A magnetic iron(III) switch with controlled and adjustable thermal response for solution processing

Article abstract of DOI:10.1039/c2dt12037b

Spin crossover requires cooperative behavior of the metal centers in order to become useful for devices. While cooperativity is barely predictable in solids, we show here that solution processing and the covalent introduction of molecular recognition sites allows the spin crossover of iron(III) sal ₂trien complexes to be rationally tuned. A simple correlation between the number of molecular recognition sites and the spin crossover temperature enabled the fabrication of materials that are magnetically bistable at room temperature. The predictable behavior relies on combining function (spin switching) and structure (supramolecular assembly) through covalent interactions in a single molecular building block. The Royal Society of Chemistry 2012.

Full text of DOI:10.1039/c2dt12037b

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 3726
www.rsc.org/dalton
PAPER
A magnetic iron(III) switch with controlled and adjustable thermal response
for solution processing†
Claudio Gandolfi,a Grace G. Morganb and Martin Albrecht*a,b
Received 26th October 2011, Accepted 16th January 2012
DOI: 10.1039/c2dt12037b
Spin crossover requires cooperative behavior of the metal centers in order to become useful for devices.
While cooperativity is barely predictable in solids, we show here that solution processing and the covalent
introduction of molecular recognition sites allows the spin crossover of iron(III) sal2trien complexes to be
rationally tuned. A simple correlation between the number of molecular recognition sites and the spin
crossover temperature enabled the fabrication of materials that are magnetically bistable at room
temperature. The predictable behavior relies on combining function (spin switching) and structure
(supramolecular assembly) through covalent interactions in a single molecular building block.
the iodine content of a porous coordination polymer comprised
Introduction
of iron(II), pyrazine, and Pt(CN)4 units.12
Molecular entities that possess two magnetically stable states are
presumed to be the key component for the next generation of
data storage and processing devices.1 While spin transition and
thermal addressing of magnetic ‘on’ and ‘off’ states at the mol-
ecular level is not uncommon,2 the designed fabrication of spin–
labile materials is often precluded by the complexity of the
numerous and intricate inter- and intramolecular interactions that
govern the spin crossover.3 As a consequence, the spin transition
of the vast majority of spin–labile compounds is gradual and
occurs over a broad (>100 K) temperature range. An abrupt tran-
sition with hysteresis loop is, however, desired for device oper-
ations1,4 and requires the metal centers to interact cooperatively.5
Up to now, only few molecular systems are known that exhibit a
strong cooperativity.2a,6
Cooperativity typically results from a combination of intra-
and intermolecular interactions such as hydrogen bonding,
π-anion, or dipolar interactions, which are generally weak.7
Therefore, even minute changes for example in the crystallizing
solvent have dramatic effects on the abruptness,8 and tailoring of
materials for specific applications has been severely hampered.9
Because of the difficulties in quantifying collective supramolecu-
lar interactions, not surprisingly, predictive models for spin tran-
sitions are rare.10 The empirical trends established thus far are
typically specific for a single class of compounds only.11 Very
recently, elegant work has demonstrated that the spin transition
temperature can be controlled over a broad temperature range by
Previous approaches to overcome the severe limitations in tai-
loring spin transition have been based on supramolecular prin-
ciples in an attempt to control the intermolecular organization of
spin–labile systems.13 In particular the introduction of amphiphi-
lic properties by charge balancing of cationic spin–labile iron
triazole polymers with lipophilic counterions has afforded dis-
crete nanoparticles with abrupt spin crossover, thus inducing an
abrupt and hysteretic spin crossover in the solution phase.14
While the magnetic activity in these nanoparticles is strongly
related to the spin crossover in analogous solid state Fe(trz)3
materials (trz = 1,2,4-triazole derivative) and is governed by the
same principles with low predictability, we considered that
covalent bonding of the modular lipophilic chain to a spin–labile
center may inherently link the magnetic function to structural
aspects. This approach offers a methodology to induce spin
crossover in the solution phase,3 where magnetic changes are
generally only gradual due to the lack of cooperativity and the
absence of intermolecular interactions between the active sites.15
Here we show that this approach provides a subtle control of the
spin crossover via supramolecular principles.
Results and discussion
Complex 1a featuring a spin–labile iron(III) center in an N4O2
coordination sphere16 was functionalized with alkyl chains R of
different length (Fig. 1).17 Solid samples of complexes 1 were
spin-stable and did not undergo a crossover upon cooling to
30 K.18 Solutions of complexes 1 are red and display an absor-
bance maximum around 495 nm (CH2Cl2) which is diagnostic
for an iron(III) high-spin (HS, S = 5/2) configuration.15 Upon
cooling, all solutions change color to dark blue due to a new
absorbance band at 650 nm, indicating a low-spin (LS, S = 1/2)
species and thus a thermally induced spin crossover.18 In the
solid state, in contrast, all alkyl-functionalized complexes 1b–g
aDepartment of Chemistry, University of Fribourg, Chemin du Musée 9,
1700 Fribourg, Switzerland
bSchool of Chemistry & Chemical Biology, University College Dublin,
Belfield, Dublin 4, Ireland. E-mail: martin.albrecht@ucd.ie; Fax: +353
1716 2501; Tel: +353 1716 2504
†Electronic supplementary information (ESI) available: UV spectra, and
plots of absorption vs. γ′ for different concentrations of 1 and for com-
plexes 2. See DOI: 10.1039/c2dt12037b
3726 | Dalton Trans., 2012, 41, 3726–3730
This journal is © The Royal Society of Chemistry 2012

View Article Online
where γ′HS = γ′LS = 0.5 (γ′ is the fraction of spin-crossover active
molecules in a given spin state). While for complex 1e compris-
ing C18 groups, the transition temperature is 235 K, T1/2 is raised
to 253 K for 1f, and even closer to ambient temperature for 1g
containing C30 tails (T1/2 = 273 K). The latter complex had to be
measured at lower concentration (0.04 mM) due to its low solu-
bility in CH2Cl2. The correlation between chain length and spin
crossover temperature brings changes of magnetization at room
temperature within reach. With the compounds based on 1,
however, further elevation of the spin transition temperature is
prevented by the restricted solubility of complexes that are func-
tionalized with longer alkyl chains. Nonetheless, this series is
unique in allowing trends to be established in spin crossover be-
havior that correlate directly with the transition temperature.12
The isosbestic points in the temperature-dependent UV-vis
spectra suggest an equilibrium between high and low-spin
species. Analysis of the temperature-dependence of this equili-
brium for 1e using the van’t Hoff equation afforded the thermo-
dynamic parameters for the LS to HS transition. Accordingly,
ΔH° = 96.5( 1.1) kJ mol−1 and ΔS° = 412( 5) J K−1 mol−1 in
the 225–240 K temperature range (Fig. S7†). For complexes 1f
and 1g extraction of the thermodynamic data is restricted by the
sharpness of the spin crossover (e.g. 4 K for 1g) and consequen-
tially by the limited amount of data points available for charac-
terizing the crossover.22
Fig. 1 Schematic representation of iron(III) sal2trien complexes functio-
nalized with different lipophilic substituents R.
A decrease of the spin transition temperature is induced upon
lowering the concentration of the spin–labile species. A five-fold
dilution of complex 1e or 1f from 0.2 to 0.04 mM reduced the
spin transition temperature from 235 to 226 K and from 253 to
245 K, respectively. The molar extinction coefficient at 650 nm
is not affected upon dilution, suggesting a complete spin change.
The consistent lowering of the transition temperature by approxi-
mately 10 K demonstrates that the concentration constitutes a
further handle to tailor the magnetic state of the complex at a
given temperature, without compromising the sharp nature of the
spin crossover.
An apparent hysteresis was observed in the first cooling–
heating cycle, but temperature-dependent UV-vis spectroscopy
provided identical curves upon cooling and heating in sub-
sequent cycles. The bistable window is dependent on the length
of the alkyl chain and varies from 6 K (for 1e) to 14 and 17 K
for solutions of complexes 1f and 1g, respectively.23 The remark-
ably large hysteresis of 1g covers the 273–290 K range (Fig. 3a),
Fig. 2 Plot of the high-spin fraction of iron(III) sal2trien complexes,
γ′HS, as a function of temperature as determined by the diagnostic
absorption at 650 nm (all complexes are 0.2 mM CH2Cl2 solutions,
except 1g at 0.04 mM; γ′HS is the molar fraction that is spin-crossover
active). The behavior of complex 1b is identical to that of 1a and 1c and
has been omitted for clarity.†
are spin stable and preserve the HS state between RT and 30 K.19
Monitoring of the extinction coefficients ε500 and ε650 at differ-
ent temperatures allows the relative HS–LS ratio to be esti-
mated.20 Accordingly, relatively large fractions of the complexes
1e–g undergo a spin crossover, whereas in solutions of 1a–d, a
significant portion remains in the HS configuration. Tempera-
ture-dependent analysis of the HS–LS distribution reveals the
presence of an isosbestic point around 580 nm, thus identifying
an equilibrium between HS and LS configurations.†
Solutions of complexes comprising short alkyl chains, viz.
complexes 1b–d, display a gradual spin transition over approxi-
mately 80 K, which may not be complete at the lowest measured
temperature of 170 K. Similar behavior has been reported for the
unfunctionalized complex 1a.15 In sharp contrast, longer alkyl
groups as in complexes 1e–g induce an abrupt HS to LS change
that is essentially complete within a 10 K cooling range
(Fig. 2).21 Variation of the length of the alkyl chain provides a
remarkably clear correlation between alkyl chain length and the
spin crossover temperature T1/2, determined as the temperature
Fig. 3 Cooling–heating cycles for self-assembled complexes in sol-
ution. (a) Hysteresis of 1g around room temperature. (b) Magnetic
response of complex 1g upon sequential cycling between 270 K and
298 K, revealing incomplete population of both the high- and the low-
spin states in cycles 2 and 3. (c) Magnetic response of complex 1f
during three cooling–heating cycles involving repetitive temperature
changes between 230 K and 270 K.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3726–3730 | 3727

View Article Online
a window that is only marginally below room temperature
(20 °C). While such properties may become particularly attrac-
tive for applications, for example for data storage, the robustness
needs improvement. Currently, the hysteresis is only pertained
during the first cooling–heating cycle and disappears in sub-
sequent cycles.9a Moreover, the reversibility of the change in
magnetization is affected with complex 1g. Repetitive cooling–
heating cycles led to a decreased absorption for both the HS and
LS states, indicating incomplete transitions upon recycling
(Fig. 3b). In contrast, the spin crossover process of solutions
containing complex 1f did not indicate any fatigue during three
consecutive cycles (Fig. 3c).
The hysteresis and the abrupt spin transition imply a coopera-
tive action of the spin crossover-active molecules in solution.5
Cryogenic scanning electron microscopy (cryo-SEM) of frozen
CH2Cl2 solutions of complexes 1c–1g indeed displayed a supra-
molecular organization of the complexes, presumably induced
by intermolecular recognition due to the amphiphilic character of
the complexes. All samples showed mutually interwoven fibers
that are approximate 40 15 nm thick (Fig. 4).
spherical particles and hence both the tertiary structural motifs as
well as the fibers seem to co-exist as supramolecular assemblies
in solutions of these complexes. The diameter of the fibers
cannot be rationalized by a simple tubular micelle-type arrange-
ment of the amphiphilic molecules, as the alkyl chain length
does not exceed 3 nm even in a fully stretched conformation of
the molecule. Hence, supercoiling or alternatively the formation
of vesicle-like aggregates may account for the observed fiber
dimensions. In either case, this supramolecular self-assembly in
solution appears to be dependent on the alkyl chain length R and
is assumed to be essential for the abruptness of the spin tran-
sition as well as for the variation in the transition temperature.
Variation of the alkyl chain length and the molar concentration
thus suggests as a simple rule that the spin transition temperature
increases with increasing number of methylene groups as recog-
nition sites available for self-assembly. Such predictive behavior
is supported by experiments with complexes comprising a larger
number of alkyl chains per spin–labile metal center. Complex 2e
containing four alkyl chains per metal center displays an 11 K
higher transition temperature than its homologue 1e featuring
only two alkyl chains. Solution processing may be an essential
prerequisite to achieve such reliable predictability and direct and
rational control of spin transition through self-assembly.
However, remarkably different larger morphologies were
observed dependent on the length of the alkyl chains. While
micrometre-sized superstructures reminiscent of tertiary protein
structures were identified for the complexes with abrupt spin
transition, 1e–1g, no such motif was found with 1c. The spheri-
cal shape of these microsize motifs is most evident in solutions
of compound 1e,18 and slightly less pronounced with 1f (average
Conclusions
We have developed a magnetically switchable system based on
alkyl-functionalized iron(III) sal2trien complexes, for which the
transition temperature can be predictably modulated via supra-
molecular principles. While solution processing has generally
been assumed to suppress spin crossover,13 the iron(III) sal2trien
system exploits the dynamics of self-assembly in solution to
induce cooperativity that is absent in the solid state. Specifically,
the number of CH2 units that are available as molecular recog-
nition sites in solution are a key parameter to control the spin
crossover temperature. These molecular recognition sites require
a minimum distance to the polar spin–labile site to be efficient
for self-assembly. In our case, C18 and longer units are appropri-
ate, while C12 units are too short. Tailoring of the spin crossover
temperature is possible by adjusting the number of recognition
sites per molecule or by variation of the molar concentration.
This methodology has been used to fabricate a solution that is
magnetically bistable around room temperature. Presumably, tai-
loring affects the self-assembly of the molecules and the type of
tubular core–shell motif comprising alkyl groups as shells
around the polar iron centers.13 The assembly in solution is
likely dynamic, yet sufficiently ordered to constrain the confor-
mation of the alkyl tails, thus imparting the intermolecular
elastic strain required for cooperative spin crossover.5c An
obvious advantage of the approach presented here is the covalent
bonding between the molecular recognition sites and the magne-
tically labile metal center, which inherently connects structure
and function. This methodology is applicable to a large diversity
of magnetically bistable systems providing access to custom-tai-
lored magnetically active solutions, which may become useful,
for example for the fabrication of soft matter devices,1 for ther-
mochromically triggered switches, and for early-stage detection
of cancer tissue due to the temperature gradient between healthy
and carcinogenic cells.24
diameter 2.3
0.9 μm). A lesser degree of aggregation into
spherical particles is observed when the solutions are diluted, as
illustrated for 1f and 1g (inset of Fig. 4a and b, respectively).
Hence, the decrease in spin transition temperature upon dilution
might be correlated to the lower tendency of the complexes to
aggregate into spherical motifs. Notably, not all fibers are part of
Fig. 4 Cryogenic scanning electron micrographs of representative sec-
tions of self-assembled species: (a) 1f from 2.5 mM solution and
0.5 mM (inset); (b) 1g from 0.5 mM solution; (c) 1e and (d) 1c from
5 mM solution in CH2Cl2 cryogenic scanning electron micrographs of
representative sections of self-assembled species: (a) 1f from 2.5 mM
solution and 0.5 mM (inset); (b) 1g from 0.5 mM solution; (c) 1e and
(d) 1c from 5 mM solution in CH2Cl2.
3728 | Dalton Trans., 2012, 41, 3726–3730
This journal is © The Royal Society of Chemistry 2012

View Article Online
stirring this mixture for 10 min at RT, 1-bromotriancontane
(502 mg, 1.0 mmol) was added as a solid. The mixture was
heated to 120 °C for 3 h under Ar. After cooling to RT, aqueous
HCl (1 M, 50 mL) was added and the mixture was stirred vigor-
ously for 15 min and filtered and the residue was washed with
water (3 × 50 mL) and acetone (50 mL) and then extracted with
warm THF (4 × 50 mL). A brownish precipitate formed upon
staying overnight at RT, which was filtered off. The filtrate was
concentrated to 20 mL and warmed to obtain a clear solution,
acetone was added dropwise to initiate precipitation. After stir-
ring for 2 h at RT, the formed precipitate was collected by cen-
trifugation. The slow precipitation was repeated three times and
the residue dried in vacuo to give the title compound as a white
solid (110 mg, 20%). Filtration of a warm solution of the title
compound in THF over SiO2 gave microanalytically pure
Experimental procedures
General
The syntheses of 1-bromotriacontane,25 complexes 1a–e,17,18
and 2c–e17 were reported previously, all other reagents were
used as received. For synthesis, THF was dried by passage
through a solvent purification column, all other reagents were
commercially available and used as received. Flash chromato-
graphy was performed using silica gel 60 (63–200 mesh) or
basic alox (0.05–0.15 mm, pH 9.5). All H and 13C{1H} NMR
1
spectra were recorded at 25 °C on Bruker or Varian spec-
trometers and referenced to residual solvent 1H or 13C reson-
ances (d in ppm, J in Hz). Assignments are based either on
distortionless enhancement of polarization transfer (DEPT)
experiments. Melting points were determined using a Mettler
Toledo TGA/SDTA 851 analyzer and are uncorrected. UV-vis
measurements were performed on a Perkin Elmer Lambda 900
instrument in CH2Cl2 solution (0.2 or 0.04 mM). Low tempera-
ture absorbance experiments were carried out with an Oxford
Instruments OptistatDN-V cryostat connected to a temperature
control unit. IR spectra were recorded on a Mattson 5000 FTIR
instrument in CHCl3 solution. High resolution mass spectra and
mass spectra were measured by electrospray ionization (ESI-MS)
in CHCl3–MeOH on a Bruker 4.7 T BioAPEX II. Elemental
analyses were performed at the ETH Zurich (Switzerland).
1
material. M.p. 83 °C. H NMR (CDCl3, 360 MHz): δ 11.48 (s,
1H, OH), 9.70 (s, 1H, CHO), 7.41 (d, 1H, J = 8.6 Hz, C6H),
6.52 (d, 1H, J = 8.6 Hz, C5H), 6.41 (s, 1H, C3H), 4.00 (t, 2H, J
= 6.5 Hz, OCH2), 1.79 (m, 2H, OCH2CH2), 1.50–1.39 (m, 2H,
OCH2CH2CH2), 1.39–1.13 (m, 52H, CH2), 0.88 (t, 3H, J = 6.4
Hz, Me). 13C{1H} NMR (CDCl3, 91 MHz): δ 194.4 (CHO),
166.6, 164.7 (2 × CarO), 135.3 (C6H), 115.1 (C1), 108.9 (C5H),
101.2 (C4H), 68.7 (OCH2), 32.1, 30.1–28.4, 26.1, 22.8 (all
CH2), 14.3 (Me). HR–MS (ESI): calcd for C37H65O3 [M − H]−
m/z = 557.4939, found m/z = 557.4930. Anal. found (calcd) for
C37H66O3 (558.93): C 79.48 (79.51); H 12.19 (11.90).
Synthesis of 4-(docosyloxy)salicylaldehyde
Synthesis of complex 1f
To a solution of 2,4-hydroxybenzaldehyde (1.86 g, 13.2 mmol)
in DMF (20 mL) was added NaHCO3 (1.11 g, 13.2 mmol). After
10 min stirring at RT, 1-bromodocosane (4.50 g, 11.0 mmol) in
DMF–THF (1 : 4 v/v, 25 mL) was slowly added. The mixture
was heated to 120 °C for 3 h under Ar. After cooling to RT,
aqueous HCl (1 M, 100 mL) was added, and the mixture was
stirred vigorously and then filtered. The residue was suspended in
acetone (100 mL), filtered, washed again with acetone (100 mL),
and then extracted with THF (4 × 50 mL). A brownish precipitate
formed upon standing for 2 d at 4 °C, which was filtered off. The
filtrate was concentrated to 50 mL and the formation of a precipi-
tate was induced by addition of EtOH under stirring. The residue
was filtered and dried in vacuo to give the title compound as a
white solid (3.65 g, 74%). Purification by column chromato-
graphy (SiO2, hexane–THF 15 : 1) afforded a microanalytically
Triethylenetetramine (43 mg, 0.30 mmol) was dissolved in
EtOH (2 mL) and treated with a solution of 4-(docosyloxy)sali-
cylaldehyde (264 mg, 0.59 mmol) in THF (10 mL). After
10 min, solid NaOMe (32 mg, 0.59 mmol) was added and the
yellowish suspension was stirred for 10 min. An ethanolic sol-
ution of Fe(NO3)3 × 9H2O (120 mg, 0.30 mmol in 3 mL) was
then added dropwise. The dark purple suspension was stirred for
30 min at RT and filtered over a short pad of silica. The product
was eluted with warm EtOH–THF 2 : 1 (3 × 60 mL) and dried
under reduced pressure. The residue was dissolved in CHCl3
(20 mL) and purified on a short pad of Al2O3 by consecutive
elution with CHCl3 (150 mL) and warm EtOH–THF 2 : 1 (3 ×
30 mL). After evaporation of the EtOH–THF fraction, the
residue was redissolved in CH2Cl2 (20 mL) and centrifuged. The
supernatant was evaporated under reduced pressure to give the
analytically pure purple solid 1f (140 mg, 42%). M.p. 197 °C
(decomp.). IR (CHCl3): 1600 cm−1 (CvN). UV-vis (CH2Cl2):
λmax (ε) = 496 nm (4100 M−1 cm−1). HR–MS (ESI): calcd for
C64H112FeN4O4 [M–NO3]+ m/z = 1056.8027, found m/z =
1056.8026. Anal. found (calcd) for C64H112FeN5O11 (1119.47):
C 68.40 (68.67); H 10.13 (10.08); N 5.97 (6.26).
1
pure white solid. M.p. 73 °C. H NMR (CDCl3, 360 MHz): δ
11.49 (s, 1H, OH), 9.70 (s, 1H, CHO), 7.41 (d, 1H, J = 8.6 Hz,
C6H), 6.52 (d, 1H, J = 8.6 Hz, C5H), 6.41 (s, 1H, C3H), 4.00 (t,
2H, J = 6.5 Hz, OCH2), 1.79 (m, 2H, OCH2CH2), 1.51–1.40 (m,
2H, OCH2CH2CH2), 1.40–1.18 (m, 36H, CH2), 0.87 (t, 3H, J =
6.4 Hz, Me). 13C{1H} NMR (CDCl3, 91 MHz): δ 194.3 (CHO),
166.5, 164.6 (CarO), 135.2 (C6H), 115.1 (C1), 108.8 (C5H),
101.1 (C4H), 68.6 (OCH2), 32.0, 30.1–28.6, 26.0, 22.8 (all
CH2), 14.2 (Me). HR–MS (ESI): calcd for C29H49O3 [M − H]−
m/z = 445.3687, found m/z = 445.3680. Anal. found (calcd) for
C29H50O3 (446.71): C 77.90 (77.97); H 11.30 (11.28).
Synthesis of complex 1g
According to procedure 1f, triethylenetetramine (14 mg,
0.10 mmol) in EtOH (2 mL) was reacted with 4-(triacontyloxy)
salicylaldehyde (110 mg, 0.20 mmol) in warm THF (16 mL),
NaOMe (11 mg, 0.20 mmol) and ethanolic Fe(NO3)3 × 9H2O
(40 mg, 0.10 mmol in 2 mL). The dark purple suspension was
stirred for 30 min at 60 °C and filtered while warm. After cooling
Synthesis of 4-(triacontyloxy)salicylaldehyde
To a solution of 2,4-hydroxybenzaldehyde (169 mg, 1.2 mmol)
in DMF (5 mL) was added NaHCO3 (101 mg, 1.2 mmol). After
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 3726–3730 | 3729

View Article Online
11644–116507; (d) B. Weber, W. Bauer and J. Obel, Angew. Chem., Int.
Ed., 2008, 47, 10098–10101.
8 (a) M. Hostettler, K. W. Törnroos, D. Chernyshov, B. Vangdal and H.-
B. Bürgi, Angew. Chem., Int. Ed., 2004, 43, 4589–4594; (b) B. Li,
R. Wei, J. Tao, R. Huang, L. Zheng and Z. Zheng, J. Am. Chem. Soc.,
2010, 132, 1558–1566.
9 (a) O. Kahn, L. Sommier and E. Codjovi, Chem. Mater., 1997, 9, 3199–
3205; (b) A. Bousseksou and G. Molnar, C. R. Chim., 2003, 6, 1175–
1183.
10 M. M. Dirtu, A. Rotaru, D. Gillard, J. Linares, E. Codjovi, B. Tinant and
Y. Garcia, Inorg. Chem., 2009, 48, 7838–7852.
11 (a) T. Forestier, A. Kaiba, S. Pechev, D. Denux, P. Guionneau,
C. Etrillard, N. Daro, E. Freysz and J. Letard, Chem.–Eur. J., 2009, 15,
6122–6130; (b) Y. Murakami, T. Komatsu and N. Kojima, Synth. Met.,
1999, 103, 2157–2158; (c) M. B. Meredith, J. A. Crisp, E. D. Brady, T.
P. Hanusa, G. T. Yee, M. Pink, W. W. Brennessel and V. G. Young, Orga-
nometallics, 2008, 27, 5464–5473; (d) M. Quesada, V. de la Pena-
O’Shea, G. Aromi, S. Geremia, C. Massera, O. Roubeau, P. Gamez and
J. Reedijk, Adv. Mater., 2007, 19, 1397–1402; (e) J. A. Kitchen and
S. Brooker, Coord. Chem. Rev., 2008, 252, 2072–2092; (f) J. A. Kitchen,
N. G. White, C. Gandolfi, M. Albrecht, G. N. L. Jameson, J. L. Tallon
and S. Brooker, Chem. Commun., 2010, 46, 6464–6466.
to RT the filtrate was centrifuged. The residue was redissolved in
warm THF (70 mL), EtOH (70 mL) was added and a precipitate
formed upon stirring at 0 °C for 30 min. After centrifugation the
residue was separated, and dissolved in warm THF. Upon stand-
ing at 0 °C for 30 min a precipitate formed which was collected
by centrifugation, and redissolved in warm THF. After filtration
through Celite, and evaporation of volatiles, 1g was obtained as a
purple solid (58 mg, 47%). M.p. 196 °C (decomp.). IR (CHCl3):
1604 cm−1 (CvN). UV-vis (CH2Cl2): λmax (ε) = 494 nm (4200
M−1 cm−1). HR–MS (ESI, MeOH): calcd for C80H144FeN4O4
[M–NO3]+ m/z = 1281.0531, found m/z = 1281.0530. Anal.
found (calcd) for C80H144FeN5O7 (1343.90) × H2O: C 70.82
(70.55); H 11.16 (10.81); N 5.21 (5.14).
Procedure for the cryo-SEM measurements
Images were acquired from 0.5–5 mM solutions in CH2Cl2
which were rapidly frozen using precooled liquid nitrogen. The
sample was transferred onto the prechamber attached to a Philips
XL30 FEG scanning electron microscope then sublimed at
−95 °C under high vacuum for 15 min, cooled and sputter
coated with platinum. The sample was moved onto the cryostat
in the main chamber of the microscope and viewed at 3–20 kV
using a secondary electron detector. A sample of a CH2Cl2 sol-
ution of 1f was further analyzed by energy dispersive X-ray
analysis at 20 kV and at 10 kV, revealing the presence of the
elements Fe, C, N, and O constituting the complex.
12 R. Ohtani, K. Yoneda, S. Furukawa, N. Horike, S. Kitagawa, A.
B. Gaspar, M. C. Muñoz, J. A. Real and M. Ohba, J. Am. Chem. Soc.,
2011, 133, 8600–8605.
13 (a) T. Fujigaya, D.-L. Jiang and T. Aida, J. Am. Chem. Soc., 2003, 125,
14690–14691; (b) O. Roubeau, A. Colin, V. Schmitt and R. Clerac,
Angew. Chem., Int. Ed., 2004, 43, 3283–3286; (c) Y. Bodenthin,
U. Pietsch, H. Möhwald and D. G. Kurth, J. Am. Chem. Soc., 2005, 127,
3110–3114; (d) C. Rajadurai, F. Schramm, S. Brink, O. Fuhr, M. Ghafari,
R. Kruk and M. Ruben, Inorg. Chem., 2006, 45, 10019–10021; (e) A.
B. Gaspar, M. Seredyuk and P. Gütlich, Coord. Chem. Rev., 2009, 253,
2399–2413; (f) I. Salitros, J. Pavlik, R. Boca, O. Fuhr, C. Rajadurai and
M. Ruben, CrystEngComm, 2010, 12, 2361–22368; (g) R. Gonzalez-
Prieto, B. Fleury, F. Schramm, G. Zoppellaro, R. Chandrasekar, O. Fuhr,
S. Lebedkin, M. Kappes and M. Ruben, Dalton Trans., 2011, 40, 7564–
7570.
14 (a) T. Fujigaya, D. L. Jiang and T. Aida, J. Am. Chem. Soc., 2005, 127,
5484–5489; (b) E. Coronado, J. R. Galan-Mascaros, M. Monrabal-
Capilla, J. Garcia-Martinez and P. Pardo-Ibanez, Adv. Mater., 2007, 19,
1359–1361; (c) T. Fujigaya, D. L. Jiang and T. Aida, Chem.–Asian J.,
2007, 2, 106–113; (d) H. Matsukizono, K. Kuroiwa and N. Kimizuka, J.
Am. Chem. Soc., 2008, 130, 5622–5623; (e) V. Martinez, I. Boldog, A.
B. Gaspar, V. Ksenofontov, A. Bhattacharjee, P. Gütlich and J. A. Real,
Chem. Mater., 2010, 22, 4271–4281; (f) K. Kuroiwa, H. Kikuchi and
N. Kimizuka, Chem. Commun., 2010, 46, 1229–1231; (g) A. Tokarev,
L. Salmon, Y. Guari, W. Nicolazzi, G. Molnar and A. Bousseksou, Chem.
Commun., 2010, 46, 8011–8013 For an instructive review, see:
(h) A. Bousseksou, G. Molnar, L. Salmon and W. Nicolazzi, Chem. Soc.
Rev., 2011, 40, 3313–3335.
Acknowledgements
We gratefully acknowledge financial support from ERA-net
Chemistry, the Swiss National Science Foundation, the Alfred
Werner Foundation, and the European Research Council (StG
208561). We thank T. Bally (Univ. Fribourg) for sharing UV-vis
spectroscopic facilities.
Notes and references
1 J.-F. Létard, Top. Curr. Chem., 2004, 235, 221–249.
15 M. F. Tweedle and L. J. Wilson, J. Am. Chem. Soc., 1976, 98, 4824–
4834.
16 B. Weber, Coord. Chem. Rev., 2009, 253, 2432–2449.
17 C. Gandolfi, N. Miyashita, D. G. Kurth, P. N. Martinho, G. G. Morgan
and M. Albrecht, Dalton Trans., 2010, 39, 4508–4516.
18 C. Gandolfi, C. Moitzi, P. Schurtenberger, G. G. Morgan and
M. Albrecht, J. Am. Chem. Soc., 2008, 130, 14434–14435.
19 Complexes 1c–g were amorphous in the solid state; all our attempts to
grow crystals of these complexes have failed.
20 C. Brady, J. J. McGarvey, J. K. McCusker, H. Toftlund and D.
N. Hendrickson, Top. Curr. Chem., 2004, 235, 1–22.
21 For a different approach to induce abrupt SCO in such complexes, see:
S. Dorbes, L. Valade, J. A. Real and C. Faulmann, Chem. Commun.,
2005, 69–71.
22 For complexes 1a–d, thermodynamic data cannot be reliably determined
since the spin transition is gradual and may not be complete at the lowest
accessible temperature for CH2Cl2 solution.
23 For hysteretic spin crossover induced by N-functionalization of the sal2t-
rien ligand, see: P. N. Martinho, Y. Ortin, B. J. Gildea, C. Gandolfi,
G. McKerr, B. O’Hagan, M. Albrecht and G. G. Morgan, submitted for
publication.
2 (a) Spin Crossover in Transition Metal Compounds I–III, ed. P. Gütlich
and H. A. Goodwin, Springer-Verlag, Berlin, 2004; (b) E. König,
G. Ritter and S. K. Kulshreshtha, Chem. Rev., 1985, 85, 219–234; (c) P.
J. Van Koningsbruggen, Y. Maeda and H. Oshio, Top. Curr. Chem., 2004,
233, 259–324; (d) O. Sato, J. Tao and Y.-Z. Zhang, Angew. Chem., Int.
Ed., 2007, 46, 2152–2187; (e) J. A. Real, A. B. Gaspar and M.
C. Muñoz, Dalton Trans., 2005, 2062–2079.
3 For a system that is dictated on the molecular scale exclusively, see:
S. Venkataramani, U. Jana, M. Dommaschk, F. D. Sönnichsen, F. Tuczek
and R. Herges, Science, 2011, 331, 445–448.
4 (a) O. Kahn and C. J. Martinez, Science, 1998, 279, 44–48; (b) J.
J. Parks, A. R. Champagne, T. A. Costi, W. W. Shum, A. N. Pasupathy,
E. Neuscamman, S. Flores-Torres, P. S. Cornaglia, A. A. Aligia, C.
A. Balseiro, G. K.-L. Chan, H. D. Abruna and D. C. Ralph, Science,
2010, 328, 1370–1373.
5 (a) H. Bolvin, O. Kahn and B. Vekhter, New J. Chem., 1991, 15, 889–
895; (b) K. S. Murray and C. J. Kepert, Top. Curr. Chem., 2004, 233,
195–228; (c) H. Spiering, Top. Curr. Chem., 2004, 235, 171–195.
6 S. Cobo, G. Molnar, J. A. Real and A. Bousseksou, Angew. Chem., Int.
Ed., 2006, 45, 5786–5789.
7 (a) J.-F. Letard, P. Guionneau, E. Codjovi, O. Lavastre, G. Bravic,
D. Chasseau and O. Kahn, J. Am. Chem. Soc., 1997, 119, 10861–10862;
(b) Z. J. Zhong, J.-Q. Tao, Z. Yu, C.-Y. Dun, Y.-J. Lui and X.-Z. You, J.
Chem. Soc., Dalton Trans., 1998, 327–328; (c) S. Hayami, Z. Gu,
H. Yoshiki, A. Fujishima and O. Sato, J. Am. Chem. Soc., 2001, 123,
24 R. N. Muller, L. Vander Elst and S. Laurent, J. Am. Chem. Soc., 2003,
125, 8405–8407.
25 R. F. Keyes, W. M. Golebiewski and M. Cushman, J. Med. Chem., 1996,
39, 508–514.
3730 | Dalton Trans., 2012, 41, 3726–3730
This journal is © The Royal Society of Chemistry 2012