Redox processes involved in the synthesis and reactivity of oxazolimlthiophenolato complexes of iron(II)/(III)

Article abstract of DOI:10.1002/ejic.201000065

The reaction of anhydrous FeCl₂ with 2-(4′,4′- dimethyloxazolin-2′-yl)thiophenolate (ox-phS^-) afforded mononuclear [Fe(ox-phS)₂] (A) and binuclear [[Fe^III(ox- phS)}₂(μ-S)₂] (B). In B, iron(III) and S^2- resulted from an unexpected redox reaction involving elemental sulfur and the iron(II) starting material. Complexes A and B co-crystallise reproducibly in a 2:1 proportion. An attempt to prepare (oxazolinylthiophenolato)-iron(III) from Li(ox-phS) and anhydrous FeCl₃ in the presence of N,N,N′,N -tetramethyletha:n.e-1,2-diamine (tmen) gave another redox reaction with disulfide D, bis{2-(4′,4′-di-methyloxazolln-2′-yl)phenyl} disulfide (ox-phS-Sph-ox), and trans-[FeCl₂(tmen)₂] (E) as 1:1 co-crystallised products. Characterisation of all complexes included Moessbauer spectroscopy and single-crystal X-ray diffraction analysis. Quantum, mechanical (TDDFT) calculations for A and cyclic voltammetry experiments carried out with A and C helped to distinguish between ligand- and metal-based electronic transitions and redox processes. Results add to the knowledge of the rich redox chemistry of early transition metals with soft S-donor ligands, with, possible consequences for catalytic and biochemical transformations.

Full text of DOI:10.1002/ejic.201000065

FULL PAPER
DOI: 10.1002/ejic.201000065
Redox Processes Involved in the Synthesis and Reactivity of
Oxazolinylthiophenolato Complexes of Iron(II)/(III)
Rúbia C. R. Bottini,[a] Rogério A. Gariani,[a] César de O. Cavalcanti,[a]
Franciele de Oliveira,[a] Nevilde de L. G. da Rocha,[a] Davi Back,[b] Ernesto S. Lang,[b]
Peter B. Hitchcock,[c] David J. Evans,[d] Giovana G. Nunes,[a] Fábio Simonelli,[a]
Eduardo L. de Sá,[a] and Jaísa F. Soares*[a]
Keywords: Iron / S ligands / Structure elucidation / Redox chemistry / Density functional calculations
The reaction of anhydrous FeCl2 with 2-(4Ј,4Ј-dimethylox- trans-[FeCl2(tmen)2] (E) as 1:1 co-crystallised products. Char-
azolin-2Ј-yl)thiophenolate (ox-phS–) afforded mononuclear
[Fe(ox-phS)2] (A) and binuclear [{FeIII(ox-phS)}2(µ-S)2] (B). In
B, iron(III) and S2– resulted from an unexpected redox reac-
tion involving elemental sulfur and the iron(II) starting mate-
rial. Complexes A and B co-crystallise reproducibly in a 2:1
proportion. An attempt to prepare (oxazolinylthiophenolato)-
iron(III) from Li(ox-phS) and anhydrous FeCl3 in the pres-
ence of N,N,NЈ,NЈ-tetramethylethane-1,2-diamine (tmen)
gave another redox reaction with disulfide D, bis{2-(4Ј,4Ј-di-
methyloxazolin-2Ј-yl)phenyl}disulfide (ox-phS-Sph-ox), and
acterisation of all complexes included Mössbauer spec-
troscopy and single-crystal X-ray diffraction analysis. Quan-
tum mechanical (TDDFT) calculations for A and cyclic vol-
tammetry experiments carried out with A and C helped to
distinguish between ligand- and metal-based electronic tran-
sitions and redox processes. Results add to the knowledge of
the rich redox chemistry of early transition metals with soft S-
donor ligands, with possible consequences for catalytic and
biochemical transformations.
containing carboxylate- or isopropyl-substituted phox, or
those with derived fluorophoric tripodal ligands, are even
more recent.[3,4]
Introduction
Metal oxazoline and bis(oxazoline) complexes have re-
ceived considerable attention in the last few years because
of their use in catalytic processes, particularly asymmetric
syntheses.[1] Among these (pre)catalysts, iron-containing
compounds have been sought in connection with their ac-
cessible synthesis, large metal availability, reasonable cata-
lytic performance, low cost and low toxicity.[2] Some of
these complexes have also been employed in the develop-
ment of fluorescent iron biosensors[3] and as small-molecule
models of natural iron-scavenging siderophores.[4–6]
Reports on the preparation of iron complexes of phenyl-
monooxazolines and their chalcoderivatives (Scheme 1) are
scarce.[7] The crystal structure of [Fe(phox)3], where Hphox
Scheme 1. Metal complexes of (chalco)phenylmonooxazolines.
To the best of our knowledge, no sulfur, selenium or tel-
is the N,O-donor 2Ј-(2-hydroxyphenyl)oxazoline, was only lurium analogues of Hphox have yet been shown to form
published in 2002.[8] Other reports on similar complexes stable complexes with iron, although a number of ZnII, CdII
and HgII adducts are known.[9,10] Considering both the rele-
vance of iron complexes in biochemical media and the es-
[a] Departamento de Química, Universidade Federal do Paraná,
81530-900 Curitiba-PR, Brazil
sential antioxidant activities of small molecule organochal-
cogens – particularly those containing sulfur and selenium –
to prevent cellular damage and death,[11] literature data on
the formation of iron(II)/(III) adducts of simple S- or Se-
donor ligands appear to be still missing.
The present work reveals that redox processes are clearly
involved in the reactivity of iron chlorides towards the mo-
nolithium salt of 2-(4Ј,4Ј-dimethyloxazolin-2Ј-yl)thiophen-
olate, Li(ox-phS). Amongst the products, two new iron ox-
Fax: +55-41-3361-3186
E-mail: jaisa@quimica.ufpr.br
[b] Departamento de Química, Universidade Federal de Santa
Maria,
97105-900 Santa Maria-RS, Brazil
[c] Department of Chemistry, University of Sussex,
Brighton BN1 9QJ, UK
[d] Department of Biological Chemistry, John Innes Centre,
Colney, Norwich NR4 7UH, UK
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201000065.
2476
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
azolinylthiophenolates were obtained and characterised by
spectroscopic, electrochemical and diffractometric tech-
niques. The general aim of this research was to contribute
to the understanding of the redox chemistry of iron ions in
the presence of relatively simple N,S-donor ligands, because
of the relevance of this chemistry for a number of organo-
metallic and bioinorganic systems.
Results and Discussion
Co-crystallised products, as well as pure new compounds,
were obtained in this work. In order to make it easier to
follow the discussion, Table 1 summarises basic information
on the products.
Table 1. Summary of the compounds described in this work.
Product Description
Formulation
Figure
1
Co-crystallised A
and B (2:1)
Pure A
A: [FeII(ox-phS)2]
B: [{FeIII(ox-phS)}2(µ-S)2]
[FeII(ox-phS)2]
1a
1b
–
2
3
2
3
4
Pure C
[ZnII(ox-phS)2]
Co-crystallised D
and E (1:1)
Pure F
D: [FeIICl2(tmen)2]
E: ox-phS-Sph-ox
“FeIICl2(tmen)”
5
–
Products 1–3
Scheme 2. Synthesis of products 1 and 2.
The new iron oxazolinylthiophenolato molecular com-
plexes, A and B, were isolated as part of two distinct crystal- the three compounds. Bands recorded at ca. 520 cm–1 were
line products: dark brown 1 and bright red 2 (Scheme 2). tentatively assigned to metal–N stretches, whereas ν(M–S)
Product 1 is reproducibly obtained upon addition of S8 could not be observed in the instrument detection range.
to the lithium salt of 4Ј,4Ј-dimethyl-2Ј-phenyloxazoline
[Li(ox)]; subsequent reaction with anhydrous FeCl2 is car-
Product 4
The reaction of Li(ox-phS) with anhydrous FeCl3 was
ried out in situ, without isolation of Li(ox-phS). If, alterna-
tively, the lithium thiophenolate is filtered off, washed,
dried and then allowed to react in a 2:1 proportion with
FeCl2, 2 is the only product.
Colourless 3 contains a zinc(II) analogue of A, which
was prepared to help understand the spectroscopic and re-
dox behaviour of the 2-(4Ј,4Ј-dimethyloxazolin-2Ј-yl)thio-
phenolato ligand in a coordination environment that is sim-
ilar to that in A. The synthesis of this complex was reported
earlier by Singh and co-workers,[10] although under dif-
ferent reaction conditions.
carried out in 1:1 and 2:1 proportions in the presence of
N,N,NЈ,NЈ-tetramethylethane-1,2-diamine (tmen). In both
cases, iron(III) was reduced to iron(II) and an organic di-
sulfide was the oxidation product [Equations (1) and (2)].
The precipitation of lithium chloride from the organic me-
dium probably contributed to the driving force of the reac-
tion. The diamine was added to the reaction mixture to help
crystallise iron-containing products, as in our earlier studies
on mono-, di- and trinuclear complexes of iron(II) with
tmen.[12]
FTIR spectra strongly support the presence of oxazolin-
ylthiophenolate as a ligand in all products, as medium-to-
strong bands at 1603, 1592 and 1595 cm–1 can be assigned
to ν(C=N) in A, B and C, respectively. Product 1 contains
two bands in this region (1603 and 1592 cm–1), probably
due to the presence of both A and B in the crystals. All
these bands are shifted to lower wavenumbers relative to
the corresponding absorption of the free phenyloxazoline
(1651 cm–1), which provides evidence for coordination in-
volving the heterocyclic nitrogen. C–S stretches occur at
735, 740 and 738 cm–1, which is higher than in Li(ox-phS)
(1)
(2)
The 1:1 reaction [Equation (1)] generated bis{2-(4Ј,4Ј-di-
(729 cm–1), whereas the δ(C–O–C) vibrations of the five- methyl oxazolin-2Ј-yl)phenyl}disulfide (ox-phS-Sph-ox),
membered rings are registered around 1052–1053 cm–1 in which co-crystallised with trans-[FeCl2(tmen)2] (4) as well
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2477

J. F. Soares et al.
FULL PAPER
as “FeCl2(tmen)” (5).[12] Product 4 was isolated from the
reaction mixture as very pale yellow, well-formed rectangu-
lar prisms, whereas “FeCl2(tmen)” was an off-white, micro-
crystalline, very air-sensitive powder.
When FeCl3 and Li(ox-phS) were allowed to react in a
1:2 proportion, [Fe(ox-phS)2] (A) was obtained together
with product 4 [Equation (2)]. All three compounds (D, E
and A) had their identities confirmed by single-crystal X-
ray diffractometry and were also analysed by FTIR, Möss-
bauer and EPR spectroscopy. trans-[FeCl2(tmen)2] and the
disulfide co-crystallise directly from the mother liquor after
several days at –20 °C, whereas complex A crystallises inde-
pendently, also at low temperature, after separation of 4
followed by careful addition of a hexane layer to the mother
liquor.
The FTIR spectrum of 4 agrees with the presence of both
the bis{2-(oxazolinyl)phenyl}disulfide and tmen in the light
yellow crystals. In this case, the very strong, diagnostic
ν(C=N) band from the oxazoline ring indicated noncoordi-
nation and was registered at almost the same wavenumber
shown by the free oxazoline [1649 and 1651 cm–1 for the
disulfide and H(ox), respectively]. This finds support in the
molecular structure determined for 4 by single-crystal X-
ray diffractometry. The diamine ν(C–N) vibration, in its
turn, gives the relatively broad, strong band recorded at
1032 cm–1, which dominates the spectrum in this region.
Synthesis and X-ray Diffraction Analyses of Products 1–3
Single-crystal X-ray diffractometry (Figures 1 and 2,
with thermal ellipsoids drawn at 50% probability) and ele- Figure 1. ORTEP-3 representations[13] of the molecular structures
of (a) [Fe(ox-phS)2] (complex A) and (b) [{Fe(ox-phS)}2(µ-S)2]
(complex B).
mental analyses of products 1–3 confirmed their composi-
tion as listed in Table 1. Table S1 (Supporting Information)
contains the crystallographic data for complex C.
Both 1 and 2 are air sensitive, although 1 is much less
reactive. In fact, Mössbauer spectra recorded after exposure
of 1 to air in the solid state for 3 d were very similar to
those obtained when the sample was kept under an atmo-
sphere of N2 (Table S2, Supporting Information).
The crystal packing in 1 consists of layers of the dimer
(perpendicular to the c axis) between layers of the mononu-
clear compound (Figure S1, Supporting Information). The
tighter molecular packing in the unit cell of 1 as compared
to 2 (Figures S1 and S2, Supporting Information), together
with the presence of iron(III) in the binuclear compound,
may explain the relative stabilities. Air sensitivity is metal
based, as the zinc(II) complex (C) is air stable. Relevant
molecular dimensions for A and B are presented in Table 2,
whereas data for products 2 and 3 are shown in Table S3
(Supporting Information).
Figure 2. ORTEP-3 representation[13] of the molecular structure of
[Zn(ox-phS)2] (complex C) with the atom numbering scheme.
In mononuclear A and C, the metal centres are coordi-
nated to nitrogen and sulfur donor atoms of two oxazolinyl-
thiophenolato ligands in a distorted tetrahedral geometry. (0.74 and 0.77 Å respectively, the latter in high-spin
Bond lengths (Table 2) involving the metal centre in A are state).[15] To the best of our knowledge, A is the first struc-
ca. 0.015–0.030 Å larger than the analogous dimensions in turally characterised iron(II) complex with an N,S-phenyl-
C (Table S3, Supporting Information), as expected from the oxazoline ligand reported to date. In C, on the other hand,
difference in the ionic radii of tetracoordinate ZnII and FeII bond lengths and angles are very close to those reported
2478
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
Table 2. Selected molecular dimensions for [Fe(ox-phS)2] (A) and consistent with a stabilising M–M interaction. This gives
[{Fe(ox-phS)2}2(µ-S)2] (B), with estimated standard deviations in
parentheses.
further support to EPR and Mössbauer spectroscopy re-
sults, discussed below. The large S4···S4Ј distance
Bond lengths about the metal centre [Å]
(3.4875 Å), on the other hand, rules out the possibility of
an S2 anion or of any significant bonding between the
2–
A
B
Fe1–N1
Fe1–N2
Fe1–S1
Fe1–S2
2.031(3)
2.054(2)
2.2820(9)
2.2807(9)
Fe2–N3
Fe2–S3
Fe2–S4
Fe2–S4Ј
2.022(2)
two sulfur atoms. The remaining distances and angles in the
central, planar 2Fe–2S ring are very much similar to those
reported earlier for [Fe2S2(LL)2]2– or [Fe2S2(L)4]2– systems,
where LL and L are bidentate(2–) and monodentate(1–) li-
gands, respectively (Table 3), confirming previous studies
where the {Fe2S2} core was found nearly invariant to the
nature of the terminal ligands.[17–19]
It is known from the literature that the reaction of FeCl3
with dithiols, hydrogen sulfide and sodium methoxide
in methanol affords the dimeric FeIII dianion
[{Fe(dithiolate)}2(µ-S)2]2–.[16] Complex B is a molecular an-
alogue of this anion, and its formation from FeCl2 probably
follows iron(II) oxidation by residual S8 [Equation (3)] in
the mixture of elemental sulfur, Li(ox) and (ox-phS)–. This
is because the reaction of sulfur with Li(ox) was carried out
in situ, and the thiolate [Li(ox-phS)] was not isolated before
the addition of FeCl2 [Scheme 2, Step 3; Equation (3)].
2.2611(9)
2.1808(8)
2.2014(8)
Bond angles about the metal centre [°]
A
B
N1–Fe1–N2
S2–Fe1–S1
N1–Fe1–S1
N2–Fe1–S2
N1–Fe1–S2
N2–Fe1–S1
115.90(10)
113.89(3)
95.76(8)
94.26(8)
117.79(8)
121.02(8)
N3–Fe2–S4Ј
S4–Fe2–S3
S4–Fe2–S4Ј
N3–Fe2–S3
S4Ј–Fe2–S3
N3–Fe2–S4
110.07(7)
110.75(3)
105.47(3)
93.99(7)
117.68(4)
119.33(7)
previously for the Pbca enantiomorph and different from
the P21 modification.[10]
In the binuclear iron(III) product (B), the molecule lies
on a crystallographic inversion centre located at the midpo-
int of the internuclear Fe2···Fe2Ј axis (Figure 1b; Figure S1,
Supporting Information). The structure consists of two
four-coordinate iron centres, each one bound to an oxazol-
inylthiophenolato ligand and two bridging sulfides. The ge-
1/4 S8 + 2FeII + 2(ox-phS)– Ǟ 2S2– + 2FeIII + ox-phS-Sph-ox (3)
Also, the addition of FeCl2 happened shortly after the
ometry about the metal centre is also distorted tetrahedral, mixture of Li(ox) and S8, when the formation of Li(ox-phS)
although less distorted than in A. The main feature of this had probably just started. This reaction is relatively slow,
structure is the planar {Fe2S2}2+ core, with unsymmetrical and precipitation of Li(ox-phS) is only observed after ca.
Fe2–S4 and Fe2–S4Ј distances of 2.1808(8) and 1 h at room temperature. A suitable yield of the lithium
2.2014(8) Å, respectively, which are close to the analogous thiophenyloxazolinate is only obtained after several more
dimensions in [{Fe(SCH2)2C6H4}2(µ-S)2]2– [2.185(2) and hours. Therefore, the conditions described in the Experi-
2.232(1) Å] and in [{Fe(SC6H4CH3)2}2(µ-S)2]2– [2.200(1) mental Section were probably just right to favour the redox
and 2.202(1) Å].[16] The M–Sthiolate and M–N bond lengths process that gave product 1.
in B [2.2611(9) and 2.022(2) Å, respectively] are very similar
According to the literature,[20] the formation of iron–sul-
to the corresponding distances in C and also consistent with fur clusters containing the {Fe2S2}2+ core from iron in low
those observed for A (Table 2), in spite of the different oxi- oxidation states and thiolate/disulfide mixtures is highly de-
dation states of the iron. The M–Ssulfide bond [av. pendent on the nature of the thiolate and the RS–/RS–SR
2.1911(8) Å] is shorter than the M–Sthiolate analogue ratio. In the present work, the presence of remaining ele-
[2.2611(9) Å], suggesting a significant degree of π donation mental sulfur in the reaction mixture apparently determined
from the electron-rich S2– ligands to the iron(III) centres. the path of the reaction, as after its consumption the
This could contribute to a softer character of the iron mother liquor gave only A. Accordingly, the combination
centres in B and explain why the metal in the formal oxi- of FeCl2 with two equivalents of isolated, purified Li(ox-
dation states +III and +II give so similar bond lengths in phS) gave a high yield of A without trace amounts of B.
B and A, respectively (Table 2).
Bond dimensions in the 2-(oxazolinyl)thiophenolato li-
The Fe2···Fe2Ј distance in B is 2.6535 Å, which is one of gands are insensitive to the Lewis acidity of the metal ion,
the shortest of this type in {Fe2S2}2+ moieties (Table 3) and as clearly shown by data compiled in Table S4 (Supporting
Table 3. Selected molecular dimensions in iron–sulfur complexes containing the {Fe2S2}2+ core. All parameters involving S atoms refer
to the bridging sulfides in the central {Fe2S2}2+ motif.[a]
Complex/Parameter
Donor set
Fe–S (av.) [Å]
S···S [Å]
Fe···FeЈ [Å]
S–Fe–SЈ [°]
Fe–S–FeЈ [°]
Ref.
[{Fe(ox-phS)2}2(µ-S)2] (B)
[{Fe(SCH2)2C6H4}2(µ-S)2]2–
[{Fe(SC6H4CH3)2}2(µ-S)2]2–
[{Fe(1,2-biphenolate)}2(µ-S)2]2–
[{Fe(pirrolate)2}2(µ-S)2]2–
[{Fe(Fe2S2CO6)}2(µ-S)2]2–
[{FeCl2}2(µ-S)2]2–
NS3
S4
S4
O2S2
N2S2
S4
2.1911(8)
2.208(2)
2.201(1)
2.215
2.18
2.198(2)
2.200(1)
3.4875
3.498(3)
3.483(3)
3.512(2)
3.57(5)
NA[a]
2.6535
2.698(1)
2.691(1)
2.699(1)
2.677(1)
2.675(2)
2.716(1)
105.47(3)
104.74(5)
104.61(4)
104.9(1)
104.3(4)
104.96(7)
103.79(3)
74.53(3)
75.27(5)
75.39(4)
74.1(1)
75.7(2)
74.98(7)
76.21(3)
this work
[16]
[16]
[18]
[18]
[19]
[17]
Cl2S2
NA[a]
[a] NA = not available.
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2479

J. F. Soares et al.
FULL PAPER
Information). This also compares well with structural data
for metal–oxazoline complexes in higher oxidation states,
such as those formed from titanium(IV) and zirconium(IV)
starting materials.[21] In all these, bond lengths and angles
within the oxazolinyl moiety indicate that the –C=N– bond
remains localised after coordination (Table S4, Supporting
Information).
The Fe1–S1 and Fe1–S2 bond lengths in A [2.2820(9)
and 2.2807(8) Å, respectively] are very similar. The average
value [2.2814(8) Å] is close to that reported for [Fe2(S2-o-
xyl)3](Et4N)2·2MeCN (S2-o-xyl = o-xylene-α,αЈ-dithiolate;
2.2901(1) Å).[22] The average Fe–N bond length
[2.0425(2) Å] is similar to that observed for [FeI2(btmgp)],
where btmgp = 1,3-bis(N,N,NЈ,NЈ-tetramethylguanidino)-
propane [2.040(1) Å][23] and smaller than those reported for
complexes of iron(II) with typical σ-donor ligands such as
tmen,[12,24] which normally range from 2.150 to 2.280 Å.
The molecular dimensions determined for complex A,
when obtained pure in product 2, are very close to those
found in 1 (Table 2; Table S3, Supporting Information), al-
lowing only for slightly distinct packing pressures.
Figure 3. Representation of the molecular structures of co-crystal-
lised trans-[FeCl2(tmen)2] and (ox-phS-Sph-ox) (product 4) with
key atoms labelled. Thermal ellipsoids are at the 50% probability
level.
Synthesis and X-ray Diffraction Analysis of Product 4
The molecular structures of D and E (product 4) are
represented in Figure 3; relevant molecular dimensions
(bond lengths and angles) are summarised in Table 4. This
is the first time D and E are formed in the same reaction
mixture, following oxidation of the thiophenolate by a tran-
sition-metal ion. Other reports on the synthesis of bis-
[(oxazolinyl)phenyl]disulfides similar to D employ O2 or
aqueous K3[Fe(CN)6] as oxidants.[10,25,26] In the present
work, thiolate oxidation was accomplished under mild con-
ditions and the use of an inert atmosphere and tmen al-
lowed the isolation of both D and E. Their unexpected –
and yet reproducible – co-crystallisation gave a significant
contribution to the understanding of this [(oxazolinyl)thio-
phenolato]iron(III) system.
Table 4. Selected bond lengths [Å] and bond angles [°] for ox-phS-
Sph-ox (D) and trans-[FeCl2(tmen)2] (E) with estimated standard
deviations (e.s.d.s) in parentheses.
Complex D
S1···N1
S2···N2
S1–S2
S1–C11
S2–C12
O1–C5
O2–C18
N1–C5
N1–C3
2.793
2.826
N2–C18
N2–C20
1.261(10)
1.488(9)
119.3(5)
119.1(5)
122.1(6)
121.4(5)
105.0(2)
105.4(2)
87.57
2.055(2)
1.793(7)
1.787(7)
1.350(9)
1.363(9)
1.257(10)
1.486(9)
C6–C11–S1
C17–C12–S2
C10–C11–S1
C13–C12–S2
C11–S1–S2
C12–S2–S1
C11–S1–S2–C12
Bond dimensions in the disulfide are similar to those re-
ported earlier for similar compounds in the pure form, sug-
gesting that packing pressures derived from E are not sig-
nificantly high.[10,25,26] The molecule assumes a staggered
conformation (C11–S1–S2–C12 torsion angle 87.57°) to
prevent repulsion between the S–S bond and lone electron
pairs. The sp3 sulfur hybridisation is evidenced by C11–S1–
S2 and C12–S2–S1 angles of ca. 105° (Table 4). The S1···N1
and S2···N2 distances (av. 2.8095 Å) are much shorter than
the summation of S and N van der Waals radii (3.85 Å),
indicating a donor–acceptor interaction. In fact, according
to the literature, the sulfur receives electron density from
trans-[FeCl2tmen2] (E)
Fe–N3
Fe–N4
Fe–N5
Fe–N6
Fe–Cl1
Fe–Cl2
2.360(7)
2.363(6)
2.371(6)
2.358(7)
2.4179(19)
2.405(2)
N3–Fe–N4
N6–Fe–N5
N6–Fe–N3
N4–Fe–N5
Cl2–Fe–Cl1
N3–Fe–Cl1
79.5(3)
80.5(2)
176.7(2)
179.4(2)
178.63(8)
88.77(18)
Electron Paramagnetic Resonance (EPR) Spectroscopy
Products 1–4 are all EPR silent (X-band) in the solid
the nitrogen through a σ* molecular orbital of the S–S state, both at room temperature and at 77 K. This is ex-
bond.[25] This helps to determine the solid-state structure of pected for diamagnetic zinc(II) complex C, and also ob-
the compound. The N1···S1–S2 and N2···S2–S1 angles of served for iron(II) species such as A and D, as accounted
170.87 and 173.98°, respectively, indicate an almost linear for by a large spin-orbit coupling in the ground state. The
N···S–S···N arrangement, compatible with a four-centre, result is also compatible with a strong antiferromagnetic
six-electron system.[25] The molecular structure of complex interaction between paramagnetic iron(III) centres in binu-
E, in its turn, is very close to the earlier report made for clear B, as reported for other compounds bearing the
trans-[FeCl2(tmen)2] by our research group.[12]
{FeIII2S2}2+ moiety.[18]
2480
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
Mössbauer Spectroscopy
Table 5. Zero-field 57Fe Mössbauer parameters determined at 77 K
for products 1, 2 and 4. Reference: iron foil at 298 K. Errors are
Ͻ0.01 mms–1.
Two 57Fe quadrupole-split doublets were recorded for
the solid samples of 1, both at room temperature and at
Parameter
Product 1
Product 2 Product 4
77 K (Figure 4 and Table 5). Doublet (a) gives Mössbauer [mms–1]
Doublet a Doublet b
parameters typical of high-spin iron(II), whereas doublet
(b) is more difficult to attribute. A high-spin iron(III) as-
signment is acceptable; little temperature dependence of
quadrupole splitting (q.s.) is detected (Table S2, Supporting
Information). The observed values fall in the range given
by other {FeIII2S2} complexes. An example is the {Fe2S2}
anion with terminal ligands derived from 2-(2-mercap-
tophenyl)benzimidazole, isomer shift (i.s.) = 0.25 and q.s. =
0.89 mms–1.[27] Similarly, for {Fe2S2} bound to 2-hy-
droxybenzyl mercaptan, i.s. = 0.28, q.s. = 0.99 mms–1.[28]
The large q.s. values seem characteristic of high-spin
iron(III) coordinated to inflexible, tight chelate rings (as the
ones generated by ox-phS), as large splittings are also found
for [Fe2S2L2]2– [L = 2,2Ј-biphenolate (1.02 mms–1), L =
thiosalicylate (1.09 mms–1) and L = dimethyl-2,2Ј-methyl-
enebis(benzimidazolate) (0.92 mms–1)] in comparison to
q.s. values of 0.32–0.49 mms–1 for coordination by mono-
dentate SPh, OPh, pyrrolate or the more flexible chelate S2-
o-xylyl. Therefore, a q.s. value of 1.07 mms–1 for product 1
is compatible with high-spin iron(III) centres, and results
from Table 5 altogether confirm the presence of a 1:1 pro-
portion of high-spin iron(II) and iron(III) in the crystals of
1.
Isomer shift
Quadrupole splitting
Γ[a]
0.71
3.20
0.13
46
0.28
1.07
0.13
54
0.73
3.11
0.13
100
1.16
2.99
0.13
100
Proportion [%]
[a] Half-width at half-height.
The isomer shifts and quadrupole splittings of the three
iron(II) complexes (1A, 2 and 4D) show the expected trends.
The lower coordination number in the tetrahedral centres
(1A and 2) gives the lowest isomer shift because the s char-
acter of the hybrid orbitals on the four-coordinate iron is
greater. The low symmetry around iron reduces delocalis-
ation of the d electrons, consistent with the large quadru-
pole splittings in both the four- and six-coordinate environ-
ments.
Electronic Spectroscopy and Time-Dependent Density
Functional Theory (TDDFT) Calculations for Complex A
Geometry optimisation and time-dependent density
functional theory (TDDFT) calculations on the electronic
structure of A employed the UB3LYP hybrid functional[29]
with LANL2DZ basis set,[30] as implemented in the
Gaussian03 suite.[31] The working space included 160 ex-
cited states. Tables S5 and S6 (Supporting Information)
contain, respectively, the description of the frontier orbitals
and electronic transitions predicted for A. Figure S3 (Sup-
porting Information) presents the atom numbering scheme
employed in the tables.
Frontier orbitals calculated for A are predominantly li-
gand based. Main contributions to the HOMO, HOMO-1
and HOMO-2 come from sulfur lone pairs and from ben-
zene π orbitals. The LUMO, LUMO+2 and LUMO+3 con-
tain a fair contribution from the iron and participate in
LMCT transitions (Table S6, Supporting Information). Fig-
ure 5 shows a comparison between the experimental and
calculated electronic spectra of A.
The experimental spectra (200–800 nm range) registered
for products 1, 2 and 3 in acetonitrile solutions are pre-
sented in Figure S4 (Supporting Information). Absorption
bands are significantly strong, with molar absorptivity (ε)
values ranging from 11600 to 480000 –1 cm–1. This is com-
patible with the ligand-to-ligand charge transfer (LLCT)
and ligand-to-metal charge transfer (LMCT) nature of the
transitions calculated by TDDFT.
In the case of C (product 3; Figure S4c, Supporting In-
formation), three intense and well-defined absorption bands
were detected with
ε
values between 24195 and
122120 –1 cm–1. These were assigned to internal pπǞpπ*
transitions (LLCT) of the oxazolinylthiophenolato ligand.
The LLCT transitions occur from molecular orbitals based
Figure 4. Zero-field 57Fe Mössbauer spectra recorded at 77 K for
products 1, 2 and 4. Reference: iron foil at 298 K.
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2481

J. F. Soares et al.
FULL PAPER
Figure 5. Experimental and calculated (TDDFT, inset) electronic
spectra of [Fe(ox-phS)2] (complex A). In the inset, only the strong-
est absorptions are identified. An extended list of calculated transi-
tions is presented in Table S6 (Supporting Information).
on S and N to π* molecular orbitals of the benzene and
oxazoline rings. Finally, the comparison between the elec-
tronic spectra of A (Figure 5; Table S6, Supporting Infor-
mation) and C (Figure S4c, Supporting Information) con-
firms the general assignment of the most intense absorp-
tions of A, which occur in the 200–250 nm range and
around 320 nm, to LLCT transitions, whereas LMCT oc-
curs above 250 nm. This conclusion agrees very well with
data reported for the tris(chelate) [Fe(phox)3].[8]
Cyclic Voltammetry
Studies on the electrochemical behaviour of H(ox) and
complexes [Fe(ox-phS)2] (A) and [Zn(ox-phS)2] (C) were
carried out at room temperature in MeCN solutions con-
taining 0.1  of (tba)[PF6]. Voltammograms were started
from 0.00 V, with scan rates of 50, 100 and 200 mVs–1 and
Figure 6. Cyclic voltammograms registered at 100 mVs–1 for (a)
[Zn(ox-phS)2] (C) and (b) [Fe(ox-phS)2] (A) in solutions of
a voltage window of –2.0 to +2.0 V. Representative voltam- (tba)[PF6] (0.1 ) in MeCN. Potentials are referenced by E1/2(Fc+/
Fc).
mograms are presented in Figure 6. No redox processes
were observed for the starting material 4,4-dimethyl-2-
phenyloxazoline [H(ox)], suggesting the absence of electro-
chemical processes relative to the oxazoline ring in the ap- electroactivity of the ligand.[32] Indeed, the oxidation of (ox-
plied potential range.
phS)– to produce the respective disulfide (ox-phS-Sph-ox)
The analysis of [Zn(ox-phS)2] (C) by cyclic voltammetry has been shown to give product 4 in the present work, con-
revealed two irreversible oxidation waves at +0.87 and firming the redox non-innocence of the sulfur-containing
+1.23 V (vs. Fc+/Fc). These were assigned to ligand-based proligand.
electrochemical processes, as no metal-centred redox reac-
For A, besides the oxidation waves at +0.18 and +1.02 V
tion would be expected for zinc(II) under the employed con- and the reduction at –1.28 V (which depends on the oxi-
ditions. The irreversible reduction wave observed at –1.29 V, dations, as in C), a pair of waves with E1/2 = –2.15 V is
which is not recorded in the first cycle (Figure 6a), probably also observed. This feature does not depend on scan rates
corresponds to a reduction process dependent on the oxi- between 50 and 200 mVs–1, thus indicating electrochemical
dations observed above +0.8 V. These three redox waves reversibility, and presents a ∆Ep value half as large as that
probably arise from the presence of sulfur as a donor atom measured for the Fc+/Fc pair under the same experimental
in the ligand, as no electrochemical activity was detected conditions. The finding is compatible with a two-electron
for the H(ox) starting material before the incorporation of process, which was tentatively assigned to the formation of
sulfur. Formation of RS· radicals and disulfide bridges (R– a FeIII dimer, according to Equation (4) and Scheme 3.
S–S–R) – following thiolate oxidation – could explain the Interestingly, the process is observed even in anodic scans
2482
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
from –2.50 to –2.10 V (Figure S5, Supporting Information). reversibility of these two processes could be explained by
This suggests a high thermodynamic tendency of the system the formation of intermolecular bonds involving the radi-
to produce the dimeric product.
cals.
A number of literature reports give support to the model
proposed in Scheme 3. Wieghardt and co-workers,[33] as an
example, carried out a detailed electrochemical and spectro-
electrochemical study on dimeric complexes of iron(III)
[N(nBu)4][Fe2L4], where L = 1,2-benzenedithiolate, and
(4)
[N(nBu)4][Fe2LBu4], where LBu
= 3,5-di-tert-butyl-1,2-
benzenedithiolate, obtained from the oxidation of the mo-
nonuclear [N(nBu)4][FeL2] and [N(nBu)4][FeLBu2], respec-
tively, in weakly coordinating solvents. The voltammograms
for the iron(III) binuclear complexes present two reversible
oxidation waves, assigned to ligand-based processes, and
one reversible reduction wave, which was attributed to a
two-electron, metal-centred process. The latter presents
E1/2 = –1.36 V and –1.45 V (vs. Fc+/Fc) for [Fe2L4]2– and
[Fe2LBu4]2–, respectively. According to the authors, the
metal-based reduction is a two-electron transformation
leading to the formation of a mononuclear iron(II) complex
from the binuclear iron(III) starting compound. Equa-
tion (4) illustrates an analogous process.
A similar study describes the preparation of a dimeric
FeIII complex with N,S-donor ligands, [Fe2(µ-HNS-Ph)2-
(H2NS-Ph)2] (HNS-Ph
= 4,6-di-tert-butyl-2-aminothio-
phenolato), from the oxidation of the related iron(II) mono-
nuclear compound.[34] Ligand-centred processes registered
2–
for [Fe2L4]2– and [Fe2LBu
]
have been assigned to the for-
4
mation of RS· radicals from the benzenedithiolate
groups.[33] Indeed, the electroactivity (redox non-innocence)
of S-donor ligands, such as thiolates, has been intensively
investigated in other systems.[35] The comparison between
the electrochemical potentials registered for the ligand-
centred processes in the literature, for example, [Fe2L4]2–
2–
(E1/2 = –0.41 and –0.10 V) and [Fe2LBu
]
(E1/2 = –0.63
4
and –0.35 V), with those observed in the cyclic voltammog-
rams of A (E = +0.18 and +1.02 V), all vs. Fc+/Fc, indicates
that the radicals generated by L = 1,2-benzenedithiolate
and LBu = 3,5-di-tert-butyl-1,2-benzenedithiolate are more
stable than those given by oxidation of 4,4-dimethyloxaz-
olinyl-2-thiophenolate (ox-phS–) in A. Also, whereas the di-
meric FeIII products described in the literature are dianions,
the dimer possibly formed from A is a 2+ cation, which is
probably much harder to oxidise than its anionic counter-
parts. This agrees with Scheme 3.
Scheme 3. Electrochemical behaviour of complex A under the ex-
perimental conditions employed in this work.
In the voltammogram shown in Figure S5a (Supporting
Information), the reduction wave at ca. –2.20 V is not ob-
served, indicating dependence on the oxidation at –2.10 V.
Accordingly, the voltammogram in Figure S5b (Supporting
Conclusions
Interestingly, in spite of the large number of reports on
Information), recorded from –2.50 to –2.10 V, shows both the preparation and catalytic performance of oxazoline
the oxidation and the reduction processes; this becomes complexes of a variety of transition-metal ions, the chemi-
clear in Figure S5c,d (Supporting Information).
cal literature revealed a relevant gap when referring to iron
Scheme 3 proposes a model for the electrochemical be- compounds with chalcogen-containing derivatives of these
haviour of complex A. After the formation of the dimeric ligands. Results presented here, generated in a simple model
FeIII complex (–2.12 V), two successive oxidation processes system, intend to add new data to this discussion, particu-
(+0.18 and +1.02 V) could lead to the formation of RS· larly in light of the widespread biological distribution of
radicals in the terminal ligands. The lack of electrochemical iron(II)/(III) and of the redox-protecting in vivo activity of
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2483

J. F. Soares et al.
FULL PAPER
Reaction with Anhydrous FeCl2: A stirred solution of 4,4-dimethyl-
2-phenyloxazoline (2.5 g, 14.3 mmol) in hexane (40 mL) was
treated dropwise with a solution of nBuLi (1.6  in hexane, 9.1 mL,
14.3 mmol). Stirring was continued for 1 h at room temperature,
giving a white precipitate, Li(ox), in an orange suspension. After
decantation, the supernatant was removed by syringe, and the white
solid was washed with cold hexane (15 mL), dried under vacuum
(1.9 g, 10.4 mmol, 73% yield) and re-dissolved in diethyl ether
(100 mL); elemental sulfur (0.33 g, 1.3 mmol S8, 10.4 mmol “S”)
was added. After stirring for 1 h at room temperature, the light
yellow reaction mixture was cooled down to 0 °C. A suspension of
anhydrous FeCl2 (0.66 g, 5.2 mmol) in toluene (35 mL) was added.
Stirring was continued for 1 h at 0 °C and then for 16 h at room
temperature; the brown-reddish solution was then filtered to re-
move lithium chloride. Dark brown crystals (0.63 g) suitable for
single-crystal X-ray analysis (product 1) were formed from this
mixture after 2 d at room temperature. They were filtered off and
dried under vacuum. Data for 1: C66H72Fe4N6O6S8 (1525.18, co-
crystallised complexes A and B): calcd. C 51.98, H 4.76, N 5.51,
Fe 14.65; found C 51.92, H 4.76, N 5.51, Fe 14.05. IR (for 1, A +
sulfur- and selenium-containing biochemicals, including en-
zymes,[11] against reactive oxygen and nitrogen species. Iron
chelation by these metabolites, or redox processes as shown
here, are highly likely to change their performance, and the
consequences still deserve close examination.
Experimental Section
General Procedures: All operations were carried out under an atmo-
sphere of N2 with the use of standard Schlenk and glove box (Vac-
uum Atmospheres Inc.) techniques. Solvents (Merck, J. T. Baker)
were dried by standard procedures[36] and distilled under an atmo-
sphere of N2 prior to use. The commercial products n-butyllithium
(1.6  in hexane), anhydrous iron(II) and iron(III) chlorides, 2-
amino-2-methylpropan-1-ol (Aldrich), potassium carbonate, ethyl-
ene glycol, glycerol and benzonitrile (Acros Organics) were used
without further purification. Zinc(II) chloride (Merck) was dried
by heating to ca. 300 °C under vacuum. Elemental sulfur (Reagen)
was recrystallised from hot toluene. N,N,NЈ,NЈ-tetramethylethane-
1,2-diamine (Amersham) was dried with KOH pellets for 24 h and
then distilled under an atmosphere of N2. 4,4-Dimethyl-2-phenyl-
oxazoline was prepared by reaction of benzonitrile with 2-amino-
2-methylpropan-1-ol in the presence of ethylene glycol, glycerol and
K2CO3.[37] The product was extracted with hexane, dried with an-
hydrous Na2SO4 and BaO, and purified by distillation. The global
yield was typically around 65%. Microanalyses were carried out by
Medac Laboratories Ltd., Egham, Surrey, UK. Iron analyses were
performed by a colorimetric method.[38] Zinc content was deter-
mined by differential pulse anodic stripping voltammetry
(DPASV). Measurements were carried out with an EG&G Prince-
ton Applied Research 394 Electrochemical Trace Analyser coupled
with an EG&G PAR 303A static mercury drop electrode. Instru-
ment conditions were described elsewhere.[39] IR data (Nujol mulls)
were recorded with Bomem Hartmann Braun (MB series) or BIO-
RAD FTS3500GX equipment in the range 400–4000 cm–1. Mulls
were spread on KBr plates. Zero-field 57Fe Mössbauer data were
recorded at 77 K and/or 300 K by using an ES-Technology MS105
spectrometer with a 57Co source in a rhodium matrix. Spectra were
referenced against iron foil at 298 K. Parameters were obtained by
fitting the data to Lorentzian bands. EPR data (X-band, 9.5 GHz)
were recorded with a Bruker ESP-300E instrument from solid sam-
ples or toluene solutions at room temperature and 77 K. Electronic
spectra were obtained from toluene solutions at room temperature
with a Hewlett–Packard HP 8452A diode array spectrophotometer
(220–820 nm range). Magnetic susceptibility measurements by a
modified Gouy method were carried out in the solid state or from
toluene solutions at room temperature by using a MKII magnetic
susceptibility balance from Johnson-Matthey. Corrections for the
diamagnetism of the ligands were applied (Pascal constants).[40] Cy-
clic voltammetry experiments were controlled by using EG&G
PAR 273 or Microquimica MQPG-01 potentiostats and were car-
ried out on one-compartment glass cells with vitreous carbon
(working), platinum (counter) and silver wire (pseudoreference)
electrodes. The potential of the reference electrode was determined
by the use of the ferrocenium/ferrocene couple (Fc+/Fc), which oc-
curs at +0.38 V in acetonitrile against the SCE reference elec-
trode.[41] All redox potentials in this work are quoted vs. the Ef(Fc+/
Fc). Tetra-n-butylammonium hexafluorophosphate, (tba)[PF6] (Al-
drich), was employed as supporting electrolyte.
B, Nujol mull): ν = 1603 (s) and 1592 [s, ν(C=N], 1366 [s, ν(C–N)],
˜
1052 [s, δ(C–O–C)], 735 [m, ν(C–S)] cm–1. UV/Vis (CH3CN): λ (ε,

–1 cm–1) = 210 (430000), 236 (480000), 274 (240000), 392 (62000),
332 (118000), 522 (35250) and 590 (29800) nm. Molar absorptivity
(ε) values for 1 were based on the concentration of A (2.24ϫ10–6 )
in the MeCN solution employed for the determination. After isola-
tion of this first batch of 1, the filtrate was left at room temperature
for 4 d, giving a mixture of small, dark brown prisms (1) and bright
red hexagonal crystals (product 2). They were filtered off and dried.
One day later, a pure, homogeneous batch of red crystals (2, 0.83 g)
was isolated by filtration from the mother liquor. Total yield (1 +
2) based on iron content: 1.46 g, 42%. Data for 2: C22H24FeN2O2S2
(468.40): calcd. C 56.41, H 5.16, N 5.98, Fe 11.92; found C 56.39,
H 5.16, N 6.09, Fe 11.43. IR (for 2, complex A, Nujol mull): ν =
˜
1595 [s, ν(C=N], 1369 [s, ν(C–N)], 1053 [s, δ(C–O–C)], 740 [m, ν(C–
S)] cm–1.
Synthesis of Pure [Fe(ox-phS)2] (Product 2, Complex A) by Reaction
of Anhydrous FeCl2 with Isolated Li(ox-phS): The preparation of
Li(ox-phS) was started as described above, from Li(ox) and elemen-
tal sulfur in diethyl ether. Soon after addition of S8, the reaction
mixture became brown-yellowish, changing to a bright yellow solu-
tion after ca. 15 min. Solid Li(ox-phS) started to precipitate after
1 h; the reaction mixture was then left to stir overnight at room
temperature. The solid product (typically 50–60% yield) was iso-
lated by filtration, dried under vacuum and stored under an atmo-
sphere of N2. Pure A was obtained from a 2:1 mixture of isolated
Li(ox-phS) (0.60 g, 2.8 mmol) in diethyl ether (50 mL) and FeCl2
(0.18 g, 1.4 mmol) in toluene (15 mL). After 24 h at room tempera-
ture, the dark red reaction mixture was filtered to remove LiCl;
hexane (20 mL) was then carefully added as a layer. Bright red
hexagonal crystals were isolated by filtration after 2 d at room tem-
perature. Yield: 0.58 g, 88.4%. The identity of A was confirmed
by FTIR and Mössbauer spectroscopy and by single-crystal X-ray
diffraction analysis. µeff = 5.41 µB (room temperature). UV/Vis
(CH3CN): λ (ε, –1 cm–1) = 212 (363000), 226 (390000), 264 sh.,
336 (79200), 392 (48500), 496 (18250), 586 (11600) nm.
Synthesis of [Zn(ox-phS)2] (Product 3, Complex C): The procedure
was analogous to that followed for pure A (product 2), employing
anhydrous ZnCl2 (0.78 g, 5.7 mmol) and Li(ox-phS) (2.4 g,
11.4 mmol). This gave colourless crystals of [Zn(ox-phS)2] (1.9 g,
68.4% yield). C22H24N2O2S2Zn (477.92): calcd. C 55.29, H 5.06, N
5.86, Zn 13.68; found C 55.48, H 5.03, N 5.87, Zn 13.47. IR (Nujol
Preparation of Co-Crystallised 2[Fe(ox-phS)2]·[{Fe(ox-phS)}2(µ-S)2]
(Product 1, Complexes A and B) and of Pure [Fe(ox-phS)2] (Product
2, Complex A) by the In situ Preparation of Li(ox-phS) Followed by
mull): ν = 1595 [s, ν(C=N)], 1364 [s, ν(C–N)], 1053 [s, δ(C–O–C)],
˜
2484
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
738 [s, ν(C–S)] cm–1. UV/Vis (CH3CN): λ (ε, –1 cm–1) = 250
(122120), 276 (48970), 364 (24195) nm.
Preparation of Co-Crystallised [FeCl2(tmen)2]·(ox-phS-Sph-ox)
(Product 4) and [Fe(ox-phS)2] (Product 2, Complex A) by Reaction
of Anhydrous FeCl3 with Isolated Li(ox-phS) and tmen (1:2:2): A
suspension of anhydrous FeCl3 (0.75 g, 4.6 mmol) in thf (30 mL)
was slowly added at room temperature to a light yellow solution
of Li(ox-phS) (2.1 g, 9.7 mmol) in thf (30 mL), and the mixture was
stirred for 1.5 h. To the deep red-brownish reaction mixture was
added tmen (1.4 mL, 1.1 g, 9.3 mmol). No precipitation was ob-
served. The solution was then left to stir at room temperature for
2 d, after which it was evaporated under vacuum to ca. 25 mL,
layered with hexane (25 mL) and cooled to –20 °C to give red crys-
tals of [Fe(ox-phS)2] (complex A, 0.56 g, 52% yield). Several subse-
quent steps of solvent evaporation, addition of hexane and cooling
at –20 °C also gave product 4 (0.82 g, 46%).
Preparation of Co-Crystallised [FeCl2(tmen)2]·(ox-phS-Sph-ox)
(Product 4) and “FeCl2(tmen)” (Product 5) by Reaction of Anhy-
drous FeCl3 with Isolated Li(ox-phS) and tmen (1:1:2): A very light
yellow solution of Li(ox-phS) (1.5 g, 7.0 mmol) in thf (25 mL) was
slowly added, at room temperature, to a stirred brown-orange sus-
pension of FeCl3 (1.1 g, 7.0 mmol) in thf (25 mL). The reaction was
immediate, giving a deep brown solution that was left to stir at
room temperature for 2 h. After this period, tmen (2.1 mL, 1.6 g,
13.8 mmol) was added dropwise to the reaction mixture, leading to
the precipitation of a light brown solid. The suspension was filtered
after 16 h of stirring at room temperature, and the solid was dried
under vacuum (0.37 g, LiCl). The brown-yellowish filtrate was then
evaporated under vacuum to ca. 30 mL and cooled to –20 °C for
3 weeks. This gave 4 as a colourless crystalline solid (0.66 g), which
was isolated by filtration. Additional batches of these crystals were
obtained after consecutive steps of evaporation of the filtrate and
cooling to –20 °C. Total yield of 4: 1.3 g, 47.5%. The identity of the
product as trans-[FeCl2(tmen)2]·(ox-phS-Sph-ox) was confirmed by
single-crystal X-ray diffraction analysis. Layering of the mother
liquor with hexane also gave an off-white microcrystalline powder
(5; 0.43 g, 50.6%), which was filtered off and dried under vacuum.
Single-Crystal X-ray Diffraction Analyses: Details on data collec-
tions and structure refinements are presented in Table 6 and the
Supporting Information. Data for 1–3 were collected at 173 K with
a Nonius Kappa CCD area detector diffractometer at the Depart-
ment of Chemistry, University of Sussex, Brighton, UK. An ab-
sorption correction (multiscan) was applied in all cases. A suitable
dark brown rectangular prism of 1 (0.15ϫ0.15ϫ0.05 mm3), a red
hexagonal prism of 2 (0.3ϫ0.25ϫ0.25 mm3) and a colourless rec-
tangular prism of 3 (0.25ϫ0.20ϫ0.20 mm3) were mounted on
glass fibres and cooled to 173(2) K. Cell dimensions were based on
all 5011, 3140 and 3387 observed reflections (I Ͼ 2σI) for 1, 2 and
3, respectively. Structures were solved by direct methods
(WinGX)[42] and refined by full-matrix least-squares on F2 with
SHELXL-97.[14] Drawings were made with ORTEP-3 for Win-
dows.[13] All non-hydrogen atoms were refined anisotropically, and
H atoms were included in riding mode. Product 2 (complex A) had
its crystal and molecular structures solved thrice, after being (i) co-
crystallised with product 1, (ii) obtained as the only reaction prod-
uct and also (iii) prepared together with product 4. X-ray data for
The FTIR spectrum of
5 is superimposable to that of
[{FeCl(tmen)}2(µ-Cl)2] prepared according to our previous re-
port.[12] Data for 4: C34H56Cl2FeN6O2S2 (771.72): calcd. C 52.90,
H 7.33, N 10.89, Fe 7.24; found C 51.59, H 7.16, N 10.60, Fe 7.40.
IR (for 4, D + E, Nujol mull): ν = 1649 [vs, ν(C=N)thiophenylox-
˜
azoline], 1364 [m, ν(C–N)thiophenyloxazoline], 1076 [m, δ(C–O–
C)thiophenyloxazoline], 1032 (vs), 1014 [sh., ν(C–N)diamine], 741
[m, ν(C–S)thiophenyloxazoline] cm–1. IR (for 5, Nujol mull): ν =
˜
1008 [s, ν(C–N)], 1025 [vs, ν(C–N)], 442 [w, ν(Fe–N)], 463 [w, ν(Fe–
N)], 486 [w, ν(Fe–N)] cm–1.
Table 6. Crystal and refinement data for 2[Fe(ox-phS)2]·[{Fe(ox-phS)}2(µ-S)2] (1), [Fe(ox-phS)2] (2) and [FeCl2(tmen)2]·(ox-phS-Sph-ox)
(4).
1 (A + B)
2 (A)
4 (D + E)
Formula
C66H72N6O6S8Fe4
1525.18
173(2)
monoclinic
P21/n
11.0426(2)
10.7840(2)
29.4789(5)
94.347(1)
3500.35(11)
1580
C22H24N2O2S2Fe
468.40
173(2)
monoclinic
P21/n
9.3542(4)
11.2933(4)
21.4225(5)
96.666(2)
2247.77(14)
976
C34H56Cl2FeN6O2S2
771.72
293(2)
monoclinic
P21/n
9.0481(3)
16.5858(5)
26.0979(9)
90.291(2)
3916.5(2)
1640
Fw [gmol–1]
T [K]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
β [°]
V [Å3]
F(000)
Z
2
1.45
1.11
4
1.38
0.88
4
ρcalcd. [mgm–3]
1.309
0.666
µ [mm–1]
θ range [°]
Measured data
Independent data
Data I Ͼ 2σI
Parameters
Completeness to θmax [%]
S on F2[a]
3.73–25.01
28641
6111
5011
406
98.9
1.055
0.046
0.116
3.56–26.02
18125
4405
3140
262
99.6
1.016
0.042
0.079
1.99–26.41
37996
7938
7138
424
98.6
1.157
0.1034
0.2448
R1 (I Ͼ 2σI)[a]
wR2 (I Ͼ 2σI)[a]
R1 (all data)[a]
wR2 (all data)[a]
0.059
0.125
0.076
0.091
0.1093
0.2473
[a] As defined by the SHELXL-97 program.[14]
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2485

J. F. Soares et al.
FULL PAPER
[8] H. Kooijman, A. L. Spek, M. Hoogenraad, E. Bouwman, J. G.
Haasnoot, J. Reedijk, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun. 2002, 58, m390–m392.
[9] a) G. Mugesh, H. B. Singh, R. J. Butcher, J. Organomet. Chem.
1999, 577, 243–248; b) G. Mugesh, H. B. Singh, R. P. Patel,
R. J. Butcher, Inorg. Chem. 1998, 37, 2663–2669.
product 4 were collected with a Bruker Kappa X8 APEX II CCD
diffractometer by using Mo-Kα radiation (0.71073 Å) at the De-
partment of Chemistry, Universidade Federal de Santa Maria, RS,
Brazil. A colourless rectangular prism of 4 (0.47ϫ0.31ϫ0.25 mm3)
was employed for data collection at 293(2) K. X-PREP was used
to perform the Gaussian absorption correction based on indexed
crystal faces.[43] Cell dimensions were based on all 7938 observed
reflections (I Ͼ 2σI, Table 6). Structure resolution (by direct meth-
ods) and refinement (by full-matrix least-squares on F2) were car-
ried out with SHELXS-97[44] and SHELXL-97 package[14] respec-
tively. Refinements employed anisotropic displacement parameters
for all non-hydrogen atoms, whereas H atoms were included in cal-
culated positions. CCDC-681814 (for 1), -681815 (for 2), -681816
(for 3), -725548 (for 4) and -725547 (for 2 crystallised from the
same reaction mixture that gave 4) contain the supplementary crys-
tallographic data for this paper. These data can be obtained free
of charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[10] G. Mugesh, H. B. Singh, R. J. Butcher, Eur. J. Inorg. Chem.
1999, 1229–1236.
[11] a) E. E. Battin, J. L. Brumaghim, Cell Biochem. Biophys. 2009,
55, 1–23; b) G. Mugesh, A. Panda, H. B. Singh, N. S. Punekar,
R. J. Butcher, J. Am. Chem. Soc. 2001, 123, 839–850; c) B. K.
Sarma, G. Mugesh, Org. Biomol. Chem. 2008, 6, 965–974.
[12] S. C. Davies, D. L. Hughes, G. J. Leigh, J. R. Sanders, J. S.
de Souza, J. Chem. Soc., Dalton Trans. 1997, 1981–1988.
[13] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
[14] G. M. Sheldrich, SHELXL-97: Program for Crystal Structure
Refinement, University of Göttingen, Germany, 1997.
[15]
J. E. Huheey, E. A. Keiter, R. L. Keiter, Inorganic Chemistry:
Principles of Structure and Reactivity, HarperCollins, New
York, 1993, pp. 114–117.
[16]
[17]
[18]
J. J. Mayerle, S. E. Denmark, B. V. DePamphilis, J. A. Ibers,
R. H. Holm, J. Am. Chem. Soc. 1975, 97, 1032–1045.
M. A. Bobrik, K. O. Hodgson, R. H. Holm, Inorg. Chem.
1977, 16, 1851–1858.
D. Coucouvanis, A. Salifoglou, M. G. Kanatzidis, A. Simo-
poulos, V. Papaefthymiou, J. Am. Chem. Soc. 1984, 106, 6081–
6082.
F. Calderoni, F. Demartin, M. C. Iapalucci, F. Laschi, G. Lon-
goni, P. Zanello, Inorg. Chem. 1996, 35, 898–905.
G. L. Lilley, E. Sinn, B. A. Averill, Inorg. Chem. 1986, 25,
1073–1075.
P. G. Cozzi, C. Floriani, A. Chiesi-Villa, C. Rizzoli, Inorg.
Chem. 1995, 34, 2921–2930.
Supporting Information (see footnote on the first page of this arti-
cle): Crystal and structure refinement data for 3; Mössbauer pa-
rameters for 1; selected molecular dimensions for 2 and 3; selected
bond lengths in the ligands for products 1–3; frontier molecular
orbitals calculated for complex A; electronic transitions calculated
for complex A; crystal structure of 1 and 2; MOLDEN representa-
tion of the molecular structure of complex A; electronic spectra for
1–3; cyclic voltammograms for complex A.
[19]
[20]
[21]
Acknowledgments
[22]
[23]
K. S. Hagen, R. H. Holm, Inorg. Chem. 1984, 23, 418–427.
S. Pohl, M. Harmjanz, J. Schneider, W. Saak, G. Henkel, J.
Chem. Soc., Dalton Trans. 2000, 3473–3479.
D. J. Evans, P. B. Hitchcock, G. J. Leigh, B. K. Nicholson,
A. C. Niedwieski, F. S. Nunes, J. F. Soares, Inorg. Chim. Acta
2001, 319, 147–158.
A. Esparza-Ruiz, A. Peña-Hueso, J. Hernández-Díaz, A. Flo-
res-Parra, R. Contreras, Cryst. Growth Des. 2007, 7, 2031–
2040.
S. Kumar, K. Kandasamy, H. B. Singh, R. J. Butcher, New J.
Chem. 2004, 28, 640–645.
We are grateful to Mr. José Luiz B. dos Santos and Mrs. Glaci
Alves for help with the preparation of products 1 and 2, to Ms.
Vanessa E. dos Anjos for zinc analyses and to Dr. Geraldo R. Frie-
dermann, Prof. Sueli Maria Drechsel and Prof. Carlos J. da Cunha
(Departamento de Química, UFPR) for helpful suggestions. This
work has been supported by the Conselho Nacional de Desenvolvi-
mento Científico e Tecnológico (CNPq 309051/2006-1 and 307535/
2009-6), the Coordenação de Aperfeiçoamento de Pessoal de Nível
Superior (CAPES) and the Universidade Federal do Paraná
(UFPR). R. C. R. B. and R. A. G. thank CAPES, and D. B.,
E. S. S., F. S. and J. F. S. thank CNPq for scholarships. D. J. E.
thanks the Biotechnology and Biological Sciences Research Coun-
cil (BBSRC) for financial support.
[24]
[25]
[26]
[27]
P. Beardwood, J. F. Gibson, J. Chem. Soc., Dalton Trans. 1992,
2457–2466.
[28]
[29]
P. Beardwood, J. F. Gibson, Polyhedron 1988, 7, 1911–1918.
a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee,
W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789; c) S. H.
Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200–1211;
d) P. J. Stephens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch,
J. Phys. Chem. 1994, 98, 11623–11627.
[1] a) C. A. Caputo, N. D. Jones, Dalton Trans. 2007, 4627–4640;
b) R. Rasappan, D. Laventine, O. Reiser, Coord. Chem. Rev.
2008, 252, 702–714; c) G. Desimoni, G. Faita, K. A. Jørgensen,
Chem. Rev. 2006, 106, 3561–3651.
[2] a) R. Ferro, S. Milione, V. Bertolasi, C. Capacchione, A.
Grassi, Macromolecules 2007, 40, 8544–8546; b) A. M. Tond-
reau, J. M. Darmon, B. M. Wile, S. K. Floyd, E. Lobkovsky,
P. J. Chirik, Organometallics 2009, 28, 3928–3940; c) M. Zhang,
R. Gao, X. Hao, W.-H. Sun, J. Organomet. Chem. 2008, 693,
3867–3877.
[3] R. Kikkeri, H. Traboulsi, N. Humbert, E. Gumienna-Kon-
tecka, R. Arad-Yellin, G. Melman, M. Elhabiri, A.-M. Al-
brecht-Gary, A. Shanzer, Inorg. Chem. 2007, 46, 2485–2497.
[4] M. D. Godbole, M. P. Puig, S. Tanase, H. Kooijman, A. L.
Spek, E. Bouwman, Inorg. Chim. Acta 2007, 360, 1954–1960.
[5] W. M. Wuest, E. S. Sattely, C. T. Walsh, J. Am. Chem. Soc.
2009, 131, 5056–5057.
[6] G. C. Hargaden, P. J. Guiry, Chem. Rev. 2009, 109, 2505–2550.
[7] M. Gómez, G. Muller, M. Rocamora, Coord. Chem. Rev. 1999,
193–195, 769–835.
[30]
[31]
P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270–283.
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R.
Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R.
Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morok-
uma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzew-
ski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V.
Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B.
Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L.
Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A.
Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
2486
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 2476–2487

Redox Processes in Oxazolinylthiophenolato Complexes of Iron(II)/(III)
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Re-
vision D.01, Gaussian, Inc., Wallingford, CT, 2008.
J. S. Pap, F. L. Benedito, E. Bothe, E. Bill, S. DeBeer George,
T. Weyhermuller, K. Wieghardt, Inorg. Chem. 2007, 46, 4187–
4196.
[37] a) J. A. Frump, Chem. Rev. 1971, 71, 483–505; b) M. Reuman,
A. I. Meyers, Tetrahedron 1985, 41, 837–860.
[38] A. I. Vogel, J. Basset, Vogel’s Textbook of Quantitative Inor-
ganic Analysis, 4th ed., Longman, New York, 1980.
[39] E. C. Prestes, V. E. dos Anjos, F. F. Sodré, M. T. Grassi, J.
Braz. Chem. Soc. 2006, 17, 53–60.
[40] G. R. Friedermann, G. G. Nunes, J. F. Soares, Quim. Nova
2005, 28, 340–344.
[41] S. J. Borg, T. Behrsing, S. P. Best, M. Razavet, X. Liu, C. J.
Pickett, J. Am. Chem. Soc. 2004, 126, 16988–16999.
[42] J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[43] G. M. Sheldrick, X-PREP, Bruker Analytical X-ray Instru-
ments Inc., Madison, Wisconsin, USA, 2006.
[32]
[33]
K. Ray, E. Bill, T. Weyhermüller, K. Wieghardt, J. Am. Chem.
Soc. 2005, 127, 5641–5654.
[34] P. Ghosh, A. Begum, E. Bill, T. Weyhermüller, K. Wieghardt,
Inorg. Chem. 2003, 42, 3208–3215.
[35] a) B. S. Kang, L. H. Weng, D. X. Wu, Z. Guo, L. R. Huang,
Z. Y. Huang, H. Q. Liu, Inorg. Chem. 1988, 27, 1128–1130; b)
C. A. Grapperhaus, S. Poturovic, Inorg. Chem. 2004, 43, 3292–
3298.
[36] D. D. Perrin, W. L. F. Armarego, Purification of Laboratory
Chemicals, 3rd ed., Butterworth-Heinemann, Oxford, 1988, pp.
66–309.
[44] G. M. Sheldrick, SHELXS-97: Program for Crystal Structure
Solution, University of Göttingen, Germany, 1997.
Received: January 22, 2010
Published Online: May 5, 2010
Eur. J. Inorg. Chem. 2010, 2476–2487
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
2487