(OAc)(OH2)]2+
with the active site of pig PAP (uteroferrin) suggests structural
similarities.17 The phenolate, l-phenolate, tertiary amine and car-
boxylate donors of HPBA3−, the l-acetate and l-oxo, O(16), model
the tyrosinate, l-OH, the histidine and the aspartate residues, and
the putative terminal hydroxide/aqua ligand, respectively, at the
Fe(III) site of uteroferrin.17 The divalent metal site of uteroferrin
is modeled by the secondary amine and pyridine nitrogen atoms,
the l-acetate and l-phenolate, the water molecule and interdimer
l-hydroxide ion.17
The vMT product at 300 K is 4.49 cm3 mol−1 K, significantly
below the value theoretically predicted for four isolated S = 5/2
spins (17.51 cm3 mol−1 K, g = 2.0), thus indicating dominant
antiferromagnetic interactions. This is also corroborated by the
constant drop of the vMT product upon cooling, down to a value
of 0.059 cm3 mol−1 K at 2 K and by the drop of the magnetic
susceptibility vM upon cooling (0.010 cm3 mol−1 at 16 K) (Fig. 5).
An increase of magnetic susceptibility below 16 K is attributed
to ferric paramagnetic impurities, which, however, do not seem to
exhibit a strict Curie behaviour, as evidenced by the lack of a clear
plateau in the low-temperature region of the vMT vs. T plot.
Mass spectrometry
High resolution ESI-MS indicates the presence of [Fe4-
(HPBA)2(l-O)(l-OH)(OAc)(CHOO)]+ (m/z calc 1197.1203;
found 1197.1018), [Fe4(HPBA)2(O)2(OAc)]+ (m/z calc 1151.1149;
found 1151.0961) and [Fe4(HPBA)2)(O)2(CHOO)]+ (m/z calc
1137.0992; found 1137.0858). The high resolution negative
ion ESI-MS suggests the presence of [Fe2(HPBA)(OAc)(O)2-
(ClO4)2]− (m/z calc 1348.9945; found 1348.9923) and
[Fe2(HPBA)(HCOO)(O)2(ClO4)2]− (m/z calc 1334.9789; found
1334.9774). After standing for 2 weeks in acetonitrile, the low
resolution mass spectrum displays a peak at m/z 642 consistent
with formation of the dimer, [Fe2(HPBA)(OAc)(OH)(OH2)2]+ in
solution, although we have been unable to isolate this complex.
Fig. 5 vM and vMT vs. T experimental data points and calculated curves
for [Fe4(HPBA)2(l-O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O according to the
model of eqn (1).
Examination of the crystal structure of [Fe4(HPBA)2(OAc)2(l-
O)(l-OH)(OH2)2]ClO4·5H2O indicates that due to the lack of sym-
metry, four exchange interactions would be rigorously required
for the interpretation of its magnetic properties. However, its
dimer-of-dimers structure allows for a simplification of the model,
considering a common intradimer interaction (J12 = J34 = J1)
Visible absorption and magnetic circular dichroism spectra
The UV-vis absorption spectrum of [Fe4(HPBA)2(OAc)2(l-O)(l-
OH)(OH2)2]ClO4·5H2O in acetonitrile exhibits bands at 481 and
313 nm (e = 6050 and 14650 M−1 cm−1, respectively) assigned as
charge transfer transitions between the ligand and the iron(III)
centres. The spectra for similar diiron(III) complexes display a
broad absorption band at 601 nm (e ∼ 7700 M−1 cm−1) and
568 nm (e ∼ 4760 M−1 cm−1),34 whilst Fe(III)Fe(II) analogues
display maxima at 540 nm (e ∼ 4840 M−1 cm−1), 516 nm
(e ∼ 4560 M−1 cm−1), and 554 nm (e ∼ 950 M−1 cm−1),34,35
assigned as phenolate to iron(III/II) charge transfer transitions.
The intermolecular l-oxide bond in [Fe4(HPBA)2(OAc)2(l-O)(l-
OH)(OH2)2]+ and similar diiron(III) complexes is believed to cause
a blue shift in 400–500 nm absorption bands arising from increased
p bonding between the iron(III) and l-oxo atoms with increase in
the Fe–O–Fe angle.36,37
(Fig. 6). As far as the interdimer interactions (J23 = J3 and J14
=
J2) are concerned, these could not be a priori considered equal, due
to the proton localization on one of the bridging oxides, and the
subsequent asymmetry of the interdimer bridges [Fe(1)–O(16)–
Fe(4) = 146.81(21)◦, Fe(2)–O(8)–Fe(3) = 144.00(22)◦]. Thus,
to test whether these are indeed equal, initial fitting attempts
considering J2 = J3 were carried out. These yielded poor results,
and it was thus decided to lift this constraint, and consider that
J2 = J3. The Hamiltonian used was therefore:
ˆ
ˆ ˆ
ˆ ˆ
ˆ ˆ
ˆ ˆ
H = −2[J1(S1S2 + S3S4) + J3S2S3 + J2S4S1]
(1)
Due to the relatively large contribution of the paramagnetic
impurities, the fits were sensitive to the temperature range of
the dataset to be fitted. The low-temperature data (below 10 K)
could not be satisfactorily fitted, probably due to the non-Curie
behaviour of the paramagnetic impurities, so these were omitted
The MCD spectrum of [Fe4(HPBA)2(OAc)2(l-O)(l-OH)-
(OH2)2]ClO4·5H2O displays only a weak signal at very low
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5132–5139 | 5135
©

View Article Online
Fig.
6 Spin-coupling scheme for [Fe4(HPBA)2(l-O)(l-OH)(OAc)2-
(OH2)2]ClO4·5H2O.
from the datasets. Satisfactory fits could be obtained above 10 K,
with noticeable improvements resulting from omission of the low
temperature (<10 K) data points. A satisfactory compromise was
achieved by considering the data between 11 and 300 K. Best-
Fig. 7 Error contour plot for various J2 vs. J3 combinations for
[Fe4(HHPBA)2(l-O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O. The two minima
are interchangeable due to the symmetry of the problem.
fit parameters were J1 = −28.8 cm−1, J2 = −39.9 cm−1, J3
=
Mo¨ssbauer spectroscopy
−13.2 cm−1, g = 2.0 (fixed), q = 0.44% with R = 8.6 × 10−5
(Fig. 5), leading to an S = 0 ground state, well separated from the
first excited S = 1 state, at 47.5 cm−1. These values are consistent
with exchange couplings between high-spin ferric (S = 5/2) ions
bridged by monatomic oxygen bridges.36
Mo¨ssbauer spectra of [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]-
ClO4·5H2O are shown in Fig. 8. These exhibit composite
quadrupole-split doublets with parameters typical of high-spin
ferric ions in octahedral environments (Table 3). An increase in
the isomer shifts upon cooling is attributed to a second-order
Doppler effect.41 The complex contains two different ferric sites
with NO5 (Fe(1)/Fe(3)) and N2O4 (Fe(2)/Fe(4)) donor-atom sets.
However, based on the crystal structure of [Fe4(HPBA)2(OAc)2(l-
O)(l-OH)(OH2)2]ClO4·5H2O (at 89 K) these coordination spheres
are not identical, since O(8) is a hydroxo- and O(16) an
oxo-bridge. Therefore, there are two types of NO5 and N2O4
Bridging between each Fe–Fe pair in [Fe4(HPBA)2(OAc)2(l-
O)(l-OH)(OH2)2]ClO4·5H2O is mediated by one monatomic
bridge (O2−, OR−, OH−), unsupported by auxiliary bridges. So
ideally, the derived couplings should be comparable to respective
couplings within singly-bridged diferric complexes with similar
unsupported monatomic bridges. However, although such exam-
ples are abundant for oxo-bridged diferric complexes, cases of
phenoxo- or hydroxo-bridged complexes are rare. Unsupported
oxo-bridged, high-spin diferric complexes display a spread of their
exchange couplings,2 which range from −42 to −195 cm−1. J2
is at the low limit of this range. It is probable that hydrogen
bonding with the proton of the adjacent hydroxide confers a partial
hydroxo character to the bridge, which thus mediates a weaker
coupling. J1 on the other hand is significantly stronger than the
exchange couplings encountered in FeIII–(l-OPh−)2–FeIII diiron
complexes (+1.2 to −8.9 cm−1).38 It should be noted however, that
in the case of [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]+ only one
phenoxo bridge is present, and that the Fe–O–Fe angle is much
larger (∼121◦) than that observed in the previous examples (97.1–
110◦). Finally, J3 is comparable to the values encountered for
dihydroxo-bridged diferric complexes (−5.5 to −11.7 cm−) and
within the range determined for singly hydroxo-bridged FeIII–FeIII
interactions (−10.7 to −21 cm−1).15,39,40
To verify the uniqueness of this solution an error-level plot
(Fig. 7) was drawn for various combinations of J2 and J3 values.
This verified our solution as unique, by showing two equivalent
minima. The shape of the contour plots around the minima was
indicative of the uncertainties of J2 and J3; while the weaker
coupling is fairly well determined, the stronger one is less so.
However, the quality of the fits and the agreement of the present
values with those previously determined, support the validity of
our solution.
Fig. 8 Mo¨ssbauer spectra of [Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]-
ClO4·5H2O at various temperatures. The spectra are fitted to the models
described in the text.
5136 | Dalton Trans., 2007, 5132–5139
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
Table 3 Mo¨ssbauer parameters for [Fe4(HPBA)2(OAc)2(l-O)(l-
OH)(OH2)2]ClO4·5H2O
although these moieties do act as bridging agents for the
formation of the tetramer. The synthetic strategy employed offers
the possibility for further elaboration introducing chiral centres
by replacing glycine, and exploring the synthesis of mixed metal
systems.
Site
T/K
293
180
78
d/mm s−1
DEQ/mm s−1
(C/2)/mm s−1
Ratio
A
B
A
B
A1
A2
B1
B2
0.38(1)
0.38(1)
0.47(1)
0.46(1)
0.51(1)
0.51(1)
0.51(1)
0.51(1)
1.10(1)
0.78(1)
1.24(1)
0.81(1)
1.43(1)
1.14(1)
0.91(1)
0.58(1)
0.16(1)
0.16(1)
0.17(1)
0.17(1)
0.16(1)
0.16(1)
0.16(1)
0.16(1)
50
50
50
50
25
25
25
25
Experimental
Chemicals
All reagents and solvents were obtained commercially and were
used without further purification.
hydroxo
hydroxo
coordination spheres (NO5oxo/NO5
and N2O4oxo/N2O4
,
respectively) and the data may in principle be analysed assuming
four quadrupole-split doublets in a 1 : 1 : 1 : 1 ratio. Due to the
similarities of the donor atom sets (NO5 and N2O4) with only
nitrogen and oxygen atoms, we consider these doublets as nested
(i.e. of similar isomer shifts and different quadrupole splittings).
In any case, however, site assignment will be ambiguous, due to
the low sensitivity of HS–Fe(III) Mo¨ssbauer parameters to the
coordination environment, and the similarities of the coordination
spheres of the four sites.
Initial examination of the spectra reveals that linewidths
decrease upon heating. Thus, the 78 K spectrum could only
be fitted to the above mentioned model of four doublets, while
higher-temperature spectra exhibited narrowing of the linewidths,
allowing a simpler model for their interpretation. The 180 and
293 K spectra could be simulated by a pair of quadrupole-split
doublets in a 1 : 1 ratio and reasonable linewidths. Although
reliable assignments are difficult, as discussed above, for purposes
of comparison, indicative fitting parameters are shown in Table 3.
This temperature-dependent behaviour suggests the influence
of dynamic effects on the Mo¨ssbauer spectra. In particular, we
could assume that upon heating, the proton trapped between
oxides O(8) and O(16) may start “hopping” between the two.
This effect, if thermally accelerated to a sufficient degree, could
surpass the Mo¨ssbauer timescale and lead to thermally averaged
spectra of the two different conformations. This has been observed
in another Fe4 cluster exhibiting a [OHO]3− bridge.15 In that
case, however, there was only one coordination sphere (N2O4),
which, when split to two (N2O4oxo/N2O4hydroxo ), yielded two well-
resolved doublets. In our case, the two coordination spheres
split to four, leading to more overlapped spectra that are harder
to analyse.
Instrumental methods
NMR spectra were measured using Bruker AV300 (300 MHz) and
AV500 (500 MHz) instruments. Spectra were recorded in CDCl3,
with chemical shifts reported in parts per million calibrated
using the resonances for CDCl3, dH(CHCl3) = 7.24 ppm and
dC(CDCl3) = 77.0 ppm. Low and high resolution positive-ion
mass spectra were obtained using a Finnigan MAT 900 XL
mass spectrometer and methanol solutions for the ligand, and
acetonitrile for the metal complex. All samples were subjected to
electrospray ionization, with voltages tuned to optimize the sig-
nals. Elemental analyses were performed using the microanalysis
facilities at The University of Queensland and The University
of Otago. UV-vis spectra were measured in acetonitrile with a
Hitachi U-3000 spectrometer (10 mm quartz cells) over the 200–
800 nm region and with a Cary17 spectrometer (1 mm quartz
cells) over the 200–2500 nm region at ambient temperature.
MCD spectra of a 0.74 mM solution of [Fe4(HPBA)2)(l-O)(l-
OH)(OAc)2(OH2)2]ClO4·5H2O in ethanol–methanol (50 : 50) were
measured at 7 T using a R669 PMT detector, calcite polarizer,
and 24 V/250 W halogen lamp over multiple temperatures.
Variable-temperature (78–293 K) Mo¨ssbauer spectra were col-
lected at the Laboratory of Mo¨ssbauer Spectroscopy of the
Institute of Materials Science of NCSR “Demokritos” with a
constant acceleration spectrometer using a 57Co(Rh) source and a
variable temperature Oxford cryostat. Home-written routines were
used to fit the Mo¨ssbauer parameters and determine their standard
deviations of statistical origin (given in parentheses). Isomer shift
values (d) are reported relative to iron foil at 300 K. Mag-
netic susceptibility measurements (2–300 K) were made using
a Quantum Design MPMS SQUID magnetometer with an
applied field of 0.1 T. The crystalline sample of [Fe4(HPBA)2)(l-
O)(l-OH)(OAc)2(OH2)2]ClO4·5H2O was enclosed in a calibrated
gelatine capsule positioned in the centre of a drinking straw fixed
to the end of the sample rod. Data were corrected with the
standard procedure for the contribution of the sample holder
and diamagnetism of the sample. The magnetic susceptibility
has been computed by exact calculation of the energy levels
associated with the spin Hamiltonian, through diagonalization
of the full matrix with a general-symmetry program.42,43 Least-
squares fittings were accomplished with an adapted version of the
Conclusion
The complex [Fe4(HPBA)2)(l-O)(l-OH)(OAc)2(OH2)2]ClO4·
5H2O has been prepared from a new binucleating ligand.
The intricate l-phenoxo, l-oxo, l-hydroxo and l-O–H–O
dimer of dimer structure has been characterised by X-ray
crystallography, Mo¨ssbauer spectroscopy and magnetochemical
studies involving consideration of the inter- and intra-dimer
interactions. Structurally, the asymmetric binuclear units have
similarities with the active site of uteroferrin although lacking the
subtlety of the metallobiosite’s ability to maintain a heterovalent
environment.17 The model does reproduce the putative terminal
hydroxide/aqua ligand of the trivalent iron site in uteroferrin,
function-minimization program MINUIT.44 The error factor R is
ꢀ
defined as R =
2 , where N is the number of experimental
(x −x
)
calc
exp
2
Nx
exp
points.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5132–5139 | 5137
©

View Article Online
Crystallographic measurements
dichloromethane (3 × 35 cm3), the combined organic layers dried
over anhydrous sodium sulfate, filtered and the solvent removed to
give the product as an orange oil (3.24 g, 92% yield). H NMR (dH,
CDCl3): 1.27 (t, 3H), 2.30 (s, 3H), 3.31 (s, 2H), 3.83 (s, 2H), 3.92 (s,
2H), 4.19 (q, 2H), 7.31 (t and d, 6H), 9.82 (s, 1H). 13C NMR (dC,
CDCl3): 14.12, 20.23, 50.76, 53.28, 57.29, 60.86, 116.32, 119.19,
120.31, 121.64, 124.83, 129.07, 129.10, 129.23, 133.32, 139.70,
157.56, 158.11, 170.68, 196.43.
X-Ray diffraction data for a crystal of [Fe4(HPBA)2)(l-O)(l-
OH)(OAc)2(OH2)2]ClO4·5H2O were collected with a Bruker Apex
II diffractometer with graphite monochromated Mo-Ka (k =
˚
0.71073 A) radiation. The structure was solved by direct methods
using SIR9745 and refined on F2 using SHELXL9746 running
within the WinGX interface.47 Plots were drawn using ORTEP3.48
The hydrogen positions for the amine, methylene and coordi-
nated water molecules were calculated and included in the final
refinement cycle. The hydrogen atoms of the water molecules of
crystallization were not located. All non-hydrogen atoms were
refined with anisotropic thermal parameters. Selected crystal data
and some details of refinements are given in Table 1. Selected bond
distances and angles are presented in Table 2.
Methyl 2-((2-hydroxy-5-methyl-3-((pyridin-2-ylmethylamino)-
methyl)benzyl)(2-hydroxybenzyl)amino)acetate,
H2HPBMA.
[(3-Formyl-2-hydroxy-5-methyl-benzyl)-(2-hydroxy-benzyl)-amino]-
acetic acid ethyl ester (5.35 g, 0.016 mol) was dissolved in
methanol (500 cm3). A solution of 2-aminomethylpyridine
(1.64 g, 0.0152 mol) in methanol (150 cm3) was added dropwise
with stirring. The opaque, bright yellow solution was stirred
at 50 ◦C for 2 h during which time it became clear. Sodium
borohydride (2 g, 0.053 mol) was added in portions with stirring,
the solution heated to reflux for 2 h and then cooled to room
temperature overnight. The solvent was then removed to leave a
bright orange/red solid. Water (100 cm3) and dichloromethane
(100 cm3) were added, and the pH of the aqueous layer adjusted to
about 4 using concentrated hydrochloric acid. The aqueous layer
was extracted using dichloromethane (2 × 100 cm3), dried over
anhydrous sodium sulfate, filtered and the solvent removed to
leave a yellow/orange solid (3.14 g, 48%). The crude product was
purified by column chromatography on silica, loading onto the
column with chloroform and eluting with a chloroform–methanol
mixture (95 : 5). The solvent was removed to give the product as
an orange-yellow oil (0.7 g, 11% yield). Found: C, 66.38; H, 6.40;
N, 8.59%. Calc. for C25H29N3O4.CH3OH: C, 66.77; H, 7.11; N,
8.99%. LRMS (CH3OH): m/z (assigned ion) 436 ([M + H+]+,
calculated for [C25H30N3O4]+ 436, 458 ([M + Na+]+, calculated
for [C25H28N3O4Na]+ 458). HRMS (CH3OH): m/z (assigned ion)
436.2251 ([M + H+]+, calculated for [C25H30N3O4]+ 436.2236. H
NMR (dH, CDCl3): 2.20 (s, 3H), 3.34 (s, 2H), 3.66 (s, 3H), 3.83
(s, 2H), 3.92 (s, 2H), 3.93 (s, 2H), 3.99 (s, 2H), 6.75 (t, 1H), 6.76
(s, 1H), 6.83 (d, 1H), 6.91 (s, 1H), 7.00 (d, 1H), 7.16 (t, 1H), 7.22
(t, 1H), 7.28 (d, 1H), 7.66 (t, 1H), 8.57 (d, 1H). 13C NMR (dC,
CDCl3): 20.39, 51.45, 51.62, 52.62, 52.72, 53.19, 56.44, 116.25,
119.02, 122.06, 122.61, 122.99, 127.93, 128.98, 129.43, 129.55,
131.20, 136.89, 149.32, 154.64, 157.61, 171.39.
Syntheses of the ligand and metal complex
2-Hydroxy-3-hydroxymethyl-5-methylbenzaldehyde and 2-hy-
droxy-3-chloromethyl-5-methylbenzaldehyde were prepared fol-
lowing a previously described method.49
N-(2-Hydroxybenzyl)glycine ethyl ester. The preparation of
this compound was adapted from the method described by Koh.50
Glycine ethyl ester hydrochloride (4.17 g, 29.9 mmol) was dissolved
in aqueous potassium hydroxide (30 cm3, 1 M). A solution of
salicylaldehyde (3.66 g, 30 mmol) in ethanol (30 cm3) was added
dropwise and the resulting yellow solution was stirred for 15 min.
Methanol (30 cm3) was added dropwise and the solution became
a clear yellow. The solution was stirred at room temperature for
another 2 h and then transferred to an ice bath for 10 min. Sodium
borohydride (2 g, 0.053 mol) was added in portions whilst the
solution remained in ice. The solution bleached and hydrogen
gas was evolved. The solution was stirred at room temperature
for a further 15 min, after which time the pH was adjusted to
about 4 using glacial acetic acid. Further sodium borohydride
was added until the yellow solution bleached, and the solvent was
removed. Water (75 cm3) was added to the flask to give a white
precipitate and the mixture was extracted with dichloromethane
(3 × 50 cm3). The combined organic layers were washed with a
saturated sodium hydrogen carbonate solution (3 × 30 cm3), dried
over anhydrous sodium sulfate, filtered and the solvent removed to
leave a pale yellow oil that crystallised in the refrigerator (yellow
crystals, 2.23 g, 36% yield). H NMR (dH, CDCl3): 1.28 (t, 3H), 3.40
(s, 2H), 3.96 (s, 2H), 4.22 (q, 2H), 6.82 (t, 1H), 6.84 (d, 1H), 6.97
(d, 1H), 7.18 (t, 1H). 13C NMR (dC, CDCl3): 14.05, 48.77, 51.87,
61.11, 116.36, 119.16, 121.78, 128.64, 128.93, 157.71, 171.26.
[Fe4(HPBA)2(OAc)2(l-O)(l-OH)(OH2)2]ClO4·8H2O. A me-
thanol solution (3 cm3) of HHPBMA (0.0477 g, 0.11 mmol) was
added dropwise with stirring to a methanolic solution (10 cm3)
of FeCl2·4H2O (0.0414 g, 0.208 mmol) to give a dark purple
solution. A methanol solution (1 cm3) of CH3COONa (0.0273 g,
0.333 mmol) was immediately added dropwise to the solution
and the solution then stirred at 40 ◦C for 15 min. A methanol
solution (1 cm3) of NaClO4 (0.0552 g, 0.451 mmol) was then added
dropwise and stirring continued at 40 ◦C for 10–15 min. Aqueous
KOH (1 M, 5 drops) was added to the solution and stirring
continued for a further 5 min at 40 ◦C. The solution was then
filtered and left to stand at room temperature to yield fine, cylin-
drical, purple crystals (24.2 mg, 4.05% yield). Found: C, 41.76;
H, 4.47; N, 5.55; Cl, 2.50%. Calc. for C52H59Fe4N6O20Cl·8H2O:
C, 41.87; H, 5.07; N, 5.64; Cl, 2.35%. LRMS (CH3CN): (as-
signed ion; m/z), ([Fe4(HPBA)2)(l-O)(l-OH)(OAc)–H+]+; 1151),
[(3-Formyl-2-hydroxy-5-methyl-benzyl)-(2-hydroxy-benzyl)-
amino]acetic acid ethyl ester. N-(2-Hydroxybenzyl)glycine ethyl
ester (2.16 g, 10.32 mmol) was dissolved in tetrahydrofuran
(100 cm3) and 2-hydroxy-3-chloromethyl-5-methylbenzaldehyde
(2.19 g, 13.18 mmol) was added with stirring to the bright yellow
solution which turned a duller yellow. Triethylamine (5.5 cm3,
39.57 mmol) was added dropwise and the solution changed from
clear to opaque with some pale yellow precipitate forming. The
mixture was left stirring overnight, gravity filtered and the solvent
removed to leave a yellow oil. Water (35 cm3) was added to the
oil to give a yellow precipitate. The mixture was extracted using
5138 | Dalton Trans., 2007, 5132–5139
This journal is
The Royal Society of Chemistry 2007
©

View Article Online
([Fe4(HPBA)2)(l-O)(l-OH)–2H+]+; 1091), ([Fe4(HPBA)2)(l-O)–
3H+]+; 1073), [Fe2(HPBA)(OAc)(OH)(H2O)2]+; 642). UV-vis
(CH3CN), kmax/nm (e/M−1 cm−1): 481 (6054), 313 (sh) (14643),
277 (25248), 234 (sh) (38524), 217 (52066).
23 D. L. Wang, R. C. Holz, S. S. David, L. Que, Jr. and M. T. Stankovich,
Biochemistry, 1991, 30, 8187.
24 P. V. Bernhardt, G. Schenk and G. J. Wilson, Biochemistry, 2004, 43,
10387.
25 A. Neves, M. A. de Brito, V. Drago, K. Griesar and W. Haase, Inorg.
Chim. Acta, 1995, 237, 131.
26 A. Horn, Jr., I. Vencato, A. J. Bortoluzzi, R. Ho¨rner, R. A. Nome Silva,
B. Spoganicz, V. Drago, H. Terenzi, M. C. B. de Oliveira, R. Werner,
W. Haase and A. Neves, Inorg. Chim. Acta, 2005, 358, 339.
27 L. Westerheide, F. K. Mu¨ller, R. Than and B. Krebs, Inorg. Chem.,
2001, 40, 1951.
28 C. Duboc-Toia, S. Me´nage, J.-M. Vincent, M. T. Averbuch-Pouchot
and M. Fontecave, Inorg. Chem., 1997, 36, 6148.
Acknowledgements
This work was funded by a grant from the Australian Research
Council (DP0558652). We thank Dr Yiannis Sanakis for the
acquisition of the Mo¨ssbauer spectra.
29 C. E. McBeth, R. Gupta, K. R. Mitchell-Koch, V. G. Young, Jr., G. H.
Lushington, W. H. Thompson, M. P. Hendrich and A. S. Borovik,
J. Am. Chem. Soc., 2004, 126, 2556.
30 T. Tanase, C. Inoue, E. Ota, S. Yano, M. Takahashi and M. Takeda,
Inorg. Chim. Acta, 2000, 297, 18.
References
1 S. J. Lippard, Angew. Chem., Int. Ed. Engl., 1988, 27, 344.
2 J. D. M. Kurtz, Chem. Rev., 1990, 90, 585.
31 R. T. Stibrany and S. M. Gorun, Angew. Chem., 1990, 102, 1195.
32 S. M. Gorun, R. T. Stibrany and A. Lillo, Inorg. Chem., 1998, 37,
836.
3 A. L. Feig and S. J. Lippard, Chem. Rev., 1994, 94, 759.
4 G. A. Clegg, J. E. Fitton, P. M. Harrison and A. Treffry, Prog. Biophys.
Mol. Biol., 1980, 36, 53.
5 K. Wieghardt, Angew. Chem., Int. Ed. Engl., 1994, 33, 725.
6 V. L. Pecoraro, M. J. Baldwin and A. Gelasco, Chem. Rev., 1994, 94,
807.
7 V. K. Yachandra, V. J. DeRose, M. J. Latimer, I. Mukerji, K. Sauer and
M. Klein, Science, 1993, 260, 675.
8 V. J. DeRose, I. Mukerji, M. J. Latimer, V. K. Yachandra, K. Sauer and
M. P. Klein, J. Am. Chem. Soc., 1994, 116, 5239.
33 D. M. Prosperio, R. Hoffman and G. C. Dismukes, J. Am. Chem. Soc.,
1992, 114, 4374.
34 A. Neves, M. A. de Brito, I. Vencato, V. Drago, K. Griesar and W.
Haase, Inorg. Chem., 1996, 35, 2360.
35 A. S. Borovik, V. Papaefthymiou, L. F. Taylor, O. P. Anderson and L.
Que, Jr., J. Am. Chem. Soc., 1989, 111, 6183.
36 R. C. Holz, T. E. Elgren, L. L. Pearce, J. H. Zhang, C. J. O’Connor and
L. Que, Jr., Inorg. Chem., 1993, 32, 5844.
37 R. C. Reem, J. M. McCormick, D. E. Richardson, F. J. Devlin, P. J.
Stephens, R. L. Musselman and E. I. Solomon, J. Am. Chem. Soc.,
1989, 111, 4688.
9 T. Tanase, T. Inagaki, Y. Yamada, M. Kato, E. Ota, M. Yamazaki,
M. Sato, W. Mori, K. Yamaguchi, M. Mikuriya, M. Takahashi, M.
Takeda, I. Kinoshita and S. Yano, J. Chem. Soc., Dalton Trans., 1998,
713.
38 R. Werner, S. Ostrovsky, K. Griesar and W. Haase, Inorg. Chim. Acta,
2001, 326, 78.
39 S. Dru¨eke, K. Wieghardt, B. Nuber, J. Weiss, E. L. Bominaar, A. Saaryn,
H. Winkler and A. X. Trautwein, Inorg. Chem., 1989, 28, 4477.
40 J. Jullien, G. Juha´sz, E. Dumas, C. R. Mayer, J. Marrot, E. Rivie`re,
E. L. Bominaar, E. Mu¨nck and F. Se´cheresse, Inorg. Chem., 2006, 45,
6922.
10 N. Mitic´, S. J. Smith, A. Neves, L. W. Guddat, L. R. Gahan and G.
Schenk, Chem. Rev., 2006, 106, 3338.
11 G. Schenk, Y. Ge, L. E. Carrington, C. J. Wynne, I. R. Searle, B. J.
Carroll, S. E. Hamilton and J. de Jersey, Arch. Biochem. Biophys., 1999,
370, 183.
12 T. Klabunde and B. Krebs, Struct. Bonding (Berlin), 1997, 1, 177.
13 M. B. Twitchett and A. G. Sykes, Eur. J. Inorg. Chem., 1999, 2105.
14 N. Stra¨ter, W. N. Lipscomb, T. Klabunde and B. Krebs, Angew. Chem.,
Int. Ed. Engl., 1996, 35, 2024.
15 A. K. Boudalis, N. Lalioti, G. A. Spyroulias, C. P. Raptopoulou, A.
Terzis, A. Bousseksou, V. Tangoulis, J.-P. Tuchagues and S. P. Perlepes,
Inorg. Chem., 2002, 41, 6474.
16 V. A. Grillo, G. R. Hanson, T. W. Hambley, L. R. Gahan, K. S. Murray
and B. Moubaraki, J. Chem. Soc., Dalton Trans., 1997, 305.
17 L. W. Guddat, A. S. McAlpine, D. A. Hume, S. E. Hamilton, J. de
Jersey and J. Martin, Structure, 1999, 7, 757.
18 Y. Lindqvist, E. Johansson, H. Kaija, P. Vihko and G. Schneider, J. Mol.
Biol., 1999, 291, 135.
41 N. N. Greenwood and T. C. Gibb, Mo¨ssbauer Spectroscopy, Chapman
and Hall, London, 1971, pp. 148–164.
42 J. Aussoleil, P. Cassoux, P. de Loth and J.-P. Tuchagues, Inorg. Chem.,
1989, 28, 3051.
43 J. M. Clemente-Juan, C. Mackiewicz, M. Verelst, F. Dahan, A.
Bousseksou, Y. Sanakis and J.-P. Tuchagues, Inorg. Chem., 2002, 41,
1478.
44 F. James and M. Roos, Comput. Phys. Commun., 1975, 10, 345.
45 A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori and R.
Spagna, J. Appl. Crystallogr., 1999, 32, 115.
46 G. M. Sheldrick, SHELXL97, University of Go¨ttingen, 1997.
47 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
48 C. K. Johnson, ORTEP, A Thermal Ellipsoid Plotting Program, Oak
Ridge National laboratory, Oak Ridge, TN, 1965.
49 G.-C. Sun, Z.-H. He, Z.-J. Li, X.-D. Yuan, Z.-J. Yang, G.-X. Wang,
L.-F. Wang and C.-R. Liu, Molecules, 2001, 6, 1001.
50 L. L. Koh, J. O. Ranford, W. T. Robinson, J. O. Svensson, A. L. C. Tan
and D. Wu, Inorg. Chem., 1996, 35, 6466.
19 J. Uppenberg, F. Lindqvist, C. Svensson, B. Ek-Rylander and G.
Andersson, J. Mol. Biol., 1999, 290, 201.
20 N. Stra¨ter, T. Klabunde, P. Tucker, H. Witzel and B. Krebs, Science,
1995, 268, 1489.
21 T. Klabunde, N. Stra¨ter, R. Fro¨hlich, H. Witzel and B. Krebs, J. Mol.
Biol., 1996, 259, 737.
22 S. K. Smoukov, L. Quaroni, X. Wang, P. E. Doan, B. M. Hoffman and
L. Que, Jr., J. Am. Chem. Soc., 2002, 124, 2595.
This journal is
The Royal Society of Chemistry 2007
Dalton Trans., 2007, 5132–5139 | 5139
©