Solution and solid state properties of Fe(III) complexes bearing N-ethyl-N-(2-aminoethyl)salicylaldiminate ligands Dedicated to Prof. Maria José Calhorda on the occasion of her 65th birthday

Article abstract of DOI:10.1016/j.jorganchem.2013.12.028

The effect of the phenolate ring derivatisation on the magnetic properties of Fe(III) complexes bearing N-ethyl-N-(2-aminoethyl)salicylaldiminate ligands both in solid state and solution have been investigated. Two new complexes [Fe(3,5-Br-salEen)₂]ClO₄.EtOH (5) and [Fe(3,5-Br-salEen) ₂]BPh₄.DMF (6) have been synthesised. SQUID magnetometry studies on these complexes showed that while complex 5 is in the low-spin (LS) state, complex 6 displays a gradual and incomplete spin crossover (SCO) transition over the temperature measured. Solution measurements on a series of six complexes - [Fe(salEen)₂]ClO₄ (1), [Fe(salEen) ₂]BPh₄·0.5H₂O (2)_, [Fe(5-Br-salEen)₂]ClO₄ (3), [Fe(5-Br-salEen) ₂]BPh₄·DMF (4), [Fe(3,5-Br-salEen) ₂]ClO₄·EtOH (5) and [Fe(3,5-Br-salEen) ₂]BPh₄·DMF (6) - were performed by UV-vis and NMR spectroscopies and cyclic voltammetry. Solution studies show that the presence of electron withdrawing groups (bromine atoms) affect the electronic density at the phenolate ring, thus influencing the ligand field strength and the separation between the t_2g and e_g* energy levels. The presence of two bromide substituents at the phenolate ring has a more pronounced effect on the magnetic behaviour in solution than in the solid state, with both complexes 5 and 6 adopting preferentially the LS state. Electrochemical studies of complexes 1-6 reveal that the reduction of the metallic centres in the complexes with electron withdrawing groups is easier, with E_1/2 values of iron moving to more positive potentials with the number of bromide substituents at the phenolate ring.

Full text of DOI:10.1016/j.jorganchem.2013.12.028

Journal of Organometallic Chemistry 760 (2014) 48e54
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Solution and solid state properties of Fe(III) complexes bearing
N-ethyl-N-(2-aminoethyl)salicylaldiminate ligands
,
a
a
a
a
Paulo N. Martinho a , Ana I. Vicente , Sara Realista , Marta S. Saraiva , Ana I. Melato ,
*
Paula Brandão b, Liliana P. Ferreira c d, Maria de Deus Carvalho e
,
a CQB, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
b Departamento de Química, CICECO, Universidade de Aveiro, 3810-193 Aveiro, Portugal
c CFMC, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
d Departamento de Física, Faculdade de Ciências e Tecnologia, Universidade de Coimbra, 3004-516 Coimbra, Portugal
e CCMM, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade de Lisboa, Campo Grande, 1749-016 Lisboa, Portugal
a r t i c l e i n f o
a b s t r a c t
Article history:
The effect of the phenolate ring derivatisation on the magnetic properties of Fe(III) complexes bearing N-
ethyl-N-(2-aminoethyl)salicylaldiminate ligands both in solid state and solution have been investigated.
Two new complexes [Fe(3,5-Br-salEen)2]ClO4.EtOH (5) and [Fe(3,5-Br-salEen)2]BPh4.DMF (6) have been
synthesised. SQUID magnetometry studies on these complexes showed that while complex 5 is in the
low-spin (LS) state, complex 6 displays a gradual and incomplete spin crossover (SCO) transition over the
temperature measured. Solution measurements on a series of six complexes e [Fe(salEen)2]ClO4 (1),
[Fe(salEen)2]BPh4$0.5H2O (2), [Fe(5-Br-salEen)2]ClO4 (3), [Fe(5-Br-salEen)2]BPh4$DMF (4), [Fe(3,5-Br-
salEen)2]ClO4$EtOH (5) and [Fe(3,5-Br-salEen)2]BPh4$DMF (6) e were performed by UVevis and NMR
spectroscopies and cyclic voltammetry. Solution studies show that the presence of electron withdrawing
groups (bromine atoms) affect the electronic density at the phenolate ring, thus influencing the ligand
field strength and the separation between the t2g and eg* energy levels. The presence of two bromide
substituents at the phenolate ring has a more pronounced effect on the magnetic behaviour in solution
than in the solid state, with both complexes 5 and 6 adopting preferentially the LS state. Electrochemical
studies of complexes 1e6 reveal that the reduction of the metallic centres in the complexes with electron
withdrawing groups is easier, with E1/2 values of iron moving to more positive potentials with the
number of bromide substituents at the phenolate ring.
Received 12 November 2013
Received in revised form
13 December 2013
Accepted 14 December 2013
Dedicated to Prof. Maria José Calhorda on
the occasion of her 65th birthday
Keywords:
Spin crossover
Fe(III) complexes
Electron withdrawing groups
Evans method
Cyclic voltammetry
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
magnetism in Fe(III) derivatives of various dithiocarbamates [5e7].
Later, the vision of Kahn and co-workers towards the application of
Schiff base ligands are very versatile for coordination to transi-
tion metal ions. Complexes with Schiff base ligands have been
widely studied [1] and coordination to first row transition metal
ions can display interesting magnetic properties, particularly SCO
or spin transition (ST) [2e4]. SCO is commonly observed for octa-
hedral metal complexes with a metal 3d4 to 3d7 electron configu-
ration which can interchange between two electronic states, high-
spin (HS) and LS, by application of an external perturbation such as
temperature, pressure, light or magnetic field. SCO was first re-
ported by Cambi and co-workers after detecting unusual
SCO compounds in data processing [8] resulted in a long and
impressive literature on SCO. This included the discovery of several
new examples of SCO compounds [3,9e11], the explanation of
different types of SCO profiles [4,12] and the modification of SCO
compounds to increase cooperativity and to direct their application
into materials science [13]. Scientists have developed SCO networks
[14e16], frameworks [17e19], gels [20e22], liquid crystals [23e25],
nanoparticles and nanocrystals [26e29], nanowires [30], thin films
[31e33] and have also applied patterning techniques to fabricate
SCO devices [34]. Among complexes displaying SCO, [Fe(salEen)2]þ
derivatives (salEen ¼ N-ethyl-N-(2-aminoethyl)salicylaldiminate)
are known to undergo thermal SCO sensitive to unit cell contents
and supramolecular packing in the crystal [35].
* Corresponding author. DQB/FCUL, Campo Grande, Ed. C8, 1749-016 Lisboa,
Portugal. Tel.: þ351 217500978; fax: þ351 217500088.
Hendrickson and co-workers reported the first examples of
salEen derivatives SCO compounds and also reported the distinct
E-mail address: pnmartinho@fc.ul.pt (P.N. Martinho).
0022-328X/$ e see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jorganchem.2013.12.028

P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
49
abrupt SCO profile for the [Fe(3-OMe-salEen)2]PF6 system [36]. This
inside the insert tube were measured and the mass susceptibility
report led to a number of new SCO compounds of the type [Fe(ꢀX-
(
cg) calculated from,
salEen)2]Y (X ¼ H, 3-OMe, 4-OMe or 5-OMe; Y ¼ Clꢀ, ClO4ꢀ; NO3
;
PF6ꢀ; or BPh4ꢀ) [35,37] displaying a range of SCO profiles from
incomplete and gradual to complete.ꢀAmong salEen derivatives, it
has been observed that Clꢀ and NO3 complexes tend to stabilise
3000 v
D
v0CM
cg
¼
c0
þ
4
p
ꢀ
where C is the concentration of the sample and M is the molecular
weight of the paramagnetic molecule. The molar susceptibility (cm
of the paramagnetic species is derived from the previous expres-
sion after appropriate diamagnetic correction.
the LS state, and BPh4 complexes adopt preferably the HS state
)
[35,36,38]. In solution the SCO process becomes an individual
molecular process and abrupt and hysteretic SCO is not expected
unless self-assembly occurs and interaction is sensed by neigh-
bouring molecules [39]. Additionally, the rigidity conferred by the
solid state intermolecular lattices is not observed and the solvent
and anion effect on the ligand field strength can be studied.
We have recently found that [Fe(5-Br-salEen)2]þ displays
different types of SCO profiles depending on the anion used and the
solvation of the unit cell [40]. We also observed that the H-bonding
between the ligand and the anion is crucial and a key to promote
SCO at room-temperature (RT). This work reports the effect of
electron withdrawing groups on the magnetic behaviour of the
metal both in solid state and solution. The ligands salEen (LA), 5-Br-
Electrochemical studies were performed using a CHI Electro-
chemical Analyser
e 620A controlled by software. A one-
compartment glass cell was used in all experiments; a platinum
wire and a platinum foil (1 cm2) were used as working and counter
electrodes, respectively.
A saturated calomel electrode (SCE,
0.244 V vs. SHE) was used as reference electrode. Before each
experiment the working electrode was cleaned by flame-annealing
in a butaneeoxygen Bunsen burner until the slide attained dark red
glowing. All solutions were deoxygenated directly in the electro-
chemical cell with a stream of N2 for 5 min. All experiments were
performed in dry acetonitrile solutions (1 mM) and tetrabuty-
lammonium tetrafluoroborate (NBu4BF4, 0.1 M). For complexes 1e
6, ferrocene (E1/2 ¼ 0.420 V vs. SCE) [42] was used as internal
standard.
ꢀ
salEen (LBꢀ) and 3,5-Br-salEen (LC) have been chosen and their ClO4
and BPh4 Fe(III) complexes, [Fe(X-salEen)2]Y (X ¼ H, 5-Br or 3,5-
Br; Y ¼ ClO4 or BPh4ꢀ), investigated. Solid state magnetic prop-
ꢀ
erties of [Fe(3,5-Br-salEen)2]ClO4$EtOH (5) and [Fe(3,5-Br-salEen)2]
BPh4$DMF (6) are reported as well as redox behaviour and mag-
netic properties in solution at RT for [Fe(salEen)2]ClO4 (1), [Fe(s-
alEen)2]BPh4$0.5H2O (2), [Fe(5-Br-salEen)2]ClO4 (3), [Fe(5-Br-
salEen)2]BPh4$DMF (4), [Fe(3,5-Br-salEen)2]ClO4$EtOH (5) and
[Fe(3,5-Br-salEen)2]BPh4$DMF (6).
Single X-ray crystal data for 6 were collected at 150(2) K by using
a Bruker SMART APEX II diffractometer equipped with a CCD area
detector and monochromated Mo-K
ꢀ
a
radiation (
l
¼ 0.71073 A). The
frames were integrated with the SAINT-Plus software package [43],
and the intensities were corrected for polarisation and Lorentz ef-
fects. A multi-scan absorption correction was also applied with the
SADABS [44]. The structure was solved by direct methods with
SHELXS-2013 [45] and refined by full-matrix least-squares based
on F2 using the program SHELXL-2013 [45]. Hydrogen atoms bound
to carbon and nitrogen were inserted at ideal geometric positions.
ORTEP and packing diagrams were drawn with the Olex2 [46] and
the Mercury software packages [47], respectively. The crystal data
together with refinement details are given in Table 1. The crystal
structure has been deposited with CCDC 976390.
2. Experimental
2.1. Materials and measurements
Salicylaldehyde,
5-bromosalicylaldehyde,
3,5-
dibromosalicylaldehyde, iron(III) nitrate nonahydrate, iron(II)
chloride, N-ethylethylenediamine, sodium perchlorate, sodium
tetraphenylborate, tetrabutylammonium tetrafluoroborate and
solvents were purchased and used without further purification. IR
spectra (KBr pellets) were recorded on a Nicolet Nexus 6700 FTIR
spectrophotometer. UVevis spectra were recorded on a Shimadzu
50/60 Hz spectrometer in acetonitrile solutions. Microanalyses (C,
H and N) were measured by elemental analysis service at the
University of Vigo, Spain.
Magnetisation measurements as a function of temperature were
performed using an SQUID magnetometer (Quantum Design
MPMS). The curves were obtained at 1000 Oe for temperatures
ranging from 10 to 370 K and the molar susceptibilities (cm) values
were corrected for diamagnetism.
2.2. Synthesis of compounds 1e4
Compounds 1 and 2 were obtained as described in the literature
[38,48]. Compounds 3 and 4 were synthesised as previously
described by our group [40].
Table 1
Crystal data parameters and selected refinement details of 6.
Empirical Formula
Mw
C49H53BBr4FeN5O3
1146.26
Monoclinic
P21/n
15.1177(4)
15.6799(4)
20.7440(5)
4906.3(2)
4
The Mössbauer spectra were recorded in transmission mode at
RT and at 78 K using a conventional constant-acceleration spec-
trometer and a 50 mCi 57Co source in a Rh matrix. The low tem-
perature measurements were performed using a liquid nitrogen
flow cryostat with a temperature stability of ꢁ0.5 K. The velocity
Crystal system
Space group
ꢀ
a/A
ꢀ
b/A
ꢀ
c/A
V/A
3
ꢀ
scale was calibrated using an a-Fe foil. The spectra were fitted to
Z
Lorentzian lines using the WinNormos software program, and the
isomer shifts reported are relative to metallic -Fe at RT.
rcalc/mg mmꢀ3
1.52
0.500
m
/mmꢀ1
a
Magnetic measurements of complexes 1e6 in solution were
performed at RT by 1H NMR using the Evans’ method [41] (Bruker
Avance 400 spectrometer operating at 400.14 MHz at a constant
temperature of 298.15 K). The measurements were performed in
NMR tubes containing the paramagnetic samples (5 mM) dissolved
in CD3CN with an inert reference of 0.03% TMS, against a reference
insert tube filled with the same solvent (0.03% TMS in CD3CN). The
difference between the shifts (in Hz) given by the TMS outside and
F(000)
2308
2
q
range for data collection/ꢂ
Index ranges
ꢀ19 ꢃ h ꢃ 16, ꢀ20 ꢃ k ꢃ 20, ꢀ27 ꢃ l ꢃ 27
Crystal size/mm
Reflections collected
Unique reflections, [Rint
Final Rindices
R1, wR2 [I > 2
R1, wR2 (all data)
0.5 ꢄ 0.12 ꢄ 0.03
65,747
]
11,722 [0.0453]
sI]
0.0346, 0.0779 [8400]
0.0630, 0.0917

50
P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
2.3. Synthesis of compound 5 e [Fe(3,5-Br-salEen)2]ClO4$EtOH
To a solution of N-ethylethylenediamine (0.105 mL, 1 mmol) in
methanol (20 mL) 3,5-dibromosalicylaldehyde (279 mg, 1 mmol)
was added and left stirring for 15 min to give a yellow solution. A
solution of iron(II) chloride (64 mg, 0.5 mmol) and sodium
perchlorate monohydrate (70 mg, 0.5 mmol) in methanol was
added to the mixture and left stirring for 1 h. The purple solution
was filtered and a dark solid was obtained after slow evaporation of
the solvent. The crude solid was recrystallised from an ethanol/
methanol mixture (50:50) yielding a microcrystalline solid after
slow evaporation of the solvent (108 mg, 13%). IR: nmax/cmꢀ1 3210
(nNH, m), 3051 (nCH, w), 1635 (nC]N, s), 1578 (dC]C, m), 1083
ð
nClO ; sÞ; 626 ðnClO ; sÞ: Anal. found (calcd) for C24H32Br4ClFeN4O7:
4
4
C 32.15 (32.05); H 3.24 (3.59); N 6.80 (6.23).
2.4. Synthesis of compound 6 e [Fe(3,5-Br-salEen)2]BPh4$DMF
To a solution of N-ethylethylenediamine (0.105 mL, 1 mmol) in
methanol (20 mL) 3,5-dibromosalicylaldehyde (279 mg, 1 mmol)
was added and the yellow solution left stirring for 15 min. A so-
lution of iron(II) chloride (64 mg, 0.5 mmol) and sodium tetra-
phenylborate (170 mg, 0.5 mmol) in methanol was added to the
mixture and left stirring for 1 h. Filtration of the reaction mixture
allowed to recover a dark microcrystalline solid. Recrystallisation
from methanol/toluene/dimethylformamide (40:20:40) produced
needle shaped crystals suitable for single crystal X-ray diffraction
(250 mg, 23%). IR: nmax/cmꢀ1 3210 (nNH, m), 3051 (nCH, w), 1633
Fig. 1. ORTEP view of the association between [Fe(3,5-Br-salEen)2]þ and DMF with two
NeH$O hydrogen bonds drawn as dashed lines and thermal ellipsoids drawn at the
50% probability level.
(
n
C]N, s), 1590 (
d
C]C, m), 744 ðnBPh ; sÞ; 708 ðnBPh ; sÞ: Anal. found
4
4
ꢀ
744 and 708 cmꢀ1 attributed to BPh4 and 1633 cmꢀ1 corre-
sponding to the C]N group.
(calcd) for C49H53BBr4FeN5O3: C 51.34 (51.34); H 4.78 (4.66); N 6.14
(6.11).
3. Results and discussion
3.2. Crystal structure of compound 6
3.1. Synthesis of complexes 1e6
The crystal structure of compound 6 is built up from an asym-
metric unit composed of one cation [Fe(3,5-Br-salEen)2]þ, one
ꢀ
Complexes 1e6 were prepared as described in Scheme 1. The
new complexes, 5 and 6 were obtained as solvated forms: [Fe(LC)2]
ClO4$EtOH (5) and [Fe(LC)2]BPh4$DMF (6). In the solid state, crys-
talline samples of both complexes were characterised by elemental
analysis, IR, 57Mössbauer spectroscopy and SQUID magnetometry
and the structure of complex 6 determined by single crystal X-ray
diffraction. Solution properties of complexes 1e6 were determined
at RT by UVevis and NMR spectroscopies and their redox behaviour
studied by cyclic voltammetry.
The IR spectrum of 5 presentedꢀa characteristic C]N stretching
band at 1633 cmꢀ1 and two ClO4 stretching bands at 1083 and
626 cmꢀ1 confirming that formation of the imine was achieved and
the exchange of the anion occurred. For complex 6 the IR spectrum
also confirmed the success of the synthesis with stretching bands at
BPh4 anion and one DMF solvent molecule. The complex cation
and DMF molecule are associated by two NeH/O hydrogen bonds
ꢀ
with N/O distances of 2.962(3) and 2.928(3) A and NeH/O angles
of 170 and 172ꢂ, respectively. The ORTEP view of this association is
presented in Fig. 1. The metal complex has a distorted octahedral
geometry with three nitrogen atoms adopting a spatial disposition
consistent with a fac-isomer. Selected bond distances and angles in
the metal coordination sphere are listed in Table 2. The FeeN
ꢀ
ꢀ
(1.9279(2)e2.0315(2) A) and FeeO (1.8679(2) and 1.8765(2) A)
distances are consistent with the existence of a Fe(III) centre in the
LS state and are equivalent to those observed for the analogous
[Fe(5-Br-salEen)2]BPh4$DMF [40].
The crystal packing of 6 is view along the b crystallographic axis,
ꢀ
shown in Fig. 2, and reveals that the BPh4 anions are arranged in
Scheme 1. General method used in the synthesis of complexes 1e6 (*for complexes 1e4).

P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
51
Table 2
ꢂ
ꢀ
Selected bond distances (A) and angles ( ) of 6.
Bond lengths
FeeO(1)
FeeO(15)
FeeN(9)
1.8679(2)
1.8765(2)
1.9279(2)
FeeN(12)
FeeN(23)
FeeN(26)
2.0283(2)
1.9222(2)
2.0315(2)
Bond angles
O(1)eFeeO(15)
O(1)eFeeN(9)
O(1)eFeeN(12)
O(1)eFeeN(23)
O(1)eFeeN(26)
O(15)eFeeN(9)
O(15)eFeeN(12)
O(15)eFeeN(23)
94.25(8)
93.38(8)
177.27(9)
86.85(8)
86.81(9)
85.65(9)
87.29(9)
92.52(8)
O(15)eFeeN(26)
N(9)eFe-N(12)
N(9)eFeeN(23)
N(9)eFeeN(26)
N(12)eFeeN(23)
N(12)eFeeN(26)
N(23)eFeeN(26)
176.45(9)
84.48(9)
178.17(9)
97.67(9)
95.35(9)
91.78(9)
84.15(9)
the crystal in open channels, which accommodates the complex
cations. The DMF crystallisation solvent molecules are positioned
ꢀ
between BPh4 anions. The shortest intermolecular Fe/Fe dis-
tance is 8.991 A indicating the absence of the electronic commu-
nication between the iron centres in the crystalline state.
Fig. 3. Temperature dependence of cmT for compounds 5 and 6, obtained under
1000 Oe, on cooling (solid symbols) and warming (open symbols).
ꢀ
Fig. 4 shows the Mössbauer spectra of complexes 5 (78 K and
290 K) and 6 (78 K), and the hyperfine parameters obtained from
the fitting procedures are presented in Table 3. Mössbauer studies
of compounds 3 and 4 are reported elsewhere [40].
3.3. Solid state characterisation of 5 and 6
The magnetic profile of complexes 1 and 2 has been previously
described [38,48]. While complex 1 showed thermal magnetic
dependence exhibiting SCO behaviour, complex 2 is stabilised in
the HS state over the measured temperature range. The thermal
magnetic behaviour of crystalline samples 3e6 was measured by
SQUID magnetometry on cooling and warming modes. Results
obtained for complexes 3 and 4 are reported elsewhere [40], and
indicated that although differently, both complexes exhibited
incomplete SCO. Complex 3 showed a gradual SCO on cooling and
an abrupt ST on warming with an asymmetric 70 K hysteresis loop.
Compound 4 displayed a gradual SCO, and its magnetic behaviour
showed to be sensitive to the applied magnetic field and sample
thermal history.
The Mössbauer spectra collected at 78 K can be resolved using a
single, but asymmetric, quadrupole doublet with isomer shifts (
d)
around 0.2 mm/s and quadrupole splittings ( EQ) near 2.8 mm/s,
D
which are consistent with Fe(III) ions in an LS state. The assignment
of this doublet to LS Fe(III) is also supported by the low temperature
magnetisation results of both samples, with cmT values close to
those expected for spin only S ¼ 1/2, 0.38 cm3 molꢀ1 K, particularly
visible for compound 6.
Fig. 3 displays the results obtained for samples 5 and 6. At low
temperature both compounds are in the LS state with cmT values of
0.59 and 0.37 cm3 molꢀ1 K, respectively. Complex 6 remains in the
LS state up to 150 K, where cmT starts to increase slowly, displaying
an incomplete thermal SCO, as indicated by the small value of cmT,
2.4 cm3 molꢀ1 K, obtained at the highest measured temperature,
370 K. In the case of complex 5 cmT values are almost constant
(varying between 0.59 and 0.86 cm3 molꢀ1 K), suggesting that the
population of the LS state is essentially maintained up to 300 K.
Fig. 2. Crystal packing diagram view along the b crystallographic axis showing the
ꢀ
assembly of BPh4 anions in open channels, which are occupied by cations [Fe(3,5-Br-
salEen)2]þ. For clarity in each channel only one molecule of the complex is shown in
ball-stick fashion. The BPh4 anions are represented in space filling mode.
ꢀ
Fig. 4. 57Fe Mössbauer spectra of complexes 5 (at 78 and 290 K) and 6 (at 78 K).

52
P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
Table 3
Estimated parameters from the Mössbauer spectra of 5 and 6.
290 K
78 K
Compound
d
(mm/s)
D
EQ (mm/s)
G
(mm/s)
W21 (mm/s)
d
(mm/s)
DEQ (mm/s)
G
(mm/s)
W21 (mm/s)
5
6
0.12(1)
0.20(5)
2.80(2)
2.6(1)
0.31(3)
0.9(2)
e
e
0.191(1)
0.209(4)
2.851(3)
2.798(7)
0.312(5)
0.52(1)
0.92(2)
0.54(1)
d
e isomer shift; DEQ e quadrupole splitting; G e FWHM line width; W21 e relative line width.
At RT, the spectra of both complexes reveal poor resonant ab-
sorption due to the small fraction of recoilless iron nuclei. In fact
one doublet is clearly perceptible in the RT spectrum of 5, Fig. 4,
agreeing with magnetisation data. The decrease observed in the
isomer shift value compared to that obtained at 78 K can be
explained by the second-order Doppler effect [3,49].
The RT spectrum resolution of 6 (not shown) is poor and it was
not possible to clearly define the characteristics of two iron sites in
a reasonable spectrum acquisition time, justifying the broad line
width (
G
¼ 0.9(2) mm/s) and the mean isomer shift value
(d
¼ 0.20(5) mm/s) obtained (Table 3). This is certainly due to a spin
interconversion process running at 290 K, as shown by the mag-
netisation data (Fig. 3).
The asymmetric line broadening of the LS Fe(III) doublet
observed for both complexes at 78 K, is estimated by parameter
W21 (Table 3) which represents the relative line width between
the higher and the lower energy absorption lines of the doublet.
Both Fig. 4 and Table 3, clearly show that at 78 K this asymmetry
is much more relevant for compound 6, and the poor resolution of
the RT spectra does not allow a meaningful quantification of W21
at this temperature. This type of line broadening asymmetry can
be attributed to fluctuating electric and magnetic fields associated
with the relaxation of paramagnetic ions, as well as to changes on
the environment surrounding the iron nuclei [50]. Taking into
account the slow but continuous spin interconversion of sample 6
above 100 K and the almost constant value of cmT for compound
5 (varying between 0.59 and 0.86 cm3 molꢀ1 K), it is reasonable to
accept that relaxation mechanisms associated with thermal SCO
in compound 6 account for the high broad line asymmetry
obtained.
Fig. 5. RT UVevis spectra of complexes 1e6 for 7 ꢄ 10ꢀ5 mol Lꢀ1 acetonitrile solutions.
for 1 and 520 nm for 2). This is followed by the compounds bearing
ligands with one electron withdrawing group (5-Br-salEen), 3 and
4, exhibiting the maximum absorbance at 534 nm. Finally, the
introduction of a second electron withdrawing group at the
phenolate ring (3,5-Br-salEen) shifted the absorption maxima to
lower energies (558 nm for 5 and 548 nm for 6). It is worth noticing
that the presence of either one or two electron withdrawing groups
at the phenolate ring causes a red shift. Bromine, with s-inductive
electron withdrawing character, restricts the electron density
available. This restriction alters the ligand field as well as the Doct
and, according to Tweedle and Wilson, such changes should be
more pronounced for substituents ortho to the oxygen atom at the
phenolate ring [51]. In fact the electronic spectra of complexes 1e6
3.4. Solution measurements
Complexes 1e6 were characterised by UVevis spectroscopy and
their redox behaviour studied by cyclic voltammetry. The magnetic
susceptibility in solution of complexes 1e6 at RT was determined
by the Evans’ method [41] using 1H NMR spectroscopy.
UVevis spectra of compounds 1e6 were recorded at RT for so-
lutions of each complex in acetonitrile (7 ꢄ 10ꢀ5 mol Lꢀ1) and can
be seen in Fig. 5.
at RT show this effect. The difference between the values of the HS
3
and LS bands is smaller for complexes 5 and 6 when compared to
those of complexes 1e4 indicating a clear effect of the phenolate
substitution on the spin equilibrium.
Magnetic susceptibilities of all complexes at RT were deter-
mined using the Evans’ method [41] where the RT 1H NMR spectra
of 5 mM solutions of 1e6 in CD3CN with 0.03% TMS were recorded
against a reference solution of 0.03% TMS in CD3CN, Table 4. The
cmT value of complex 1 (4.41 cm3 molꢀ1 K) shows that this complex
The UVevis spectra of all complexes display charge transfer and
intraligand
p
ep
* transitions below 400 nm with varying between
3
z4000 (6) and z12,000 mol Lꢀ1 cmꢀ1 (2). For all six complexes,
two additional absorbance bands appear in the region 450e850 nm
possibly from the ligand-to-metal charge transfer (LMCT) transi-
tions, Table 4. Electronic studies on Fe(III) complexes bearing
(sal)2trien (trien ¼ triethylenetetramine) ligands showed that in
the 450e850 nm region there is a higher energy band corre-
sponding to the HS species and a second band at lower energy
assigned to the LS species [51]. Taking this into account, we attri-
Table 4
Electronic absorption maxima and extinction coefficients in acetonitrile and cm
T
values obtained from 1H NMR results in CD3CN for 1e6 at RT.
ꢀ
ꢀ
ClO4
R1 ¼ H R1 ¼ H
BPh4
R1 ¼ Br R1 ¼ H R1 ¼ H
R1 ¼ Br
R2 ¼ H R2 ¼ Br R2 ¼ Br R2 ¼ H R2 ¼ Br R2 ¼ Br
bute the bands between 519 nm < lmax < 548 nm to the HS species
1
3
5
2
4
6
and those between 655 nm < lmax < 680 nm to the LS species.
For complexes 1e6 the band attributed to the HS species shows
a pronounced difference directly related to the degree of substi-
tution at the phenolate ring. The non-substituted compounds (1
and 2) present the maximum absorbance at higher energy (519 nm
lmax (nm)
(mol Lꢀ1 cmꢀ1
lmax (nm)
(mol Lꢀ1 cmꢀ1
519
2271
678
569
4.4
534
3571
678
558
2806
655
520
3670
678
959
3.1
534
2556
680
548
1343
673
797
2.0
3
)
)
3
1547
2227
1206
cmT (cm3 molꢀ1 K)
2.7
1.6
3.1

P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
53
Fig. 6. Cyclic voltammograms of a Pt electrode in acetonitrile solutions with 0.1 M of NBu4BF4 as supporting electrolyte containing 1 mM of a) complexes 1 (solid line), 3 (dash
dotted line) and 5 (dashed line) and b) complexes 2 (solid line), 4 (dash dotted line) and 6 (dashed line). Cyclic voltammogram of 0.1 M of NBu4BF4 in acetonitrile (dotted line). Scan
rate: 200 mV sꢀ1
.
Table 5
is essentially in the HS state at RT, while the values obtained for all
the other complexes indicate a mixture of both HS and LS states.
Complexes 5 and 6, for which R1 ¼ R2 ¼ Br, are those exhibiting the
lowest cmT values, indicating a major fraction of LS state at RT. For
complexes 2, 3 and 4 the cmT values indicate a major fraction of HS
state, with values of 67, 59 and 69% respectively. These results
clearly show the ligand influence on the magnetic behaviour of the
metal in solution, as the cmT value, and thus of HS fraction, de-
creases with the presence of bromide substituents at the phenolatꢀe
ring. This fact is observed for the three complexes containing ClO4
Electrochemical data of complexes 1e6.a
Complex
EFe(III)/Fe(II)/V EFe(II)/Fe(III)/V E1/2/V
DEp/V
1
R1 ¼ R2 ¼ H
ꢀ0.478
ꢀ0.360
ꢀ0.360
ꢀ0.260
ꢀ0.185
ꢀ0.357
ꢀ0.280
ꢀ0.160
ꢀ0.414 0.059
ꢀ0.310 0.100
ꢀ0.250 0.070
ꢀ0.441 0.079
ꢀ0.320 0.070
ꢀ0.225 0.070
ꢀ
ClO4
3 [40] R2 ¼ Br
5
2
R1 ¼ R2 ¼ Br ꢀ0.325
R1 ¼ R2 ¼ H
ꢀ0.515
ꢀ0.350
ꢀ
BPh4
4 [40] R2 ¼ Br
6
R1 ¼ R2 ¼ Br ꢀ0.300
a
Acetonitrile solution, 0.1 M NBu4BF4 as supporting electrolyte: E1/2 vs. SCE, Pt
working electrode, v ¼ 200 mV sꢀ1 Fcþ/Fc (E1/2 ¼ þ0.420 V [42]) as internal stan-
ꢀ
dard. E1/2 ¼ ½ (Epa þ Epc), Ep ¼ Epa ꢀ Epc (mV).
D
and also for the solvated complexes 4 and 6 (BPh4 as anion).
Within the same ligand, a variation of cmT is also observed when
solvates are present and the anion changes. These observations
indicate that the solvent and/or anion interact with the complex
cation, affecting the cmT value.
These results are in good agreement with those of the UVevis
measurements confirming that the second bromide at the pheno-
late ring has an effect on the ligand field strength increasing the
separation between the t2g and the eg* energy levels. Further work
is now underway to investigate the solvent and temperature effect
on the magnetic behaviour of these complexes.
significantly affect the E1/2 values of the iron centres, Table 5. While
complex 1 exhibits a reversible redox process (
D
Ep ¼ 59 mV) all
Ep > 59 mV)
other complexes show a quasi-reversible behaviour (
D
[52].
The redox processes of the ligands analysed by the results ob-
tained for ClO4ꢀ complexes (1, 3 and 5, Fig. 6a), show a shift to more
positive values with the increase of bromide substituents at the
phenolate ring. This can be attributed to the presence of electron
withdrawing groups at the ring, restricting the electron density
available at the ligand. For the BPh4ꢀ complexes (2, 4 and 6, Fig. 6b)
it has to be noted that the anion itself can also be oxidised over the
same potential window as the ligand (Fig. S1 of the Supporting
information), challenging the clear identification of the processes
involving both ligand and anion.
The redox behaviour of complexes 1e6 in acetonitrile was
determined at RT by cyclic voltammetry (Fig. 6). Well-defined
anodic and cathodic peaks with E1/2 values between ꢀ0.441
and ꢀ0.225 V (Table 5) are related to the iron centres. Additionally,
complexes 2, 4 and 6 show peaks between 0.750 and 1.7 V that can
be related to redox processes of both ligands and BPh4ꢀ: Within the
same potential range complexes 1,ꢀ3 and 5 only display ligand
related redox processes, as the ClO4 anion did not show a redox
response in the potential range studied. Attempts to isolate the li-
gands in order to clarify their redox processes were unsuccessful.
The fact that the electrochemical behaviour of the complexes is
affected by the number of electron withdrawing groups at the
phenolate ring is noteworthy. Namely, all redox processes shift to
more positive values with the increase of bromide substituents at
the ring, Fig. 6. This fact is clearly perceptible when the E1/2
values of the iron redox processes (Table 5) are compared: a value
of E1/2 ¼ ꢀ0.414 V is observed for complex 1, and the presence of
one bromide substituent at the phenolate ring (complex 3) shifts
this value by 100 mV to more positive values (ꢀ0.310 V). Addi-
tionally an even higher shift to positive values (ꢀ0.250 V) is
observed when a second bromide (complex 5) is present at the
ligand. This trend is also observed for the BPh4ꢀ complexes 2, 4 and
6. These results demonstrate that the reduction of the iron centre is
easier with the increase of bromide subꢀstituents at the ring.
However, the nature of the anion ðClO4 or BPh4ꢀÞ does not
4. Conclusions
Two new complexes, [Fe(3,5-Br-salEen)2]ClO4$EtOH (5) and
[Fe(3,5-Br-salEen)2]BPh4$DMF (6) were synthesised and their
magnetic properties investigated. The crystal packing diagram of
complex 6 showed that the interaction between the metal centres
is obstructed by the anion, affecting the SCO behaviour. SQUID
magnetometry and Mössbauer spectroscopy showed that while 6
presents a gradual and incomplete SCO, 5 stabilises in the LS state
up to 300 K. Comparison of the solid state magnetic profiles of
newly reported complexes 5 and 6 with previously reported com-
plexes 1e4 did not allow to draw conclusions on the influence of
ligand substituent (s-inductive electron withdrawing groups e
bromine) on their magnetic behaviour. However, both spectro-
scopic and electrochemical studies of the complexes in solution
have shown that there is a clear effect on their properties
depending on the degree of substitution at the phenolate ring.
Interestingly, both UVevis measurements and Evans magnetic

54
P.N. Martinho et al. / Journal of Organometallic Chemistry 760 (2014) 48e54
[16] Y. Garcia, V. Niel, M.C. Munoz, J.A. Real, Top. Curr. Chem. 233 (2004) 229e257.
[17] M. Ruben, U. Ziener, J.M. Lehn, V. Ksenofontov, P. Gutlich, G.B.M. Vaughan,
Chem. Eur. J. 11 (2004) 94e100.
[18] T. Kitazawa, Y. Gomi, M. Takahashi, M. Takeda, M. Enomoto, A. Miyazaki,
T. Enoki, J. Mater. Chem. 6 (1996) 119e121.
[19] V. Niel, J.M. Martinez-Agudo, M.C. Munoz, A.B. Gaspar, J.A. Real, Inorg. Chem.
40 (2001) 3838.
[20] O. Roubeau, A. Colin, W. Schmitt, R. Clerac, Angew. Chem. Int. Ed. 43 (2004)
3283e3286.
susceptibility determinations at concentrations separated by a
factor of z100 (0.07 mM for UVevis and 5 mM for Evans’ method)
confirmed the effect of the ligand substituents on their magnetic
character. The results indicate that in solution complex 1 is
essentially in the HS state, and both HS and LS forms are present for
the other complexes. When compared to the unsubstituted com-
plexes (1 and 2), a significant decrease on cmT is observed when
two bromide substituents are present at the phenolate ring,
showing a clear preference for the LS state, and indicating an
increased separation between the t2g and eg* energy levels. The
bromide substituent effect has also been observed on the electro-
chemical behaviour of the complexes. The reduction of the metallic
centres proved to be easier for ligands with increasing number of
electron withdrawing groups, as the E1/2 values of iron shift to more
positive potentials.
[21] P. Grondin, O. Roubeau, M. Castro, H. Saadaoui, A. Colin, R. Clerac, Langmuir 26
(2010) 5184e5195.
[22] T. Fujigaya, D.L. Jiang, T. Aida, Chem. Asian J. 2 (2007) 106e113.
[23] Y. Galyametdinov, V. Ksenofontov, A. Prosvirin, I. Ovchinnikov, G. Ivanova,
P. Gütlich, W. Haase, Angew. Chem. Int. Ed. 40 (2001) 4269e4271.
[24] Y. Bodenthin, G. Schwarz, Z. Tomkowicz, M. Lommel, T. Geue, W. Haase,
H. Mohwald, U. Pietsch, D.G. Kurth, Coord. Chem. Rev. 253 (2009) 2414e2422.
[25] A.B. Gaspar, M. Seredyuk, R. Gutlich, Coord. Chem. Rev. 253 (2009)
2399e2413.
[26] E. Coronado, J.R. Galan-Mascaros, M. Monrabal-Capilla, J. Garcia-Martinez,
P. Pardo-Ibanez, Adv. Mater. 19 (2007) 1359.
[27] T. Forestier, S. Mornet, N. Daro, T. Nishihara, S. Mouri, K. Tanaka, O. Fouche,
E. Freysz, J.F. Letard, Chem. Commun. (2008) 4327e4329.
[28] F. Volatron, L. Catala, E. Riviere, A. Gloter, O. Stephan, T. Mallah, Inorg. Chem.
47 (2008) 6584e6586.
[29] I. Boldog, A.B. Gaspar, V. Martinez, P. Pardo-Ibanez, V. Ksenofontov,
A. Bhattacharjee, P. Gutlich, J.A. Real, Angew. Chem. Int. Ed. 47 (2008) 6433e
6437.
[30] P.N. Martinho, T. Lemma, B. Gildea, G. Picardi, H. Muller-Bunz, R.J. Forster,
T.E. Keyes, G. Redmond, G.G. Morgan, Angew. Chem. Int. Ed. 51 (2012)
11995e11999.
[31] A. Ruaudelteixier, A. Barraud, P. Coronel, O. Kahn, Thin Solid Films 160 (1988)
107e115.
Acknowledgements
We thank Fundação para Ciência e Tecnologia for financial
support (PEst-OE/QUI/UI0612/2013, PEst-OE/QUI/UI0536/2011,
PEst-OE/FIS/UI0261/2011 and PTDC/QUI-QUI/101022/2008) and
fellowships to AIM (SFRH/BPD/69526/2010), PNM (SFRH/BPD/
73345/2010), MSS (SFRH/BPD/88082/2012).
Appendix A. Supplementary material
[32] P. Coronel, A. Barraud, R. Claude, O. Kahn, A. Ruaudelteixier, J. Zarembowitch,
J. Chem. Soc. Chem. Commun. (1989) 193e194.
[33] S. Cobo, G. Molnar, J.A. Real, A. Bousseksou, Angew. Chem. Int. Ed. 45 (2006)
5786e5789.
[34] M. Cavallini, I. Bergenti, S. Milita, G. Ruani, I. Salitros, Z.R. Qu, R. Chandrasekar,
M. Ruben, Angew. Chem. Int. Ed. 47 (2008) 8596e8600.
[35] M.S. Haddad, M.W. Lynch, W.D. Federer, D.N. Hendrickson, Inorg. Chem. 20
(1981) 123e131.
CCDC 976390 contains the supplementary crystallographic 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.
[36] M.S. Haddad, W.D. Federer, M.W. Lynch, D.N. Hendrickson, J. Am. Chem. Soc.
102 (1980) 1468e1470.
Appendix B. Supplementary information
[37] T.A. Ivanova, I.V. Ovchinnikov, A.N. Turanov, Phys. Solid State 49 (2007) 2132e
2137.
[38] M.S. Haddad, W.D. Federer, M.W. Lynch, D.N. Hendrickson, Inorg. Chem. 20
(1981) 131e139.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jorganchem.2013.12.028.
[39] P.N. Martinho, Y. Ortin, B. Gildea, C. Gandolfi, G. McKerr, B. O’Hagan,
M. Albrecht, G.G. Morgan, Dalton Trans. 41 (2012) 7461e7463.
[40] A.I. Vicente, S. Realista, L.P. Ferreira, M.D. Carvalho, A. Melato, P. Brandão, M.J.
Calhorda, P.N. Martinho, (2013), submitted for publication.
[41] D.F. Evans, J. Chem. Soc. (1959) 2003e2005.
References
[1] P.A. Vigato, S. Tamburini, Coord. Chem. Rev. 248 (2004) 1717e2128.
[2] H. Toftlund, J.J. McGarvey, Top. Curr. Chem. 233 (2004) 151e166.
[3] P.J. van Koningsbruggen, Y. Maeda, H. Oshio, Top. Curr. Chem. 233 (2004)
259e324.
[4] P. Gutlich, H.A. Goodwin, Top. Curr. Chem. 233 (2004) 1e47.
[5] L. Cambi, L. Szegö, Ber. Dtsch. Chem. Ges. 64 (1931) 2591e2598.
[6] L. Cambi, L. Szegö, Ber. Dtsch. Chem. Ges. 66 (1933) 656e661.
[7] L. Cambi, L. Malatesta, Ber. Dtsch. Chem. Ges. 70 (1937) 2067e2078.
[8] O. Kahn, E. Codjovi, Y. Garcia, P.J. van Koningsbruggen, R. Lapouyade,
L. Sommier, ACS Sym. Ser. 644 (1996) 298e310.
[9] P. Gutlich, P.J. van Koningsbruggen, F. Renz, Struct. Bond. 107 (2004) 27e75.
[10] H.A. Goodwin, Top. Curr. Chem. 234 (2004) 23e47.
[11] Y. Garcia, P. Gutlich, Top. Curr. Chem. 234 (2004) 49e62.
[12] A. Hauser, Top. Curr. Chem. 233 (2004) 49e58.
[13] M.A. Halcrow, Spin-crossover Materials: Properties and Applications, John
Wiley & Sons Ltd, 2013.
[42] M.A. Vorotyntsev, M. Casalta, E. Pousson, L. Roullier, G. Boni, C. Moise, Elec-
trochim. Acta 46 (2001) 4017e4033.
[43] Bruker, SAINT-plus, Bruker AXS Inc, Madison, Wisconsin, USA, 2007.
[44] G.M. Sheldrick, SADABS, University of Göttingen, Germany, 1996.
[45] G.M. Sheldrick, Acta Crystallogr. A64 (2008) 112.
[46] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard, H. Puschmann, J. Appl.
Crystallogr. 42 (2009) 339e341.
[47] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock,
L. Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, Appl. Crystallogr.
41 (2008) 466e470.
[48] C.F. Sheu, S.M. Chen, G.H. Lee, Y.H. Liu, Y.S. Wen, J.J. Lee, Y.C. Chuang, Y. Wang,
Eur. J. Inorg. Chem. (2013) 894e901.
[49] P. Gütlich, E. Bill, A.X. Trautwein, Mössbauer Spectroscopy and Transition
Metal Chemistry, Springer-Verlag, Berlin, Heidelberg, 2011.
[50] M. Blume, Phys. Rev. Lett. 14 (1965) 96e98.
[14] S.B. Erenburg, N.V. Bausk, L.G. Lavrenova, V.A. Varnek, L.N. Mazalov, Solid
State Ionics 101 (1997) 571e577.
[15] Y. Garcia, P.J. van Koningsbruggen, R. Lapouyade, L. Fournes, L. Rabardel,
O. Kahn, V. Ksenofontov, G. Levchenko, P. Gutlich, Chem. Mater. 10 (1998)
2426e2433.
[51] M.F. Tweedle, L.J. Wilson, J. Am. Chem. Soc. 98 (1976) 4824e4834.
[52] D. Pletcher, R. Greff, R. Peat, L.M. Peter, J. Robinson, Instrumental Methods in
Electrochemistry, Ellis Horwood Ltd, 1990.