Iron(II) spin-transition complexes with dendritic ligands, part II

Article abstract of DOI:10.1002/ejic.201000067

The dendritic triazole-based complexes [Fe(G1-BOC)₃](triflate) ₂·xH₂O (1; G1-BOC = tert-butyl {3-[3-(3-tert- butoxycarbonylaminopropyl)-5-([1,2,4]triazol-4-ylcarbamoyl)-phenyl]propyl} carbamate, triflate = CF₃SO

Full text of DOI:10.1002/ejic.201000067

FULL PAPER
DOI: 10.1002/ejic.201000067
Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II[‡]
YongLi Wei,[a] Prashant Sonar,[b] Matthias Grunert,[a] Joachim Kusz,[a,c]
A. Dieter Schlüter,[b] and Philipp Gütlich*[a]
Dedicated to Professor Rolf W. Saalfrank on the occasion of his 70th birthday
Keywords: Spin crossover / Iron / Dendrimers / Moessbauer spectroscopy / Magnetic properties
The dendritic triazole-based complexes [Fe(G1-BOC)3](tri-
flate)2·xH2O (1; G1-BOC = tert-butyl {3-[3-(3-tert-butoxy-
carbonylaminopropyl)-5-([1,2,4]triazol-4-ylcarbamoyl)-
phenyl]propyl}carbamate, triflate = CF3SO3–), [Fe(G1-BOC)3]-
(tosylate)2·xH2O(2;tosylate=p-CH3PhSO3–),[Fe(G1-DPBE)3]-
rimetry, respectively. All dendritic complexes under study
show different spin-transition behaviour with respect to the
nature of different dendritic ligands and counteranions.
Complexes 1 and 2 have pronounced effects of a spin-state
change during the first heating process and gradual spin-
transition properties for further temperature treatments,
(triflate)2·xH2O {3; G1-DPBE
= 3,5-bis(3,5-didodecaoxy-
benzyloxy)-N-[1,2,4]triazol-4-ylbenzamide}, [Fe(G1-DPBE)3]- whereas 3 and 4 exhibited a very sharp spin-state change in
(tosylate)2·xH2O (4) and [Fe(G1-DPBE)3](BF4)2·xH2O (5) were
designed and synthesized. Magnetic and thermal properties
of these novel complexes were characterized by magnetic
susceptibility measurements, 57Fe Mössbauer spectroscopy
and thermogravimetric analysis or differential scanning calo-
the first heating procedures. Complex 5 showed a gradual
spin-transition curve. In this paper, we report how the
magnetic properties of these complexes are correlated with
noncoordinated water molecules and their effects on spin
states.
used for the electrochemical reaction of CO2 to CO[13,14]
and iron dendritic complexes were used as spatially encum-
bered models of nonheme iron proteins.[15] Our motivation
also falls in the same line to make metal–dendritic-system-
based complexes with various interesting properties de-
pendent on different ligands, metal ions and counteranions.
It is well documented that iron(II)-containing dendritic
complexes may exhibit thermal spin-crossover (SCO) be-
haviour. This particular unique property with an ac-
companying change of colour and magnetic behaviour is
very useful for various applications such as display devices,
optical switches and magneto-optical storage systems.[16,17]
Spin crossover is related to its electronic transition between
Introduction
Dendritic materials are interesting due to their unique
macromolecular monodispersed nature, generation-depend-
ent properties and the possibilities they offer for anchoring
various groups for functionalization. These materials have
received much attention recently with regards to fundamen-
tal research and because of their applications in various
fields.[1–3] For example, incorporation of metals in a den-
dritic moiety may provide an opportunity to generate new
metallodendrimer materials,[4–8] dendritic boxes and other
supramolecular dendritic arrangements.[9–11] These metallo-
dendrimers can be used as soluble catalysts for various or-
ganic reactions. There are a few reports in the literature:
aryl–nickel(II) dendrimers have been used as an effective
catalyst for the Karasch addition reaction of polyhaloalk-
anes to olefins,[12] P-based polypalladium complexes were
t2g6eg [low spin (LS)] and t2g4eg [high spin (HS)]. In some
cases, this electronic transition causes a switching of colour
of the complexes with respect to its transition state. The
electron transition can be triggered by using external per-
turbations such as change of temperature, pressure, light
irradiation and magnetic field.[16a,18,19] There are numerous
examples of the SCO behaviour of organometallic com-
plexes using various ligands but very few reports about
using dendritic ones. Still, it is quite challenging to design
and synthesize novel ligands that can show synergic effects
in terms of thermal and spin-crossover properties. Among
the reported wide range of ligand systems, 4-R-substituted
1,2,4-triazoles were the most studied, in which the func-
tional R group may be any kind of aliphatic or aromatic
0
2
[‡] Part I: Ref.[21]
[a] Institut für Anorganische Chemie und Analytische Chemie,
Johannes Gutenberg Universität Mainz,
Staudingerweg 9, 55099 Mainz, Germany
Fax: +49-6131-3922990
E-mail: guetlich@uni-mainz.de
[b] Laboratory of Polymer Chemistry, Department of Materials,
Swiss Federal Institute of Technology, ETH Zürich, HCI J541
8093 Zürich, Switzerland
E-mail: ads@mat.ethz.ch
[c] University of Silesia, Institute of Physics,
ul. Uniwersytecka 4, 40007 Katowice, Poland
View this journal online at
wileyonlinelibrary.com
3930
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
organic moiety. Its combination with iron to provide the complexes because of differences in the connection between
FeIIN6 spin-crossover chromophore is the most attractive the triazole ring and the dendron (NH–C=O and CH2 spac-
one due to its relatively high chemical stability and striking ing). Some important rules for the design and the synthesis
colour change upon spin transition[20] that originates either of the dendritic ligands and their iron(II) complexes with
from the dd HS/LS bands or the more intensive metal-to- magnetic thermal properties are discussed in this paper.
ligand charge-transfer (MLCT) bands, which in many cases
overlap the thermochromic effect associated with the dd
HS/LS bands.
Results and Discussion
Earlier we published the first part of this study using
various triazole-based ligand systems combined with
iron(II) and its spin transition.[21] In this second part, our
main focus is the utilization of various triazole-based den-
dritic ligands (as shown in Figure 1) for making iron(II)
complexes with different counteranions to arrive at a struc-
ture–property relationship with respect to its spin-crossover
behaviour. We have synthesized and characterized various
triazole dendritic ligands and their iron(II) complexes, all
the while keeping SCO behaviour as the main objective.
This study is an extension of our earlier work[21] and also a
comparative study with that of some of the previously re-
ported dendritic iron(II) complexes by Fujigaya et al. (see
structure B in Figure 1).[22] They highlighted a generation
number (n) dependency of the abruptness of the spin-state
change with temperature and showed that [(G1-trz)Fe] (trz
= triazole) was the best-behaved complex in terms of the
spin-crossover properties (cooperativity) in terms of event-
ual applications.[21] We have chosen different counteranions
Synthesis of the Dendritic Triazole Ligands
The first-generation dendritic triazole G1-BOC (C), was
synthesized by heating to reflux the corresponding tert-
butoxycarbonyl (BOC)-protected dendron D[26] with 4-
amino-1,2,4-triazole (atrz) in the presence of diphenyl (2,3-
dihydro-2-thioxo-3-benzooxazolyl) phosphonate (DBOP)
as a catalyst in a mixture of triethylamine and dry THF
under nitrogen (Scheme 1). The reaction was worked up by
using standard extraction techniques and the compound
was isolated from the organic phase. The crude product was
purified by column chromatography and the final com-
pound was obtained in 72% yield. The ligands A and B
were synthesized according to previously reported publica-
tions of Sonar et al.[21] and Fujigaya et al.,[22] respectively.
The dendritic triazoles in Figure 1 were prepared on the
gram scale as analytically pure compounds. The purity of
the materials was confirmed by elemental analysis,
MALDI-TOF spectrometry and NMR spectroscopy.
–
–
such as triflate (CF3SO3 ) and tosylate (p-CH3PhSO3 ) be-
cause of their interesting influences on spin-crossover be-
haviour.[23–25] The G1-PBE [PBE = poly(benzyl ether)] den-
dritic triazole-based complexes (see structure A in Figure 1)
are already covered in our earlier publication. In this paper,
we are mainly focussing on two dendritic triazole ligands
(structures B and C in Figure 1) with different dendritic
branches attached to triazole and their iron(II) complexes.
Structural differences of dendritic ligands play a crucial role
in “tuning” the thermal and magnetic properties of iron(II)
Scheme 1.
Synthesis of the Iron(II) Complexes with Dendritic
Triazoles
All the iron(II) complexes with dendritic triazoles,
[Fe(G1-BOC)3](triflate)2·xH2O (1), [Fe(G1-BOC)3](tosyl-
ate)2·xH2O (2), [Fe(G1-DPBE)3](triflate)2·xH2O {3; G1-
DPBE
= 3,5-bis(3,5-didodecaoxybenzyloxy)-N-[1,2,4]tri-
azol-4-ylbenzamide}, [Fe(G1-DPBE)3](tosylate)2·xH2O (4)
and [Fe(G1-DPBE)3](BF4)2·xH2O (5), were synthesized in
high yield. Ascorbic acid was used as antioxidant during
the synthesis of all complexes to avoid the oxidation of
iron(II) ions. The reactions were carried out using methanol
and THF solvent mixtures (1:5) at room temperature. After
completion of the reaction, we did not observe any precipi-
tate formation in the solvent mixture, which may be due
to similar solubility behaviour of dendritic triazole and its
complexes at room temperature. Due to this reason, we
could not purify the final iron(II) complexes, as it was diffi-
cult to separate the products by precipitation from the reac-
tion solution. After modifying the synthesis procedure, we
Figure 1. Perspective molecular structures of different dendritic tri-
azoles: G1-PBE (A),[21] G1-DPBE (B) and G1-BOC (C).
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3931

Y. Wei, P. Sonar, M. Grunert, J. Kusz, A. D. Schlüter, P. Gütlich
FULL PAPER
were able to separate the pure dendritic triazole iron(II) = 3.281°. This temperature effect was also noticed for other
complexes at very low temperature. Specifically for the G1- small diffraction peaks at the positions 5.812, 6.729, 12.386
BOC-based iron(II) complexes, when the solvent of the re- and 13.715° (Figure 2, b). When the sample is heated again
action mixture was partially removed, the remaining reac- to the original temperature at 300 K, the position of the
tion solution segregated easily into two layers. After cooling strong signal changes only a little (to 2θ = 3.243°). Other
in liquid nitrogen to around –100 °C, the pink solid G1- small peaks appear at 5.788, 6,738, 12.334 and 13.718° (Fig-
BOC-based iron(II) complex was recovered as a product at ure 2, c), respectively. All these signals can be indexed as
the bottom of the flask. The same procedure was carried (1,0), (1,1), (2,0), (3, 1) and (4,0) and belong to a two-di-
out several times to get higher yields. For other dendritic mensional hexagonal lattice with lattice parameters a equal
triazole-based iron(II) complexes, we used the same pro- to 29.71, 31.15 and 31.51 Å for 300, 250 and 300 K, respec-
cedure as mentioned above and it was much easier to obtain tively. After heating the iron(II) complexes again to 350 K,
pure complexes in higher yields.
the complex shows a strong diffraction peak at 2θ = 3.247°
and only two small peaks appear at positions at 5.805 and
6.743° (Figure 2, d). During the heating and cooling cycles,
we observed the irreversibility of the appearances of the
peak. This is due to the loss of water after heating to 350 K
Powder X-ray Diffraction Experiments of the Dendritic
Triazole Iron(II) Complexes
The powder X-ray diffraction experiments were carried [see the thermogravimetric analysis (TGA) and differential
out at various temperatures (first at 300 K, then at 250, 300, scanning calorimetric (DSC) measurements below]. By
350 and again at 300 K, respectively) using fresh samples of cooling again to 300 K, the strong peak was observed at
[Fe(G1-BOC)3](tosylate)2·xH2O (2) and [Fe(G1-DPBE)3]- position 2θ = 3.306° and two small peaks remained at posi-
(tosylate)2·xH2O (4) to compare the effect of different den- tions 5.906 and 6.845° (Figure 2, e). These peaks can be
dritic triazoles. The X-ray powder diffraction pattern shows indexed as (1, 0), (1, 1) and (2, 0) and belong to a two-
in each case a single strong diffraction peak beside some dimensional hexagonal lattice with lattice parameters a
small diffraction peaks. The broad halo peak appeared at equal to 31.47 and 30.91 Å for 350 and 300 K, respectively.
about 2θ–20° and originated from the voluminous dendritic
The G1-DPBE-based iron(II) complex [Fe(G1-DPBE)3]-
branches attached to triazole. It was not possible to deter- (tosylate)2·xH2O (4) was also analyzed by X-ray powder
mine the crystal structures because only eight diffraction diffraction under similar conditions to the first sample mea-
peaks of the poorly structured complexes under study could surement. (Figure 3). The X-ray diffraction pattern was first
be recorded.
analyzed at 300 K using a fresh sample of 4. The peak ap-
Figure 2 shows the X-ray powder diffraction pattern peared at 2θ = 4°; other small peaks were also observed at
measured in the 2θ range between 2 and 20° at various tem- various positions at 6.199, 6,988, 8.405 and 12.382°. These
peratures for [Fe(G1-BOC)3](tosylate)2·xH2O (2). As shown can be indexed in two-dimensional hexagonal lattice as (2,
in Figure 2 (a), the diffraction pattern at 300 K shows the 1), (3, 0), (3, 1) and (4, 2), respectively (Figure 3, a). On
strong peak at 2θ = 3.440°, whereas some other small peaks cooling this sample to 250 K, the small peaks appeared at
appear at the positions of 2θ = 5.975, 6.912, 12.546 and positions 4.674, 6,209, 7.032, 8.422 and 12.425°, which can
13.874°. After cooling the samples to 250 K, the strong dif- be indexed as (2,0), (2,1), (3,0), (3,1) and (4,2), respectively
fraction signal changes its position from 2θ = 3.440° to 2θ (Figure 3, b). After heating the sample again to 300 K, the
Figure 2. X-ray powder diffraction pattern of a fresh sample of 2 measured at (a) 300, (b) 250 and (c) 300 K after cooling and (d) 350
and (e) at 300 K after heat treatment. The vertical line shows the changes in the 2θ position of diffraction peak (1).
3932
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
Figure 3. X-ray powder diffraction pattern of a fresh sample of 4 measured at (a) 300, (b) 250 and (c) 300 K after cooling and (d) 350
and (e) at 300 K after heat treatment. The vertical line shows the changes in the 2θ position of the diffraction peak.
small peaks appeared at positions 4.131, 4.767, 6.338, 7.171 χMT value initially remains constant and then, at about
and 8.543° and this is indexed as (1,1), (2,0), (2,1), (3,0) and 210 K, starts to lower more gradually compared to the
(3,1), respectively (Figure 3, c). During the next heating cy- first heating process. It finally reaches
a value of
cle from 300 to 350 K, the complex showed a strong peak 0.28 cm3 Kmol–1 at 1.7 K. The spin-transition temperature
at 2θ = 2.476°. Small peaks at this temperature were ob- TSC (at which 50% of all complex molecules actively in-
served at positions 4.262, 4.909, 6.516, 7.376 and 8.695°, volved in the thermal spin transition have changed the spin
which can be indexed as (1,0), (1,1), (2,0), (2,1), (3,0) and states from high spin to low spin on cooling) of complex 1
(3,1), respectively (Figure 3, d). After cooling the sample is about 30 K. After the first heating process, the χMT
from 350 to 300 K, the first strong peak appeared at posi- curves in both cooling and heating directions are com-
tion 2θ = 2.458° and the small peaks appeared at positions pletely reversible and no hysteresis was observed. The shape
4.228, 4.926, 6.485, 7.341 and 8.793°. The smaller peaks of the χMT curves in this case, and somewhat similar also
can be indexed as (1,0), (1,1), (2,0), (2,1), (3,0) and (3,1), in the cases of Figure 4 (b and d), resemble very much the
respectively (Figure 3, e).
curves for thermal variation of the high-spin fraction in
All visible peaks can be indexed in a two-dimensional SCO systems with low-spin transition temperatures (i.e.,
hexagonal lattice with lattice parameters a equal to 43.64, with small energy gaps between HS and LS states). It has
43.56, 42.68, 41.14 and 41.50 Å for 300, 250, 300, 350 and been proposed that in such cases the SCO characteristics,
again 300 K, respectively. Similar results were obtained for or rather the magnetic response functions χMT, may be es-
the liquid-crystalline dendritic triazole iron(II) complexes sentially determined by the molecular vibrations due to the
reported by Fujigaya[22] and Seredyuk.[27] Our earlier paper closeness of the effective vibrational gap to the electronic
on this work also reported some common observations.[21]
gap.[28]
The magnetic properties of 2 were studied following the
temperature sequence of 200Ǟ350Ǟ6Ǟ350 K. The χMT
versus T values are shown in Figure 4b. In the first heating
process from 200 to 350 K, the χMT values increase
from 1.7 (at 200 K) to 2.45 cm3 Kmol–1 (at 305 K). In the
temperature range 280–305 K, the χMT values change
sharply and merge into a plateau with a χMT value of
2.8 cm3 Kmol–1. After this first heating process, the χMT
versus T curve shows a gradual spin transition upon cool-
ing from 350 to 6 K, at which point the χMT value reaches
0.5 cm3 Kmol–1. The second heating curve from 6 to 350 K
matches well with the cooling curve but cannot reproduce
the same sharp transition that was observed during the first
heating process.
Magnetic Measurements of the Dendritic Triazole Iron(II)
Complexes
The magnetic properties of 1 were investigated with a
superconducting quantum interference device (SQUID)
magnetometer at variable temperatures following the se-
quence 150Ǟ350Ǟ1.7Ǟ350 K. The results are plotted as
χMT versus T in Figure 4 (a). For a first heating process
from 150 to 250 K, the χMT curve shows a gradual
increase in the χMT values. After heating above 250 K, the
curve ends in a plateau with an χMT value around
2.8 cm3 Kmol–1, which is about 20% below that expected
for the spin-only value of iron(II) in the high-spin state. On
cooling, the χMT values match perfectly those of the heating
The magnetic properties of the freshly synthesized
branch in the temperature between 350 and 250 K. When iron(II) complex 3 were characterized by following the tem-
the temperature is lowered from 250 K down to 1.7 K, the perature sequence of 80Ǟ350Ǟ5.5Ǟ350 K; the results
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3933

Y. Wei, P. Sonar, M. Grunert, J. Kusz, A. D. Schlüter, P. Gütlich
FULL PAPER
Figure 4. Thermal dependence of χMT of complexes (a) 1, (b) 2, (c) 3, (d) 4 and (e) 5.
are shown in Figure 4 (c). The χMT values increase mildly matches very well that of the cooling mode. The χMT values
with increasing temperature in the region 80 to 350 K, start- start rising sharply between 5.5 and 50 K. Further heating
ing from an χMT value of 0.85 cm3 Kmol–1 at 80 K. Appar- leads to much more gradual spin transition until the sample
ently, the thermal spin transition is very gradual due to is heated up to 300 K. During this cycle the χMT values
weak cooperativity between the complex molecules in this change from 1.0 to 2.3 cm3 Kmol–1. Another sharp transi-
material. After crossing the temperature from 300 K on- tion appeared between 300 and 317 K, at which the χMT
ward, the spin state changes very sharply and merges into values change from 2.3 to 2.8 cm3 Kmol–1. A hysteresis
a plateau with an χMT value of 2.8 cm3 Kmol–1 at 320 K. loop of around 15 K width has been observed in the tem-
Upon cooling from 350 to 5.5 K, the χMT value remains perature region between 270 and 300 K.
nearly constant between the 350 and 300 K zone but on
The magnetic characteristics of 4 were measured in the
further cooling down to 280 K, χMT decreased rather grad- temperature sequence of 200Ǟ350Ǟ6Ǟ350 K, as shown
ually from 2.7 to 2.2 cm3 Kmol–1. At 50 K an χMT value of in Figure 4 (d). During the first heating step from 200 to
1.0 cm3 Kmol–1 was recorded. Below 50 K the χMT value 350 K, this iron(II) complex shows an extremely sharp spin-
dropped down to 0.35 cm3 Kmol–1 at 5.5 K most likely due state change from the low-spin to the high-spin state. The
to the well-known zero-field-splitting effect. In the second χMT values change from 0.15 to 2.9 cm3 Kmol–1 within a
heating process from 5.5 to 350 K, the transition curve relatively narrow temperature range of 290 to 350 K. Dur-
3934
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
ing the subsequent cooling and heating processes, the χMT
The Mössbauer spectrum of complex 2 recorded at 80 K
versus T curve is much more gradual relative to the first is displayed in Figure 5 (b). The fitted parameters are δ =
heating process. As the temperature is lowered from 350 to 0.54 mms–1 and ∆EQ = 0.31 mms–1 with reference to the
300 K, χMT first remains constant and then decreases more low-spin state and δ = 1.16 mms–1 and ∆EQ = 3.42 mms–1
strongly between 330 and 300 K; the χMT values decrease for the high-spin state of the iron(II) sites. These parameter
from 2.4 to 1.7 cm3 Kmol–1. With a further decrease in tem- values are typical for LS and HS iron(II) sites in FeIIN6
perature down to 6 K, the χMT values decrease to core complexes. The area fraction of about 35% of HS in
0.84 cm3 Kmol–1, thereby revealing an incomplete HSǞLS the sample of 2 at 80 K is much lower than that of 1 at the
conversion. The drop of χMT at temperatures below 50 K same temperature. This can be explained by the hydrogen-
is again most likely due to the occurrence of zero-field split- bonding interaction being more dominant in compound 2
ting of the remaining HS iron(II) ions. There was no obvi- than in 1, which plays a crucial role in stabilizing the LS
ous difference in the χMT values between the cooling and state.[29] A structural proof was given by single-crystal X-ray
heating curves and hysteresis was not observed.
diffraction on a trinuclear triazole iron(II) SCO compound.[30]
Figure 4 (e) shows the thermal dependence of the χMT
The Mössbauer spectra of 3 (fresh sample) were recorded
values of the iron(II) complex 5. The magnetic properties at 80 and 4 K, respectively, and the fitted spectra are shown
were measured in the temperature regime between 400 and in Figure 5 (c). The parameters obtained from the fitting
4 K following the sequence of 330Ǟ4Ǟ400 K. At room procedure of the 80 K spectrum are δ = 0.49 mms–1 and
temperature, the χMT value is 2.5 cm3 Kmol–1, which points ∆EQ = 0.21 mms–1 for the LS state and δ = 1.19 mms–1 and
to the presence of a small fraction of low-spin molecules. ∆EQ = 3.30 mms–1 for the HS state of iron(II). The area
Cooling the sample leads to a gradual χMT curve with val- fraction AHS/AHS+LS comes out to be γ = 0.36 at 80 K and
ues of 1.5 cm3 Kmol–1 at 50 K and 0.44 cm3 Kmol–1 at 4 K. only slightly lower with γ = 0.29 at 4 K.
The χMT versus T response is completely reversible. During
The Mössbauer spectra of 4 (fresh sample) were also re-
these measurements we did not find any plateau despite corded at 80 and 4 K, as shown in Figure 5 (d). At 80 K
heating the sample up to 400 K. The χMT value at 400 K the spectrum shows a major doublet (with area fraction of
was around 2.7 cm3 Kmol–1.
94%) with an isomer shift of 0.47 mms–1 and a quadrupole
From all these measurements of the dendritic triazole splitting of 0.21 mms–1, which is assigned to the iron(II) LS
complexes under study, it is quite clear that the magnetic sites. The minor contribution of 6% with isomer shift of
behaviour depends on various factors such as the nature 1.16 mms–1 and a large quadrupole splitting of 3.50 mms–1
of the counteranions, the structure of the dendrons attached refers to high-spin iron(II) sites. Corresponding to the
to the triazoles and the type of the spacer between the SQUID data, the value of 0.15 cm3 Kmol–1 (at 200 K) of a
triazole nitrogen and the dendron (NH–C=O and CH2 fresh sample at low temperature probably arises from a
spacing).
small fraction of HS sites located at hydrated chain-end po-
sitions.[22,31] Therefore, the average degree of polymerization
(Dp) can be estimated to be about 37 using the equation Dp
[27]
= 2χTHS/χTLS
.
However, with Mössbauer spectroscopy
57Fe Mössbauer Spectroscopy Study of the Dendritic
Triazole Iron(II) Complexes
data it is possible to calculate Dp at very low temperatures,
at which the zero-field splitting prevails in many instances
The measured Mössbauer spectra of all dendritic triazole and vitiates the determination of the magnetic susceptibili-
iron(II) complexes under study are shown in Figure 5. The ties that originate from the spin transition. It turned out
corresponding hyperfine parameters obtained from least- that the 6% of iron(II) in the HS state determined by Möss-
squares fitting to Lorentzians are collected in Table 1.
bauer spectroscopy at 80 K had converted totally (to less
A representative Mössbauer spectrum of complex 1 re- than two percent, which is approximately the detection limit
corded at 80 K is shown in Figure 5 (a). The spectrum of Mössbauer spectroscopy) into low spin upon further
shows two quadruple doublets. The best fits could be ob- cooling; the spectrum recorded at 4 K contains no more
tained with isomer shift δ = 0.55 mms–1 and quadrupole resonances of FeII in the HS state. By taking an error limit
splitting ∆EQ = 0.28 mms–1 for the inner doublet, which is of maximum 2% in the high-spin signal into account, we
typical for iron(II) in the low-spin state, whereas the param- can estimate that the average chain length would contain
eter values δ = 1.17 mms–1 and ∆EQ = 3.35 mms–1 for the about one hundred iron(II) centres.
doublet that consists of the outer two lines that are assigned
An estimate of Dp was possible only for complex 4, be-
to the high-spin state of the iron(II) site. The large quadru- cause it was the only one among the five compounds under
pole splitting mainly arises from a noncubic valence elec- study that showed an extended plateau in the χMT versus T
tron contribution to the electric field gradient and calls for plot of the low-spin phase. For the other complexes, it can-
a compressed octahedral coordination sphere around not be decided whether the residual high-spin species in the
iron(II) sites. The area ratio AHS/AHS+LS (AHS = area of the low-temperature region arise mainly from iron(II) com-
HS doublet; AHS+LS = total area of the HS and LS dou- plexes at the end of the iron–triazole chains or whether
blets) shows that still 70% of HS exists in the sample at other influences are operative such as various kinds of de-
80 K. The estimation is based on the assumption that the fects in the crystal lattice that are responsible for the incom-
Lamb–Mössbauer factors for HS and LS states are similar. plete spin transition.
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3935

Y. Wei, P. Sonar, M. Grunert, J. Kusz, A. D. Schlüter, P. Gütlich
FULL PAPER
Figure 5. 57Fe Mössbauer spectra of the iron(II) dendritic complexes (a) 1, (b) 2, (c) 3, (d) 4 and (d) 5 at the temperatures indicated in
the figures.
Table 1. 57Fe Mössbauer spectrum parameters for all the iron(II) complexes under study.
Complexes
Low-spin state
∆ELQS
High-spin state
∆EHQS
T
[K]
δ
Γ/2
δ
Γ/2
AHS/AHS+LS
[mms–1]
[mms–1]
[mms–1]
[mm/s]
[mms–1]
[mms–1]
[%]
[Fe(G1-BOC)3](triflate)2·xH2O (1)
[Fe(G1-BOC)3](tosylate)2·xH2O(2)
[Fe(G1-DPBE)3](triflate)2·xH2O (3)
[Fe(G1-DPBE)3](triflate)2·xH2O (3)
[Fe(G1-DPBE)3](tosylate)2·xH2O (4)
[Fe(G1-DPBE)3](tosylate)2·xH2O (4)
[Fe(G1-DPBE)3](BF4)2·xH2O (5)
80
80
80
4
80
4
0.55
0.54
0.49
0.49
0.47
0.48
0.53
0.28
0.31
0.21
0.21
0.21
0.21
0.31
0.14
0.14
0.15
0.17
0.13
0.15
0.15
1.17
1.16
1.19
1.17
1.16
–
3.35
3.42
3.30
3.26
3.50
–
0.18
0.17
0.21
0.24
0.16
–
70
35
36
29
6
0
49
80
1.19
3.35
0.19
A sample of complex 5 showed nearly half of low-spin 1.19 mms–1 and ∆EQ = 3.35 mms–1. It is worth noting that
and high-spin states at 80 K as can be distinguished from the relatively large quadrupole splitting of 3.35 mms–1 for
Figure 5 (e). The parameters for the complex molecules in HS-FeII sites, which seems to arise mainly from a noncubic
the LS state are δ = 0.53 mms–1 and ∆EQ = 0.31 mms–1, distribution of valence electrons around the iron(II) centres,
whereas for those in the HS state these values are δ = points to a relatively high symmetry (close to Oh) of the
3936
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
FeN6 chromophore despite the bulkiness of the dendritic curves of 1 and 2 show similar weight-loss profiles (see Fig-
ligands. There appears to be little perturbation of the FeN6 ure 6, a and b). Both complex compounds started to show
cores that leads to a rather small lattice contribution to the weight loss at around 50 °C and had weight loss of about
electric field gradient, which usually opposes the valence 5 and 6% when the temperature reached around 100 °C.
electron contribution.
The weight can be attributed to the loss of about five and
six molecules of water from the iron(II) complexes 1 and 2.
In each complex there are two regions of sharp weight loss
above 150 °C, which can be attributed to the decomposition
of the alkyl chain of the complexes. During water release,
Thermal Properties of the Dendritic Triazole Iron(II)
Complexes
The TGA and DSC curves of the iron(II) complexes 1 to the DSC curves show strong endothermic peaks around
5 are presented in Figures 6 and 7, respectively. The TGA 70 °C and one further peak at about 150 °C. Both signals
Figure 6. TGA data of the iron(II) dendritic complexes (a) 1, (b) 2, (c) 3, (d) 4 and (e) 5.
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3937

Y. Wei, P. Sonar, M. Grunert, J. Kusz, A. D. Schlüter, P. Gütlich
FULL PAPER
Figure 7. DSC data of the iron(II) dendritic complexes (a) 1, (b) 2, (c) 3, (d) 4 and (e) 5.
are of rather complicated shape, which points to the occur- namically more stable configuration. Therefore, the ther-
rence of a combination of water release and structural re- mally induced release of noncoordinated water molecules
arrangement. The latter peak around 150 °C might be due alters markedly the magnetic behaviour of these complex
to decomposition of the compound. After the first heating compounds.
process, there were no endothermic or exothermic peaks de-
In the first heating branch of complex 3, there were two
tected in the range between 50 and 250 °C. Only a small sharp endothermic peaks in the DSC curve observed at 40
change in the slope of the curve can be recognized around and 50 °C (Figure 7, c). These changes are in accordance
150° for compound 1 and at about 170 °C for compound 2, with the sharp increases in the χMT versus T curve in the
which probably indicates small structural changes due to same temperature region. After the first warming, the χMT
alkyl-chain-length reduction. The calculation of the en- versus T curve is irreversible because of loss of crystal
thalpy and entropy of the heating and cooling processes water. This complex shows about 1.3% loss of weight before
was not possible due to the overlap of the signals in the decomposition of the sample starts above 250 °C, as seen in
calorimetric data.
the TGA curve (Figure 6, c). The declining mass of the
The results obtained from the thermal analysis are in iron(II) complex assigns a loss of about 2.7 molecules of
agreement with the interpretation of the magnetic proper- water. During the second cycle of DSC, the calorimetric
ties. Water loss detected by TGA indicates a drastic change feature showed much weaker endothermic and exothermic
of spin state in the magnetic response function χMT versus peaks. The different temperature peak maxima in the sec-
T, which is irreversible due to the loss of noncoordinated ond cycle of DSC can be attributed to the observed small
water. The complex signals around 70 °C observed by hysteresis in the χMT curves. In this case, it seems that the
DSC further confirm the irreversible nature due to struc- change of spin state is a direct consequence of the release
tural rearrangement of the fresh sample into a thermody- of water and would probably not occur without it. There-
3938
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
fore, the increasing χMT values are not an indication of aforementioned explanation. It is conceivable that the struc-
thermal spin transition, but moreover an irreversible change ture of a thermally cycled sample has a different hydrogen-
of spin state.
bonding network than a sample before thermal treatment.
In the case of 4, the effect of loss of water on the spin For the complex 4, the lattice parameter a is bigger for the
state is even more pronounced relative to the other com- LS state, which is consistent with similar compounds re-
plexes. The TGA data (Figure 6, d) signal the loss of water ported in the literature.[33] Complex 5 shows a significant
around 60 °C, which amounts to 1.6 molecules of water per loss of water below 350 K; it is therefore difficult to
formula unit. As an immediate consequence, this complex correlate the lattice parameter a with the spin-state
shows abrupt and irreversible change of spin state from low change.
spin to high spin at about 60 °C. The DSC data did not
According to the results of the TGA measurements, the
show any noticeable change, which is most likely due to the content of noncoordinated water molecules per formula
fact that an extra dried sample was used particularly for unit is different in all the complex compounds under study
this DSC analysis. After the synthesis of this complex, the and amounts to x = 5 in 1, x = 6 in 2, x = 2.7 in 3, x = 1.6
sample was dried under high vacuum for a long time. The in 4 and x = 1 in 5. As these numbers are only approximate
TGA curve of 5 is shown in Figure 6 (e). The gradual estimates, which most likely vary from one preparation to
weight loss of the complex is due to evaporation of the sol- another of one and the same compound, we have preferred
vent from the complex. The weight loss between 200 and not to articulate the water content in the chemical formulae
250 °C is related to the decomposition of the compound. of the five systems under study. The common feature in all
The water content of the fresh complex is less than 1%, cases, however, is the more or les dramatic change of spin
which corresponds to the loss of less than one molecule state on losing water by warming the substance.
of water per formula unit. A rapid loss in weight due to
Fujigaya et al. published[22] a report on (Gn-trz)Fe-based
decomposition of the complex was observed in the tempera- (n = 0–2) dendritic triazole complexes that showed spin-
ture range of 250–380 °C. The DSC curve of 5 showed transition dependence on the generation number (n) of the
moderate and broad (∆T of around 50 °C) exothermic and dendritic unit. They observed a similar behaviour for their
endothermic peaks at about –30 °C during heating and (G1-trz)Fe complexes with a most abrupt spin transition.
cooling cycles. These peaks not only arise from a HSǞLS We also selected the same triazole dendritic ligands for
transition but also presumably from structural rearrange- making iron(II) complexes with different counteranions and
ments.[32]
tried to correlate the influence of various anions on the
From all these observations it is clear that in each series spin-transition behaviour. From the behaviour of the χMT
of complexes, the abruptness of spin-state changes is more versus T function, it is clear that noncoordinated water mo-
pronounced in the freshly prepared tosylate-based iron(II) lecules play a crucial role. Attention should be drawn to the
complexes than in the triflate compounds. The only excep- possibility that the ligand influences strongly depend on
tion is 5, which showed a gradual change of spin state. This how the noncoordinated water interacts with the ligands
is most likely due to the absence or little content of nonco- through hydrogen bonding. It is known that if water is hy-
ordinated water in this complex. Kahn et al. previously re- drogen-bonded to the ligand through N–H···OH2 interac-
ported this kind of gradual change of spin state as a func- tion close to the coordinating triazole ring of the ligand,
tion of temperature, which they suggested to arise from a the complex tends to stabilize in the LS state; if the hydro-
combination of thermal spin-state change and dehydration gen bonding occurs in the mode N···H2O, the HS state is
process.[23] The dehydrated complex is stable under normal favoured.[29] Comparing the magnetic properties of [Fe(G1-
conditions, and the sharp spin-state change is not reversible DPBE)3](A)2·xH2O (A = triflate and tosylate) and the com-
after the heating process. The influence of noncoordinated plexes [Fe(G1-BOC)3](A)2·xH2O (A = triflate and tosylate),
water on the spin state is much more pronounced in the G1- the NH groups indeed exist in both structures (G1-BOC
DPBE derivatives in which the difference in the magnetic and G1-DPBE complexes), whereas the earlier studied G1-
behaviour between a fresh sample and a heated one is more PBE derivatives do not have such NH groups. The influence
evident than in the others. For example, the freshly pre- of water in both derivatives {[Fe(G1-DPBE)3](A)2·xH2O
pared sample of 4 was diamagnetic at the beginning and (A = triflate and tosylate) and [Fe(G1-BOC)3](A)2·xH2O (A
became paramagnetic after heating. Thus the observed = triflate and tosylate)} seems to be much more predomi-
spin-state change that occurred above room temperature nant than in the earlier studied G1-PBE derivatives because
was driven by the release of the noncoordinated water mole- of the formation of N–H···OH2 hydrogen-bonding interac-
cules upon heating, and this procedure is not reversible. The tion that involves the noncoordinated water molecules.
dehydrated forms of the iron(II) complexes do no show Furthermore, G1-BOC-based dendritic triazole has three –
abrupt spin-state changes around room temperature but NH groups in the structure, among which only one is con-
rather a gradual and reversible thermal spin crossover be- nected to the triazole ring. In this case, the –NH group
low room temperature. This can be understood as a conse- connected to the triazole ring in G1-BOC-based iron(II)
quence of changes of lattice parameters due to loss of water. complexes is less suited to the formation of N–H···OH2 hy-
In this compound the loss of water is particularly small, drogen bonding than the –NH group in G1-DPBE. As a
whereas the changes of the two-dimensional hexagonal lat- result, G1-DPBE derivatives show much more abrupt spin
tice parameters a are remarkably strong, which supports the transition in the first warming procedure. On the contrary,
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3939

Y. Wei, P. Sonar, M. Grunert, J. Kusz, A. D. Schlüter, P. Gütlich
FULL PAPER
immersed in helium gas in a helium cryostat. In this case the used
source was 57Co in a Rh matrix with an activity of about 5 mCi
kept at 4 K. It was impossible to record Mössbauer spectra at room
temperature within acceptable measuring time because of the soft-
ness of the material with too low Debye temperatures. The isomer
shift values are given with reference to α-iron.
the magnetic properties of G1-PBE derivatives show almost
no difference between the fresh sample and the heated
one.
Conclusion
Synthesis of Ligands and Complexes
We have synthesized first-generation dendritic triazole-
based ligands and their iron(II) complexes with different
counteranions. Their thermal, magnetic and spin-transition
properties were studied in detail using various characteriza-
tion techniques. All complexes showed spin-transition be-
haviour that is dependent on the coordinated water mole-
cules present in the iron(II) complex. Spin-state conversion
was triggered by thermally induced water release and a
gradual spin transition took place after a first heat treat-
ment to 350 K. The [Fe(G1-DPBE)3](BF4)2·xH2O (5)
iron(II) complex shows exceptional behaviour due to lack
or little content of crystal water in the initial sample. There-
fore, this particular iron(II) complex did not show any
change in the spin-state change behaviour between initial
and first heating cycle, whereas complexes 1 to 4 clearly
show noticeable changes in magnetic behaviour upon heat-
ing. The influence of noncoordinated water is more pre-
dominant in iron(II) complexes with tosylate than with
triflate anions. Similar effects are also observed for G1-
DPBE-based iron(II) complexes than G1-BOC-based deriv-
atives. The nature of the counteranions and the type of den-
dritic ligands play an important role for the magnetic be-
haviour as reflected by the χMT versus T curve and by vari-
able-temperature Mössbauer spectroscopy.
Synthesis of G1-BOC (A): Triethylamine (1 mL, 6.88 mmol), 4-
amino-1,2,4-triazole (0.89 g, 10.6 mmol) and diphenyl (2,3-dihy-
dro-2-thioxo-3-benzoxazolyl) phosphonate (DBOP) (2.63 g,
6.87 mmol) were added to a solution of AЈ (3.00 g, 6.88 mmol) in
150 mL of THF at room temperature. The mixture was heated at
reflux overnight and allowed to cool to room temperature. The
organic solvent was evaporated and the crude compound purified
by column chromatography using hexane/ethyl acetate mixture as
eluent (gradient from 100:0 to 90:10). The fraction was collected
and the solvents evaporated to dryness to give compound 5 as white
1
powder (yield: 72%, 2.50 g). H NMR (CDCl3): δ = 8.56 (s, 2 H,
Trz), 7.61 (s, 2 H, Ph), 7.33 (s, 1 H, Ph), 4.70 (s, 3 H, NH), 3.09 (t,
4 H, 2CH2), 2.73 (t, 4 H, 2CH2), 1.86 (t, 2 H, 2CH), 1.32 (s, 18 H,
6CH3) ppm. 13C NMR (CDCl3): δ = 168.14, 157.92, 143.65, 134.45,
131.31, 126.14, 79.72, 48.90, 40.35, 33.35, 31.96, 30.82, 28.69,
24.07 ppm. MALDI-TOF calcd. for C25H38N6O5: [M + H]+
502.62; found 502. C25H38N6O5 (502.61): calcd. C 59.74, H 7.62,
N 16.72; found C 59.14, H 7.09, N 16.28.
Synthesis of [Fe(G1-BOC)3](triflate)2·xH2O (1): A solution of
Fe(triflate)2·6H2O (0.13 mmol, 0.062 g) and ascorbic acid [0.014 g;
to prevent iron(II) from oxidation] in methanol (1 mL) were added
dropwise to a solution of G1-BOC (0.4 mmol, 0.2012 g) in meth-
anol (2 mL) at 60 °C while stirring. The reaction mixture was
stirred for one hour at 60 °C and then was cooled down to room
temperature. The solvent was removed under reduced pressure, and
the colourless colloidal substance precipitated on cooling with li-
quid nitrogen. The product was fast-washed with cold MeOH and
THF and dried in vacuo; yield 93%. C77H114F6FeN18O21S2
(1861.81): calcd. C 49.63, H 6.12, N 13.53, S 3.44; found C 47.15,
Experimental Section
Materials and Characterization: Commercially available chemicals
were reagent grade and used without further purification. The syn-
thesis of dendritic ligands were published by Fujigaya et al.[22] and
Sonar et al.[21] 1H (300 MHz) and 13C (62.5 MHz) NMR spec-
troscopy was carried out on a Bruker spectrometer at room tem-
perature. Elemental analysis (C, H, N and S Analyzer) and mass
spectrometry (Ionspec Ultrima spectrometer) techniques were uti-
lized for the confirmation and purity of the samples. 1H-1,2,4-tri-
azole-1-propanenitrile was synthesized from 1H-1,2,4-triazole and
acrylonitrile as described in the literature.[31] Infrared spectra were
recorded by using a Brooker Tensor 27 with samples prepared as
pellets in KBr. DSC measurements were performed on Perkin–El-
mer, Norwalk, Connecticut, whereas TGA analyses were carried
out on TA Instruments, New Castle, Delaware. The elemental
analyses of all complexes were performed on a Vario EL (Elemen-
tal) for C, H, N and S determination. For structural investigations
powder X-ray diffraction were recorded at 300, 250 K and
then again at 300, 350 and 300 K, with Cu-Kα radiation using a
PANalytical X’Pert PRO diffractometer equipped with the Paar
HTK 1200. The temperature-dependent magnetic susceptibilities
were measured on a MPMS SQUID device of Quantum Design.
The magnetic data were corrected for magnetization of the sample
holder and diamagnetic contributions. 57Fe Mössbauer spec-
troscopy was carried out at 80 K using a constant-acceleration con-
ventional spectrometer with a nitrogen cryostat. The source used
was 57Co in a Rh matrix with an activity of about 10 mCi kept
at room temperature. For measurements at 4 K the samples were
H 5.93, N 12.10, S 4.14. IR (KBr pellets): ν = 2978.6 (m, CH ),
˜
3
2936.0 (–CH2–), 758.8, 715.2, 640.1 (m, substituted Ar), 3365.8,
–
1690.2 (s, –CONH–), 1029.8 (s, CF3), 1277.9, 1251.5 (m, R–SO3 ),
1524.5, 1455.1, 1394.2, 1367.4, 1168.8 (triazole) cm–1.
Synthesis of [Fe(G1-BOC)3](tosylate)2·xH2O (2): A similar pro-
cedure as described above was employed to obtain the complex
2, except for the quantity of the Fe(tosylate)2·6H2O salt (0.068 g,
0.13 mmol); yield 96%. C89H128FeN18O21S2 (1906.07): calcd. C
56.03, H 6.72, N 13.22, S 3.36; found C 54.66, H 6.96, N 11.31, S
3.33. IR (KBr pellets): ν = 2977.4 (m, CH ) 2932.6 (–CH ) 684.0,
˜
3
2
625.1 (m, substituted Ar) 3358.0, 1692.5 (s, –CONH–) 1275.1,
–
1251.5 (m, R–SO3 ), 1523.7, 1453.6, 1392.7, 1366.2, 1170.7 (tri-
azole) cm–1.
Synthesis of [Fe(G1-DPBE)3](triflate)2·xH2O (3): A methanol solu-
tion of Fe(triflate)2·6H2O (0.0310 g, 0.067 mmol) that contained a
small amount of ascorbic acid was added dropwise to a solution
of G1-DPBE (0.227 g, 0.2 mmol) in 20 mL of THF at 70 °C. The
reaction mixture was stirred at this temperature for about 30 min,
then cooled down to room temperature. The solvent was partially
removed under reduced pressure until the solution was separated
into two layers. The pink gelatine-form product was cooled with
liquid nitrogen, washed with cold methanol and was dried under
vacuum; yield 89%. C215H348F6FeN12O27S2 (3767.13): calcd. C
68.49, H 9.24, N 4.46, S 1.70; found C 69.13, H 10.27, N 4.40, S
2.39. IR (KBr pellets): ν = 1606.2, 1594.8 (s, Ar), 2921.3, 2853.6
˜
(–CH2–), 813.4, 720.8, 681.3, 639.9 (m, substituted Ar), 1074.0 (s,
3940
www.eurjic.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2010, 3930–3941

Iron(II) Spin-Transition Complexes with Dendritic Ligands, Part II
–
[14] M. Slany, M. Bardaji, A.-M. Caminade, B. Chaudret, J. P. Ma-
joral, Inorg. Chem. 1997, 36, 1939–1945.
C–O), 1699.0 (s, –CONH–), 1157.2 (s, CF3), 1252.8 (m, R–SO3 ),
1465.3, 1377.7, 1349.5, 1327.7, 1174.6 (triazole) cm–1.
[15] M. Enomoto, T. Aida, J. Am. Chem. Soc. 2002, 124, 6099–
Synthesis of [Fe(G1-DPBE)3](tosylate)2·xH2O (4): A similar pro-
cedure was employed to prepare 4 using Fe(tosylate)2·6H2O
(0.0338 g, 0.067 mmol) and G1-DPBE (0.2271 g, 0.2 mmol); yield
93%. C227H362FeN12O27S2 (3811.39): calcd. C 71.47, H 9.50, N
4.41, S 1.68; found C 71.31, H 9.51, N 4.30, S 1.79. IR (KBr pel-
6108.
[16] a) Topics in Current Chemistry, vol. 233–235, Spin Crossover in
Transition Metal Compounds, parts I–III (Eds.: P. Gütlich,
H. A. Goodwin), Springer, Berlin, Heidelberg, 2004; b) P.
Gütlich, A. Hauser, H. Spiering, Angew. Chem. Int. Ed. Engl.
1994, 33, 2024–2054; Angew. Chem. 1994, 106, 2109–2141.
[17] J.-F. Létard, P. Guionneau, L. Goux-Capes, Top. Curr. Chem.
2004, 235, 221–249.
[18] a) O. Kahn, Curr. Opin. Solid State Mater. Sci. 1996, 1, 547–
556; b) P. Gütlich, Y. Garcia, T. Woike, Coord. Chem. Rev.
2001, 219–221, 839–879; c) A. Bousseksou, G. Molnär, J.-P.
Tuchagues, N. Menéndez, E. Codjovi, F. Varret, C. R. Chim.
2003, 6, 329–334; d) J. A. Real, A. B. Gaspar, V. Niel, M. C.
Muñoz, Coord. Chem. Rev. 2003, 236, 121–141; e) A. Boussek-
sou, F. Varret, M. Goiran, K. Boukheddaden, J.-P. Tuchagues,
Top. Curr. Chem. 2004, 235, 65–84 and references cited therein;
f) E. König, Struct. Bonding (Berlin) 1991, 76, 51–152; g) P.
Gütlich, Y. Garcia, H. A. Goodwin, Chem. Soc. Rev. 2000, 29,
419–427.
lets): ν = 1599.3 (s, Ar), 2923.5, 2853.3 (–CH –), 683.7 (m, substi-
˜
2
tuted Ar), 1054.6 (m, C–O) 1696.4 (s, –CONH–), 1217.3 (m,
R–SO3 ), 1466.0, 1378.5, 1348.9, 1326.2, 1164.5 (triazole) cm–1.
–
Synthesis of [Fe(G1-DPBE)3](BF4)2·xH2O (5): The same procedure
as for 4 was employed to prepare 5 using Fe(BF4)2·6H2O (0.0231 g,
0.067 mmol) and G1-DPBE (0.2252 g, 0.2 mmol), and was dried
under vacuum; yield 90%. C223H348B2F8FeN12O21 (3762.72): calcd.
C 70.23, H 9.62, N 4.61; found C 70.28, H 9.50, N 4.41.
Acknowledgments
[19] L. Salmon, A. Bousseksou, B. Donnadieu, J.-P. Tuchagues, In-
org. Chem. 2005, 44, 1763–1773.
[20] P. J. van Koningsbruggen, Top. Curr. Chem. 2004, 233, 123–
149.
[21] P. M. Sonar, C. M. Grunert, Y.-L. Wei, J. Kusz, P. Gütlich,
A. D. Schlüter, Eur. J. Inorg. Chem. 2008, 1613–1622.
[22] T. Fujigaya, D.-L. Jiang, T. Aida, J. Am. Chem. Soc. 2005, 127,
5484–5489.
We are thankful to B. Mathiasch for the performance of infrared
spectroscopy, to M. Colussi for help with DSC and TGA measure-
ments and to Dr. Christian Baerlocher for the XRD measurements
(ETH Zürich). We acknowledge the financial support from the
Deutsche Forschungsgemeinschaft (DFG), Priority Program no.
1137 (“Molecular Magnetism”).
[23] O. Kahn, C. J. Martinez, Science 1998, 279, 44–48.
[24] E. Codjovi, L. Sommier, O. Kahn, C. Jay, New J. Chem. 1996,
20, 503–505.
[25] Y. Garcia, P. J. van Koningsbruggen, E. Codjovi, R. Lapouy-
ade, O. Kahn, L. Rabardel, J. Mater. Chem. 1997, 7, 857–858.
[26] A. Zhang, L. L. Okrasa, T. Pakula, A. D. Schlüter, J. Am.
Chem. Soc. 2004, 126, 6658–6666.
[27] M. Seredyuk, A. B. Gaspar, V. Ksenofontov, S. Reiman, Y. Ga-
lyametdinov, W. Haase, E. Rentschler, P. Gütlich, Chem. Mater.
2006, 18, 2513–2519; A. B. Gaspar, M. Seredyuk, P. Gütlich,
Coord. Chem. Rev. 2009, 253, 2399–2413.
[28] J.-P. Tuchagues, A. Bousseksou, G. Molnar, J. J. McGarvey, F.
Varret, Top. Curr. Chem. 2004, 235, 85–103.
[1] D. J. Díaz, G. D. Storrier, S. Bernhard, K. Takada, H. D. Ab-
ruña, Langmuir 1999, 15, 7351–7354.
[2] G. R. Newkome, C. N. Moorefield, F. Vögtle, Dendrimers and
Dendrons: Concepts, Syntheses, Applications, VCH, Weinheim,
2001. J. M. J. Fréchet, D. A. Tomalia (Eds.), Dendrimers and
Other Dendritic Polymers, Wiley, Chichester, 2001.
[3] F. Zeng, S. C. Zimmerman, Chem. Rev. 1997, 97, 1681–1712.
[4] D. A. Tomalia, A. M. Naylor, W. A. Goddard III, Angew.
Chem. Int. Ed. Engl. 1990, 29, 138–175.
[5] M. Fischer, F. Vögtle, Angew. Chem. Int. Ed. 1999, 38, 884–
905.
[6] J.-M. J. Fréchet, Science 1994, 263, 1710–1715.
[7] M. Benito, O. Rossell, M. Seco, G. Segalés, V. Maraval, R.
Laurent, A.-M. Caminade, J.-P. Majoral, J. Organomet. Chem.
2001, 622, 33–37.
[29] M. Sorai, J. Ensling, K. M. Hasselbach, P. Gütlich, Chem.
Phys. 1977, 20, 197–208; K. H. Sugiyarto, H. A. Goodwin,
Aust. J. Chem. 1988, 41, 1645–1648; K. H. Sugiyarto, K.
Weitzner, D. C. Graig, H. A. Goodwin, Aust. J. Chem. 1997,
50, 869–873.
[8] E. C. Constable, C. E. Housecroft, M. Neuburger, S. Schaffner,
L. J. Scherer, Dalton Trans. 2004, 2635–2642.
[9] C. Díaz, M. Barbosa, Z. Godoy, Polyhedron 2004, 23, 1027–
1035.
[10] M. T. Reetz, G. Lohmer, R. Schwickardi, Angew. Chem. Int.
Ed. Engl. 1997, 36, 1526–1539.
[11] F. G. A. Jansen, E. M. M. de Brabander-van den Berg, E. W.
Meiser, Recl. Trav. Chim. Pays-Bas 1995, 114, 225–230.
[12] J. W. J. Knapen, A. W. van der Made, J. C. de Wilde,
P. W. N. M. van Leeuven, P. Wijkens, D. M. Grove, G.
van Koten, Nature 1994, 372, 659–663.
[13] a) A. Miedaner, C. J. Curtis, R. M. Barkley, D. L. DuBois, In-
org. Chem. 1994, 33, 5482–5490; b) A. M. Herring, B. D. Stef-
fey, A. Miedaner, S. A. Wander, D. L. DuBois, Inorg. Chem.
1995, 34, 1100–1109.
[30] Y. Garcia, P. Guionneau, G. Bravic, D. Chasseau, J. A. K.
Howard, O. Kahn, V. Ksenofontov, S. Reiman, P. Gütlich, Eur.
J. Inorg. Chem. 2000, 1531–1538.
[31] J. J. A. Kolnaar, G. Dijk, H. Kooijiman, A. L. Spek, V. G.
Ksenofontov, P. Gütlich, J. G. Haasnoot, J. Reedijk, Inorg.
Chem. 1997, 36, 2433–2440.
[32] Y. Garcia, P. J. van Koningsbruggen, R. Lapouyade, L.
Fournès, L. Rabardel, O. Kahn, V. Ksenofontov, G. Levchenko,
P. G ütlich, Chem. Mater. 1998, 10, 2426–2433.
[33] T. Fujigaya, D.-L. Jiang, T. Aida, J. Am. Chem. Soc. 2003, 125,
14690–14691.
Received: January 22, 2010
Published Online: July 14, 2010
Eur. J. Inorg. Chem. 2010, 3930–3941
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
3941