Hexa- and octanuclear iron(iii) salicylaldoxime clusters

Article abstract of DOI:10.1039/c0dt01593h

The syntheses, structures and magnetic properties of six iron complexes stabilised with the derivatised salicylaldoxime ligands Me-saoH₂ (2-hydroxyethanone oxime) and Et-saoH₂ (2-hydroxypropiophenone oxime) are discussed. The four hexanuclear and two octanuclear complexes of formulae [Fe₈O₂(OMe)₄(Me-sao)₆Br ₄(py)₄]·2Et₂O·MeOH (1·2Et₂O·MeOH), [Fe₈O₂(OMe) _3.85(N₃)_4.15(Me-sao)₆(py) ₂] (2), [Fe₆O₂(O₂CPh-4-NO ₂)₄(Me-sao)₂(OMe)₄Cl ₂(py)₂] (3), [Fe₆O₂(O ₂CPh-4-NO₂)₄(Et-sao)₂(OMe) ₄Cl₂(py)₂]·2Et₂O·MeOH (4·2Et₂O·MeOH), [HNEt₃]₂[Fe ₆O₂(Me-sao)₄(SO₄) ₂(OMe)₄(MeOH)₂] (5) and [HNEt₃] ₂[Fe₆O₂(Et-sao)₄(SO ₄)₂(OMe)₄(MeOH)₂] (6) all are built from a series of edge-sharing [Fe₄(μ₄-O)]¹⁰⁺ tetrahedra. Complexes 1 and 2 display a new μ₄-coordination mode of the oxime ligand and join a small group of Fe-phenolic oxime complexes with nuclearity greater than six. The Royal Society of Chemistry 2011.

Full text of DOI:10.1039/c0dt01593h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 2875
www.rsc.org/dalton
PAPER
Hexa- and octanuclear iron(III) salicylaldoxime clusters†
Kevin Mason,a Ian A. Gass,a Fraser J. White,a Giannis S. Papaefstathiou,b Euan K. Brechin*a and
Peter A. Tasker*a
Received 16th November 2010, Accepted 17th January 2011
DOI: 10.1039/c0dt01593h
The syntheses, structures and magnetic properties of six iron complexes stabilised with the derivatised
salicylaldoxime ligands Me-saoH2 (2-hydroxyethanone oxime) and Et-saoH2 (2-hydroxypropiophenone
oxime) are discussed. The four hexanuclear and two octanuclear complexes of formulae
[Fe8O2(OMe)4(Me-sao)6Br4(py)4]·2Et2O·MeOH (1·2Et2O·MeOH), [Fe8O2(OMe)3.85(N3)4.15(Me-
sao)6(py)2] (2), [Fe6O2(O2CPh-4-NO2)4(Me-sao)2(OMe)4Cl2(py)2] (3), [Fe6O2(O2CPh-4-NO2)4(Et-
sao)2(OMe)4Cl2(py)2]·2Et2O·MeOH (4·2Et2O·MeOH), [HNEt3]2[Fe6O2(Me-sao)4(SO4)2-
(OMe)4(MeOH)2] (5) and [HNEt3]2[Fe6O2(Et-sao)4(SO4)2(OMe)4(MeOH)2] (6) all are built from a series
of edge-sharing [Fe4(m4-O)]10+ tetrahedra. Complexes 1 and 2 display a new m4-coordination mode of the
oxime ligand and join a small group of Fe-phenolic oxime complexes with nuclearity greater than six.
Salicylaldoxime ligands (Fig. 1) have been studied extensively
Introduction
for their use in extractive hydrometallurgy, showing great selectiv-
ity for Cu2+ and accounting for approximately 25% of the world’s
copper production.11–13 Our research into Fe-salicylaldoximate
complexes is fuelled not only by an interest in their magnetic
properties, but also due to their role as anti-corrosives in pro-
tective coatings, where phenolic oximes have been used to treat
lightly oxidised Fe surfaces.14 It is postulated that the corrosion
inhibition is due to the formation of polynuclear complexes on
the surface and thus an extensive knowledge of the coordination
chemistry of such ligands with Fe would tell us more about
possible modes-of-action of the corrosion inhibition ability of the
ligands. Herein we report the syntheses, structures and magnetic
behaviour of four hexa- and two octametallic iron complexes
built with the derivatised salicylaldoxime ligands Me-saoH2 (2-
hydroxyethanone oxime) and Et-saoH2 (2-hydroxypropiophenone
oxime).
Polymetallic clusters of iron are being synthesised and studied
for a host of reasons. In bioinorganic chemistry, for example,
they have been employed as models for iron-containing enzymes
such as methane monooxygenase and hemerythrin which both
contain diiron cores, whilst the formation of larger oxy-hydroxy
stabilised molecules can give insight into the formation and
function of Ferritin, a protein containing up to ~4500 Fe centres
that stores and regulates iron in living organisms.1–3 The presence
of five unpaired electrons in high spin Fe3+ ions also makes them
attractive for making magnetically interesting complexes, and
the triangular [Fe3O]7+ and tetrahedral [Fe4O]10+ building blocks
common to the cores of many polymetallic Fe3+ cluster compounds
can lead to fascinating frustration effects.4 Geometric frustration is
at the origin of a variety of phenomena such as spin glass behaviour
or the appearance of unusual jumps and plateaus in the field-
dependence of the magnetisation observed in antiferromagnetic
[Fe3] triangles, [Fe13] Keggin ions and [Fe30] icosidodecahedra –
properties reminiscent of those observed in extended frustrated
lattices such as the Kagome´ lattice.5–7 An added effect of spin
frustration is the presence of degenerate or low-lying excited
spin states, a property that can be exploited in low-temperature
magnetic refrigeration,8 with clusters such as [Fe14] displaying an
enhanced magnetocaloric effect.9,10
aSchool of Chemistry, The University of Edinburgh, West Mains Road,
Edinburgh, UK EH9 3JJ. E-mail: ebrechin@staffmail.ed.ac.uk, Peter.
Tasker@ed.ac.uk; Tel: +44 (0) 131-650-7545
bLaboratory of Inorganic Chemistry, Department of Chemistry, National and
Kapodistrian University of Athens, Panepistimiopolis, 157 71, Zogragfou,
Greece
† CCDC reference numbers 667681, 667684, 794871–794875. For
crystallographic data in CIF or other electronic format see DOI:
10.1039/c0dt01593h
Fig. 1 General structure of the (derivatised) salicylaldoxime ligands.
Me-saoH2, R1 = Me and R2, R3 = H; Et-saoH2 R1 = Et and R2 and
R3 = H.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2875–2881 | 2875
©

View Article Online
[HNEt3]2[Fe6O2(OMe)4(Et-sao)4(SO4)2(MeOH)2] (6)
Experimental
The synthesis was identical to 5 using Et-saoH2 (165 mg,
1.00 mmol) instead of Me-saoH2. The yield was approximately
30%. Elemental analysis found (calc.%) for C54H88Fe6N6O24S2: C
40.38 (40.42), H 5.24 (5.53), N 5.20 (5.24).
Syntheses
All manipulations were performed under aerobic conditions using
chemicals as received, unless otherwisestated. 2-Hydroxyethanone
oxime (Me-saoH2) and 2-hydroxypropiophenone oxime (Et-
saoH2) were synthesised via the reaction of the appropriate ketone
with hydroxylamine and sodium acetate in EtOH, as described in
the literature.15 Complex 5 has been reported previously16 but is
included here to aid discussion.
Physical measurements
Elemental analyses (C, H, N) were performed by the EaStCHEM
microanalysis service. Variable temperature magnetic susceptibil-
ity measurements were made on powdered polycrystalline samples
restrained in eicosane using a Quantum Design MPMS-XL
SQUID magnetometer equipped with a 7 T magnet. Diamagnetic
corrections were applied using Pascal’s constants.
[Fe8O2(OMe)4(Me-sao)6Br4(py)4]·2Et2O·MeOH
(1·2Et2O·MeOH)
Single crystal X-ray crystallography was performed using a
Bruker Smart Apex CCD diffractometer equipped with an Oxford
Cryosystems LT device, using Mo radiation. Data collection
parameters and structure solution and refinement details are listed
in Table 1. Full details can be found in the CIF files provided in
the supporting information (CCDC 794871–7947875).†
FeBr3 (148 mg, 0.50 mmol) and Me-saoH2 (226.5 mg, 1.50 mmol)
were dissolved in MeOH (25 ml) in the presence of pyridine (2 ml).
The dark red solution was stirred for 2 h, filtered and then diffused
with Et2O, producing X-ray quality crystals of 1 after 1 week, in
approximately 20% yield. Elemental analysis found (calc.%) for
C81H98Br4Fe8N10O21: C 41.73 (42.04), H 4.16 (4.27), N 5.97 (6.05).
[Fe8O2(OMe)3.85(N3)4.15(Me-sao)6(py)2] (2)
Results and Discussion
FeF3·3H2O (167 mg, 1.00 mmol) and Me-saoH2 (151 mg,
1.00 mmol) were dissolved in a mixture of MeOH (25 ml) and
pyridine (2 ml) and stirred for 5 min. NaN3 (130 mg, 2.00 mmol)
was then added and the dark red solution was stirred for a further
120 min. The solution was filtered and left to evaporate slowly,
producing X-ray quality crystals of 2 after 3 days in approximately
20% yield. Elemental analysis found (calc.%) for C72H73Fe8N23O18:
C 43.14 (43.34), H 3.42 (3.69), N 16.02 (16.15).
Synthesis
Reacting FeBr3 with Me-saoH2 in a 1 : 3 ratio in a MeOH–pyridine
solution yields the octanuclear complex [Fe8O2(OMe)4(Me-
sao)6Br4(py)4] (1), which has a metallic skeleton of two edge-
sharing [Fe4O] tetrahedra, edge-capped by two additional Fe3+
ions. The oxime ligand displays a m4-coordination mode – the
first time this has been observed. We can replace the terminal
bromide ligands with terminal azides by simply introducing NaN3
into the reaction mixture. This also has the effect of partially
substituting a m-methoxide bridge with an end-on m-azide bridge
affording the complex [Fe8O2(OMe)3.85(N3)4.15(Me-sao)6(py)2] (2).
Further attempts to fully substitute the m-methoxide bridges have
proved unsuccessful, even when employing a large excess of NaN3.
Attempts to repeat these reactions with other iron halide salts have
failed thus far. It is interesting to note that (in Fe3+ chemistry) only
two examples of end-on bridging azides appear in the CCDC
database – and both display ferromagnetic exchange coupling
between the Fe3+ centres.17,18
[Fe6O2(OMe)4(O2CPh-4-NO2)4(Me-sao)2Cl2(py)2] (3)
FeCl3·6H2O (270 mg, 1.00 mmol), Me-saoH2 (151 mg, 1.00 mmol)
and NaO2CPh-4-NO2 (189 mg, 1.00 mmol) were dissolved in
MeOH (25 ml) in the presence of pyridine (2 ml). The dark
red solution was stirred for 2 h, filtered and then diffused with
Et2O, producing X-ray quality crystals of 3 after 1 week. The yield
was approximately 30%. Elemental analysis found (calc.%) for
C61H61Cl2Fe6N8O28: C 40.94 (41.26), H 3.05 (3.49), N 6.26 (6.37).
Introducing carboxylates (in the form of Na(O2CPh-4-NO2))
to the reaction of FeCl3·6H2O and Me-saoH2 in a 1 : 1 : 1 ratio
in a MeOH–py solution leads to the related but lower nuclear-
ity hexanuclear complex [Fe6O2(OMe)4Cl2(O2CPh-4-NO2)4(Me-
sao)2(py)2] (3). The core again consists of two edge-sharing
tetrahedra but the replacement of the m4-bridging oximes with
m-bridging carboxylates prevents the addition of two addi-
tional edge-capping Fe3+ ions. Repeating the same reaction
with Et-saoH2 instead of Me-saoH2 simply gives the analogous
compound, [Fe6O2(OMe)4(O2CPh-4-NO2)4(Et-sao)2Cl2(py)2] (4).
Despite many attempts, we failed to isolate compounds with
carboxylates other than 4-NO2-benzoate. It is unclear why this
should be so, but we presume it to be due to the steric bulk of the
4-substituent and/or the stabilising inter-molecular interactions
propagated between neighbouring NO2 groups (vide infra).
The introduction of co-ligands, such as carboxylates, thus
appears to favour the formation of smaller clusters and
[Fe6O2(OMe)4(O2CPh-4-NO2)4(Et-sao)2Cl2(py)2]·2Et2O·MeOH
(4·2Et2O·MeOH)
The synthesis was identical to that of compound 3 using Et-
saoH2 (165 mg, 1.00 mmol) instead of Me-saoH2. The yield
was approximately 30%. Elemental analysis found (calc.%) for
C69H80Cl2Fe6N8O29: C 43.45 (43.82), H 4.04 (4.26), N 5.77 (5.92).
[HNEt3]2[Fe6O2(OMe)4(Me-sao)4(SO4)2(MeOH)2] (5)
Fe2(SO4)3·6H2O (508 mg, 1.00 mmol) and Me-saoH2 (151 mg,
1.00 mmol) were dissolved in a solution of MeOH (30 ml) in
the presence of NEt3 (404 mg, 4 mmol). The solution was stirred
for 1 h and then diffused with Et2O, producing X-ray quality
crystals after 3 days. The yield was approximately 30%. Elemental
analysis found (calc.%) for C50H80Fe6N6O24S2: C 38.96 (38.78), H
5.10 (5.21), N 5.38 (5.43).
2876 | Dalton Trans., 2011, 40, 2875–2881
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 1 Crystallographic details for complexes 1–4, 6
1·2Et2O·MeOH
2
3
4·2Et2O·MeOH
6
M/g mol--1
Crystal system
Space group
2314.13
1985.63
Monoclinic
P21/c
11.9923(4)
28.0040(10)
13.3217(5)
90
114.382(2)
90
4074.8(3)
150
1760.34
Triclinic
P1
1891.41
Monoclinic
P21/c
13.5512(19)
29.198(4)
21.050(3)
90
101.368(6)
90
8165(2)
100
1604.52
Orthorhombic
Pbca
20.4582(4)
14.4149(3)
22.6752(4)
90
90
90
6687.0(2)
150
4
Hexagonal
¯
¯
R3
˚
a/A
42.1099(11)
42.1099(11)
13.8971(4)
90
16.1705(4)
27.5304(7)
29.7123(8)
114.5420(10)
90.8730(10)
93.4590(10)
11998.6(5)
100
˚
b/A
˚
c/A
a/◦
b/◦
g /◦
90
120
21341.4(10)
150
9
3
˚
V/A
T/K
Z
2
6
4
r
calc/g cm--3
1.621
1.618
1.462
1.539
1.594
Crystal shape and colour
Crystal size/mm
m/mm--1
Black plate
0.31 ¥ 0.19 ¥ 0.07
2.943
Black plate
0.43 ¥ 0.28 ¥ 0.12
1.463
Black block
0.44 ¥ 0.44 ¥ 0.26
1.205
Black plate
0.28 ¥ 0.25 ¥ 0.08
1.187
Black block
0.26 ¥ 0.24 ¥ 0.15
1.410
Unique data
12 493
8336
48 626
15 647
5902
Unique data, (I > 2s(F))
7412
0.0560
0.0466, 0.1181
0.929
6650
0.0536
0.0424, 0.1012
1.034
31507
0.0531
0.0508, 0.1284
1.002
9754
0.0830
0.0701, 0.2073
1.111
5602
0.0689
0.0826, 0.1547
1.362
Rint
R1a, wR2b
Goodness of fit
ꢀ
ꢀ
ꢀ
ꢀ
a R1) (jFoj - jFcj)/ (jFoj) for observed reflections. b wR2) { [w(Fo - Fc2)2]/ [w(Fo2)2]} for all data.
2
1/2
m-coordination of the phenolic oxime. This is corroborated
when sulfate anions are introduced to the reaction mixture.
The reaction of Fe2(SO4)3·6H2O with Me-saoH2 or Et-saoH2 in
the presence of NEt3 in MeOH produces the hexanuclear com-
plexes [HNEt3]2[Fe6O2(Me-sao)4(SO4)2(OMe)4(MeOH)2] (5) and
[HNEt3]2[Fe6O2(Et-sao)4(SO4)2(OMe)4(MeOH)2] (6). The metallic
core of both (analogous) molecules again comprises two edge-
sharing [Fe4O] tetrahedra but on this occasion each tripodal
2-
SO4 anion caps one of the triangular faces of the tetrahedron,
preventing further growth.
Description of structures
¯
Complex 1 crystallises in the trigonal space group R3with nine
molecules in the unit cell (Fig. 2). The metallic skeleton consists
of two edge-sharing tetrahedra (Fe1–Fe2–Fe2¢–Fe3 and symmetry
equivalent, s.e.) with the Fe1 ◊ ◊ ◊ Fe3 (and s.e) vertices capped by
another Fe3+ ion (Fe4). The shared edge of the tetrahedra is
defined by Fe2–Fe2¢. Each [FeIII4] tetrahedron houses a central
m4-O2- ion (O123 and symmetry equivalent), with the bonding
along the edges consisting of a combination of single O-atom
bridges from m-OMe- ions and double N–O atom bridges from
Me-sao2- ligands. The latter display three different bonding modes:
Fig. 2 The molecular structure of 1 (top) and its metallic core (bottom).
Colour code: Fe = olive green; O = red; N = dark blue; C = gold; Br = light
blue.
h :h :h :m4, h :h :h :m3 and h :h :h :m. The m-Me-sao2- ligands
bridge Fe2¢–Fe3 (and s.e.) through the NO oximic group; the m3-
Me-sao2- ligands bridge Fe1–Fe4 and Fe3–Fe4 through the two
atom N—O bridge and Fe1–Fe3 through the m-oximic O-atom;
and the m4-Me-sao2- ligand – seen here for the first time – bridges
Fe1, Fe4 and Fe4 through the NO double atom bridge and Fe1
and Fe2 through the phenolic O-atom. This tetranucleating motif
for the salicylaldoxime may provide an explanation for how such
ligands attach to lightly corroded iron surfaces when they are used
as anti-corrosives.14
2
1
2
1
1
2
1
1
1
hedron (Fe3), and the other bridges across one edge of the
central tetrahedron (Fe1–Fe2 and s.e). Each Fe3+ ion is in a dis-
torted octahedral geometry (cis, 74.3(1)-107.1(1)◦; trans, 156.4(1)-
177.12(7)◦) with FeO5N (Fe2, Fe3), FeO5Br (Fe1) and FeO3N2Br
(Fe4) coordination spheres. The remaining coordination sites on
Fe1, Fe3 and Fe4 are filled with a combination of terminally
bonded Br- ions and/or pyridine molecules.
There are two symmetry inequivalent OMe- ions: one bridges
the edge-capping peripheral Fe3+ ion (Fe4) to the central tetra-
In the crystal, the molecules interact through two comple-
mentary C–H ◊ ◊ ◊ p interactions [C5E–H5E ◊ ◊ ◊ p (C1A, C2A, C3A,
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2875–2881 | 2877
©

View Article Online
◦
˚
C4A, C5A, C6A), C ◊ ◊ ◊ centroid 3.543 A, C–H ◊ ◊ ◊ centroid 124 ]
to form one-dimensional (1D) chains running along the c-axis
(Fig. 3). Despite the absence of any other inter-molecular hydrogen
bonds or p ◊ ◊ ◊ p interactions [the closest inter-molecular contacts
being between the phenyl ring of the m4-Me-sao2- ligand and a
˚
neighbouring pyridine molecule (C ◊ ◊ ◊ C, 3.321(7) A) and between
methyl and phenyl groups on opposing m3-Me-sao2- ligands
˚
(C ◊ ◊ ◊ C, 3.362(5) A)], when viewed down the c-axis the chains
of 1 pack in groups of three around the three-fold screw axes
(either left or right handed) imposed by the rhombohedral lattice
˚
to create large hexagonal cavities approximately 11 A in diameter
(Fig. 3).
Fig. 4 The molecular structure of 3 (top) and its metallic core (bottom).
Colour code: Fe = olive green; O = red; N = blue; C = gold; Cl = bright
green.
sao2- ligands bridging Fe1–Fe2 through their oximic NO moieties.
The m-oximic O-atom of one then bridges Fe1–Fe4 while the other
bridges Fe2–Fe3. The four m-O2CPh-4-NO2 and four m-OMe-
ligands bridge the remaining vertices, the carboxylates bridging
Fe2–Fe3, Fe3–Fe5, Fe4–Fe5 and Fe4–Fe6 in their familiar syn,
syn, m-mode and the m-OMe- bridging Fe1–Fe5, Fe2–Fe3, Fe3–
Fe5 and Fe4–Fe6. The two remaining coordination sites on Fe5
and Fe6 are both occupied with one pyridine molecule and one
halide ligand. Each Fe3+ ion is a distorted octahedral geometry
(cis, 77.9(1)–103.9(1)◦; trans, 157.6(1)–177.7(1)◦) with FeO6 (Fe3,
Fe4), FeO5N (Fe1, Fe2) and FeO4NCl (Fe5, Fe6) coordination
spheres. In the crystal there are a number of inter-molecular
interactions propagated through neighbouring 4-NO2 groups of
Fig. 3 The 1D chains of 1 along the c-axis formed through C–H ◊ ◊ ◊ p
interactions (top). Packing diagram of the chains of 1 (in red, blue and
green) around the left handed (yellow arrow) and right handed (cyan)
three-fold screw axes, emphasizing the hexagonal cavities (bottom).
Complex 2 crystallises in the monoclinic space group P21/c
with two molecules in the unit cell. The molecule differs only
slightly from 1. Besides the small changes in bond lengths and
angles, the obvious differences are the replacement of the terminal
bromide ions with terminally bonded azide ligands and the partial
˚
the carboxylates (O ◊ ◊ ◊ O, 3.011(4)–3.184(4) A), as well as between
replacement of a m-OMe- bridge with an end-on m-N3 bridge
the chloride ions and neighbouring pyridine molecules (Cl ◊ ◊ ◊ C,
-
˚
(15% on Fe1–Fe2). This is an interesting observation since the
introduction of end-on bridging azides to Fe3+ cluster chemistry
is likely to produce molecules exhibiting ferromagnetic nearest
neighbour exchange.17,18 All Fe3+ ions have distorted octahedral
geometries with cis angles in the range 74.09(9)–107.78(9)◦ and
trans angles in the range 156.75(9)–179.4(1)◦. There are no inter-
molecular hydrogen bonds in the extended lattice with the closest
contacts being between the phenyl ring of the m4-Me-sao2- ligand
3.400(6) A).
Complex 4 crystallises in monoclinic P21/c space group with
four molecules in the unit cell. 4 differs from 3 only in the identity
of the phenolic oxime, Et-sao2- replacing Me-sao2- and the intra-
molecular bond lengths and angles are very similar in both com-
pounds. In the crystal lattice there are inter-molecular hydrogen
bonds between the 4-NO2 group of carboxylate ligands and MeOH
˚
solvent molecules (O ◊ ◊ ◊ C, 3.00(1) A), with the shortest contact
˚
˚
and a neighbouring pyridine molecule (C ◊ ◊ ◊ C 3.370(5) A).
Complex 3 (Fig. 4) crystallises in the triclinic space group P1with
between neighbouring cluster molecules (3.40(1) A) being between
C-atoms on the phenyl rings of adjacent carboxylate and Me-sao2-
¯
three molecules in the asymmetric unit and two asymmetric units
in the unit cell. The metallic core consists of two edge-sharing
(Fe3–Fe4) tetrahedra (Fe1, Fe3, Fe4, Fe5 and Fe2, Fe3, Fe4,
Fe6), each housing a central m4-O2- ion (O134, O234). The two
ligands.
Complex 6 (Fig. 5) crystallises in the orthorhombic space
group Pbca with four molecules in the unit cell. Its structure is
analogous to that of 5 (but containing Et-sao2- rather than Me-
sao2-), which we have reported in a previous paper.16 6 is another
1
1
2
tetrahedral subunits are joined together by the two h :h :h :m3-Me-
2878 | Dalton Trans., 2011, 40, 2875–2881
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Discussion
There are now a number of Fe3+/R-saoH2 clusters in the literature
(Table 2).14,16,19–26 They range in size (£ [Fe12]) and structure,
but there are undoubtedly some pervading themes. The body of
published work in this area (Table 2) clearly shows that, akin to
Mn3+/R-saoH2 chemistry, the [Fe3(m3-O)]7+ triangle is the most
frequently encountered building block.26 The related tetrahedral
subunit [Fe4(m4-O)]10+ is the next most common, and this has no
counterpart in Mn3+ chemistry. The variance in size of the [Fex]
cluster can be attributed to a number of factors, including the
presence or absence of co-ligand (here carboxylates and sulfates)
and the steric bulk of the ketoxime group. This is another area in
which we see a difference in comparison with the Mn3+ chemistry,
where all of the above have little or no effect upon the products of
the reaction.27–29 The addition of (relatively large) bridging co-
ligands tends to favour smaller nuclearity clusters (they edge-
and face-cap preventing further growth) and encourages the m-
bridging mode of the phenolic oxime. In the absence of such co-
ligands higher order bridging modes of R-sao2- are seen (and
higher nuclearity clusters as a consequence) and in complexes 1
and 2 the m4-bridging mode is seen for the very first time. The
bridging modes depicted in Fig. 5 thus offer some insight into
possible ligand bonding modes on lightly corroded iron surfaces
when salicylaldoximes are used as anti-corrosives.14
Fig. 5 The molecular structure of 6 (top) and its metallic core (bottom).
Colour code: Fe = olive green; O = red; N = blue; C = gold; S = yellow.
Variations in the bulk of the ketoxime group can also change the
topology of the cluster greatly in Fe3+ chemistry,26 whereas in Mn3+
chemistry the [M3O(R-sao)3]+ unit is almost always retained and
the only differences observed appear to be in the twisting of the
Mn–N–O–Mn torsion angles.27 For example, both complex 1 and
[Fe3O(OMe)(Ph-sao)2Br2(py)3]·Et2O26 are made by reacting FeBr3
with the appropriate R-saoH2 pro-ligand in a MeOH–pyridine
solvent mix. The bulky Ph-sao2- ligand restricts the size of the
cluster to an [Fe3] triangle, but the Me-sao2- ligand allows the
growth of an octametallic cluster, [Fe8].
example of an [FeIII6] cluster whose metallic core describes two
edge sharing tetrahedra. The common edge is Fe1–Fe1¢ with an
inversion centre at its midpoint. The tetrahedra are built upon
two central m4-O2- ions (O123 and s.e.) and connected via four m-
OMe- ions (O1A, O14 and s.e.) creating a [Fe6O2(OMe)4]10+ core
1
1
1
2-
similar to that observed in 3 and 4. The h :h :h :m3-SO4 ligands
cap the Fe1–Fe2–Fe3 (and s.e.) triangular face of a tetrahedron,
1
1
1
whilst the four h :h :h :m-Me-sao2- ligands bridge the Fe1–Fe2
and Fe1–Fe3 (and s.e.) edges. The remaining coordination sites
on Fe1 and Fe1¢ are filled by terminal MeOH molecules. The
Fe3+ ions have FeO6 (Fe1) or FeO5N coordination spheres and
lie in distorted octahedral geometries with cis angles in the range
76.9(2)–108.1(2)◦ and trans angles in the range 152.6(2)–175.4(2)◦.
An examination of the extended lattice reveals short contacts
When employing carboxylates in Fe3+/R-saoH2 chemistry one
might expect the structures to resemble the basic Fe3+ carboxylates
of general formula [Fe3O(O2CR)6L3]+ (L = solvent) in which
m
2/3-bridging R-sao2- ligands simply replace the m-bridging car-
boxylates. Indeed this is true in [Fe6] clusters where the [Fe3(m3-
O)]7+ building block predominates,23,24,26 and the m3-bridging oxime
promotes oligomerisation of the basic triangles. Complexes 3
and 4 differ, the major change in reaction conditions being
the introduction of pyridine. This results in the formation of a
cluster adopting a metallic skeleton comprising two edge-sharing
tetrahedra. Pyridine is present in excess and thus acts as base,
ligand and co-solvent – a strategy that has been shown previously
to aid the growth of very large mineral-like Fe3+ clusters.30
+
2-
˚
between the NEt4 cation and the SO4 (N ◊ ◊ ◊ O, 2.82(1) A), with
the closest distances between neighbouring clusters being between
the phenyl ring of the Et-sao2- ligand and a SO42- O-atom (C ◊ ◊ ◊ O,
˚
3.412(8) A). There are also intra-molecular H-bonds present –
between a terminally bonded MeOH molecule and an oximic
˚
O-atom (O ◊ ◊ ◊ O, 2.737(6) A). A scheme depicting the observed
bridging modes of the R-sao2- ligands is given in Fig. 6.
Magnetism
Direct current (dc) magnetic susceptibility measurements were
performed on microcrystalline samples of representative 1 and
3 in a field of 0.1 T and in the 5–300 K temperature range. The
magnetic behaviour of complex 5 is analogous to that of complex
6 which we have reported before, so will not be repeated here.17
For 1 (Fig. 7) the room temperature cmT value of 12.38 cm3 K
mol-1 is significantly below the value of 35 cm3 K mol-1 expected
for eight non-interacting high spin (S = 5/2) Fe3+ ions, indicative
Fig. 6 The coordination modes of the phenolic oximes in complexes 1–6.
Colour code: Fe = olive green; O = red; N = blue; C = gold.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2875–2881 | 2879
©

View Article Online
Table 2 The structurally characterised Fe/R-sao2- complexes in the CCDC database
Molecule
Core
Oxime coordination
Ref.
[HNEt3][Fe2(OMe)(Ph-sao)2(Ph-saoH)2]·5MeOH
[Fe2(sao)3(tmtacn)]·MeOH
[Fe2(OMe)(NO)2]3+ dimer
[Fe2(NO)3]3+ dimer
2 ¥ m, 1 ¥ NO chelate
16
21
21
23
23
16
26
26
20
16
14
26
25
22
24
24
3 ¥ m
[Fe2(3-5-di-tbut-sao)3(tmtacn)]·3.5 CHCl3
[Fe3O(O2CPh)5(sao)(MeOH)2]·1.25MeOH·1.05H2O
[Fe3O(O2CPh)5(sao)(EtOH)(H2O)]·EtOH
[Fe3O (O2CPh)5(Et-sao)(MeOH)2]·3MeOH
[Fe3O(OMe)(Ph-sao)2Cl2(py)3]·2MeOH
[Fe2(NO)3]3+ dimer
3 ¥ m
[Fe3(m3-O)]7+ triangle
[Fe3(m3-O)]7+ triangle
[Fe3(m3-O)]7+ triangle
[Fe3(m3-O)]7+ triangle
[Fe3(m3-O)]7+ triangle
[Fe3(m3-O)]7+ triangle
Distorted [Fe4(NO)4]8+ cube
Distorted [Fe4(NO)4]8+ cube
[Fe4(NO)4F4] square
1 ¥ m
1 ¥ m
1 ¥ m
1 ¥ m
[Fe3O(OMe)(Ph-sao)2Br2(py)3]·Et2O
1 ¥ m
[HNEt3][Fe3O(salmpH3)(sao)2(saoH)].2H2O
[Fe4(Me-sao)4(Me-saoH)4]·MeOH
1 ¥ m, 1 ¥ NO chelate
4 ¥ m3, 4 ¥ NO chelate
[Fe4(Me-sao)4(Me-saoH)4]·saoH2·C8H10
4 ¥ m3, 4 ¥ NO chelate
4 ¥ m
[Fe4(Ph-sao)4F4(py)4]·1.5MeOH
[Fe4O2(O2CCH3)3(sao)2(tmtacn)2][PF6]
[Fe4(m3-O)2]8+ butterfly
[Fe4(m3-O)2]8+ butterfly
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ [Fe3(m3-O)]7+ triangles
2 ¥ edge sharing [Fe4(m4-O)]10+
tetrahedra
2 ¥ m
[Fe4O2(O2CC(OH)Ph2)3(sao)2(tmtacn)2][ClO4]
[Fe6O2(O2CPh)10(sao)2(MeCONH2)2]·6MeCN
[Fe6O2(O2CPh)10(sao)2(H2O)2]·2MeCN·3H2O
[Fe6O2(OH)2(O2CPh)6(Et-sao)2(Et-saoH)2]
[HNEt3]2[Fe6O2(OH)2(O2CPh(Me)2)6(Et-sao)4]·2MeCN
[Fe6Na3O(OH)4(OMe)3(Me-sao)6(H2O)3(MeOH)6]·MeOH
[Fe6O2(O2CPh)10(3-tbut-5-NO2-sao)2(H2O)2]·2MeCN
[Fe6O2(O2CCH2Ph)10(3-tbut-sao)2(H2O)2]·5MeCN
[HNEt3]2[Fe6O2(OMe)4(Me-sao)4(SO4)2(MeOH)2]
2 ¥ m
2 ¥ m3
2 ¥ m3
2 ¥ m, 2 ¥ m3
2 ¥ m, 2 ¥ m3
6 ¥ m
26
26
26
26
2 ¥ m
2 ¥ m
26
4 ¥ m
16/This paper
[HNEt3]2[Fe6O2(OMe)4(Et-sao)4(SO4)2(MeOH)2]
[Fe6O2(O2CPhNO2)4(OMe)4(Me-sao)2Cl2(py)2]
2 ¥ edge sharing [Fe4(m4-O)]10+
tetrahedra
4 ¥ m
2 ¥ m3
2 ¥ m3
3 ¥ m
3 ¥ m
3 ¥ m
This paper
This paper
This paper
16
2 ¥ edge sharing [Fe4(m4-O)]10+
tetrahedra
[Fe6O2(OMe)4(O2CPhNO2)4(Et-sao)2Cl2(py)2]
·2Et2O·MeOH
2 ¥ edge sharing [Fe4(m4-O)]10+
tetrahedra
[Fe8O3(O2CMe)3(Me-sao)3(tea)(teaH)3]
3 ¥ common-edge sharing
[Fe4(m4-O)]10+ tetrahedra
3 ¥ common-edge sharing
[Fe4(m4-O)]10+ tetrahedra
3 ¥ common-edge sharing
[Fe4(m4-O)]10+ tetrahedra
4 ¥ [Fe4(m4-O)]10+ tetrahedra
2 ¥ bicapped [Fe4(m4-O)]10+
tetrahedra
[Fe8O3(O2CMe)3(Et-sao)3(tea)(teaH)3]
[Fe8O3(O2CMe)3(Ph-sao)3(tea)(teaH)3]
16
16
[Fe8O4(sao)8(py)4]·4py
4 ¥ m, 4 ¥ m3
2 ¥ m, 2 ¥ m3, 2 ¥ m4
25
[Fe8O2(OMe)4(Me-sao)6Br4(py)4]·2Et2O·MeOH
This paper
[Fe8O2(OMe)3.85(N3)4.15(Me-sao)6(py)2]
2 ¥ bicapped [Fe4(m4-O)]10+
tetrahedra
2 ¥ m, 2 ¥ m3, 2 ¥ m4
6 ¥ m, 6 ¥ m3
This paper
26
[HNEt3]2[Fe12Na4O2(OH)8(OMe)6(sao)12(MeOH)10]
4 ¥ [Fe3(m3-O)]7+ triangles
Abbreviations: saoH2, Ph-saoH2, see Fig. 1. tmtacn, 1,4,7-trimethyl-1,4,7-triazacyclononane; salmpH3, 2(bis(salicylideneamino)methyl)phenolate(-3);
teaH3, triethanolamine.
The room temperature cmT value of complex 3 is 11.23 cm3 K
mol-1, lower than the expected value for six non-interacting
Fe3+ ions (26.25 cm3 K mol-1). The cmT value then decreases
constantly with decreasing temperature to a value of 0.25 cm3 K
mol-1. This is again indicative of antiferromagnetic exchange
interactions between the Fe3+ ions and the stabilisation of a
diamagnetic ground state. A Curie–Weiss analysis of the 1/cm vs.
T plot affords q = -386 K. The complex nature of the structures
precludes fitting of the susceptibility data by standard procedures.
Of the thirty two complexes listed in Table 2 only two structural
types (and a total of eight molecules) have been reported to
possess non-zero spin ground states – the [Fe3] triangles and
complexes 5 and 6. Both are the result of spin frustration effects
from antiferromagnetic exchange within symmetric metallic cores.
This suggests that future attempts to build Fe3+/R-sao2- clusters
Fig. 7 Plot of cMT vs. T for complexes 1 (squares) and 3 (triangles).
with non-zero spin ground states should focus on the use of
high temperature/high pressure (e.g. solvothermal or microwave)
reaction conditions which are likely to produce highly symmetric
molecules.10
of relatively strong antiferromagnetic exchange interactions. The
cmT value decreases steadily with decreasing temperature reaching
0.79 cm3 K mol-1 at 5 K, consistent with an S = 0 ground state. A
plot of 1/cm vs. T affords q = -115 K.
2880 | Dalton Trans., 2011, 40, 2875–2881
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
9 M. Evangelisti, A. Candini, A. Ghirri, M. Affronte, E. K. Brechin and
E. J. L. McInnes, Appl. Phys. Lett., 2005, 87, 072504.
10 R. Shaw, R. H. Laye, L. F. Jones, D. M. Low, C. Talbot-Eeckelaers,
Q. Wei, C. J. Milios, S. Teat, M. Helliwell, J. Raftery, M. Evangelisti,
M. Affronte, D. Collison, E. K. Brechin and E. J. L. McInnes, Inorg.
Chem., 2007, 46, 4968.
11 P. A. Tasker, P. G. Plieger and L. C. West, Comprehensive Coordination
Chemistry II, 2004, 9, 759.
12 G. A. Kordosky, Proceedings of the International Solvent Extraction
Conference, Cape Town, South Africa, March 17–21, 2002, p. 853.
13 P. J. Mackey, CIM Mag., 2007, 2, 35.
14 J. M. Thorpe, R. L. Beddoes, D. Collison, C. D. Garner, M. Helliwell,
J. M. Holmes and P. A. Tasker, Angew. Chem., Int. Ed., 1999, 38, 1119.
15 R. Dunsten and T. A. Henry, J. Chem. Soc. Trans., 1899, 75, 66.
16 I. A. Gass, C. J. Milios, A. Collins, F. J. White, L. Budd, S. Parsons, M.
Murrie, S. P. Perlepes and E. K. Brechin, Dalton Trans., 2008, 2043.
17 K. Rajender Reddy, M. V. Rajasekharan and J. –P. Tuchagues, Inorg.
Chem., 1998, 37, 5978.
18 G. De Munno, T. Poerio, G. Viau, M. Julve and F. Lloret, Angew.
Chem., Int. Ed. Engl., 1997, 36, 1459.
19 P. Chaudhuri, M. Winter, P. Fleischhauer, W. Haase, U. Florke and
H.-J. Haupt, Inorg. Chim. Acta, 1993, 212, 241.
Conclusions
The two hexa- and four octa-metallic Fe3+ salicylaldiminato
clusters presented here all have a common building block, the
tetrahedral [Fe4O]10+ moiety. Each contains a central core of two
such edge-sharing tetrahedra, with 1 and 2 having two vertices
additionally capped by Fe3+ ions, as a result of a unique m4-Me-
sao2- coordination mode. In contrast to Mn3+ complexes, the self-
assembly of these Fe3+ molecules – and the coordination mode of
the phenolic oxime ligand – appears to be highly dependent upon
the presence of co-ligands and the steric bulk of the ketoxime
group. The magnetic behaviour of all complexes is [perhaps
unsurprisingly] dominated by relatively strong antiferromagnetic
exchange interactions, as seen for almost all reported Fe3+/R-
saoH2 complexes. However, the observation in complex 2 of
the partial replacement of a m-bridging OMe- ion with an end-
-
on m-N3 ion, and the symmetric cores of complexes 5 and 6,
suggests targeting both azide-based clusters and highly symmetric
complexes, since they may pave the way for isolating compounds
displaying frustration effects, and/or ferro- or ferrimagnetic
exchange interactions.
20 E. Bill, C. Krebs, M. Winter, M. Gerdan, A. X. Trautwein, U. Florke,
H.-J. Haupt and P. Chaudhuri, Chem.–Eur. J., 1997, 3, 193.
21 C. N. Verani, E. Bothe, D. Burdinski, T. Weyhermuller, U. Florke and
P. Chaudhuri, Eur. J. Inorg. Chem., 2001, 2161.
22 P. Chaudhuri, E. Rentschler, F. Birkelbach, C. Krebs, E. Bill, T.
Weyhermuller and U. Florke, Eur. J. Inorg. Chem., 2003, 541.
23 C. P. Raptopoulou, Y. Sanakis, A. K. Boudalis and V. Psycharis,
Polyhedron, 2005, 24, 711.
Notes and References
1 H. Basch, K. Mogi, D. G. Musaev and K. Morokuma, J. Am. Chem.
Soc., 1999, 121, 7249.
2 E. C. Theil, Annu. Rev. Biochem., 1987, 57, 289; B. Xu and N. D.
Chasteen, J. Biol. Chem., 1991, 266, 19965.
24 C. P. Raptopoulou, A. K. Boudalis, Y. Sanakis, V. Psycharis, J. M.
Clemente-Juan, M. Fardis, G. Diamantopoulos and G. Papavassiliou,
Inorg. Chem., 2006, 45, 2317.
25 I. A. Gass, C. J. Milios, A. G. Whittaker, F. P. A. Fabiani, S. Parsons,
M. Murrie, S. P. Perlepes and E. K. Brechin, Inorg. Chem., 2006, 45,
5281.
26 K. Mason, I. A. Gass, S. Parsons, A. Collins, F. White, A. M. Z. Slawin,
E. K. Brechin and P. A. Tasker, Dalton Trans., 2010, 39, 2727.
27 R. Inglis, L. F. Jones, C. J. Milios, S. Datta, A. Collins, S. Parsons,
W. Wernsdorfer, S. Hill, S. P. Perlepes, S. Piligkos and E. K. Brechin,
Dalton Trans., 2009, 3403.
28 R. Inglis, L. F. Jones, K. Mason, A. Collins, S. A. Moggach, S. Parsons,
S. P. Perlepes, W. Wernsdorfer and E. K. Brechin, Chem.–Eur. J., 2008,
14, 9117.
3 S. A. Malone, A. Lewin, M. A. Kilic, D. A. Svistuneko, C. E. Cooper,
M. T. Wilson, N. E. Le Brun, S. Spiro and G. R. Moore, J. Am. Chem.
Soc., 2004, 126, 496.
4 J. Schnack, Dalton Trans., 2010, 39, 4677.
5 J. Schnack, M. Luban and R. Modler, Europhys. Lett., 2001, 56, 863.
6 V. O. Garlea, S. E. Nagler, J. L. Zarestky, C. Stassis, D. Vaknin, P.
Ko¨gerler, D. F. McMorrow, C. Niedermayer, D. A. Tennant, B. Lake,
Y. Qiu, M. Exler, J. Schnack and M. Luban, Phys. Rev. B: Condens.
Matter Mater. Phys., 2006, 73, 024414.
7 J. van Slageren, P. Rosa, A. Caneschi, R. Sessoli, H. Casellas, Y. V.
Rakitin, L. Cianchi, F. Del Giallo, G. Spina, A. Bino, A.-L. Barra,
T. Guidi, S. Carretta and R. Caciuffo, Phys. Rev. B: Condens. Matter
Mater. Phys., 2006, 73, 014422.
29 L. F. Jones, R. Inglis, M. E. Cochrane, K. Mason, A. Collins, S. Parsons,
S. P. Perlepes and E. K. Brechin, Dalton Trans., 2008, 6205.
30 G. W. Powell, H. N. Lancashire, E. K. Brechin, D. Collison, S. L. Heath,
T. Mallah and W. Wernsdorfer, Angew. Chem., Int. Ed., 2004, 43, 5772.
8 M. Evangelisti and E. K. Brechin, Dalton Trans., 2010, 39, 4672.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 2875–2881 | 2881
©