This document is the Accepted Manuscript version of a Published Work that
appeared in final form in Journal of the American Chemical Society, copyright ©
American Chemical Society after peer review and technical editing by the
publisher. To access the final edited and published work see
http://pubs.acs.org/doi/pdf/10.1021/jacs.6b09778

Easy excited state trapping and record TT IESST (250 K) in a spin crossover polyanionic
FeII trimer**

By Ver´ nica G´ mez,†,§ Cristina S´ enz de Pipa´ n,§ Pilar Maldonado-Illescas, Jo˜ o Carlos
o
o
a
o
a
Waerenborgh, Eddy Martin, Jordi Benet-Buchholz, and Jos´ Ram´ n Gal´ n-Mascar´ s*
e
o
a
o

[*] Prof. Dr. J. R. Gal´ n Mascar´ s, Dr. Ver´ nica G´ mez, Dr. Cristina S´ enz de Pipa´ n, Dr.
a
o
o
o
a
o
Eddy Martin, Dr. Jordi Benet-Buchholz, Pilar Maldonado-Illescas,
Institute of Chemical Research of Catalonia (ICIQ)
Av. Paisos Catalans 16, E-43007 Tarragona (Spain)
E-mail: jrgalan@iciq.es

[*] Prof. Dr. Jo˜ o Carlos Waerenborgh,
a
Centro de Ciˆ ncias e Tecnologias Nucleares, Instituto Superior T´ cnico
e
e
Universidade de Lisboa, 2695-066 Bobadela LRS, (Portugal)

[*] Prof. Dr. J. R. Gal´ n Mascar´ s
a
o
Catalan Institution for Research and Advanced Studies (ICREA)
Passeig Lluis Companys, 23, E-08010, Barcelona (Spain)

†

Present address: Institute of Nanotechnology, Karlsruhe Institute of Technology (KIT),

Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany.
§

Both authors contributed equally.

[**] We thank the ﬁnancial support of the EU (ERC Stg grant 279313, CHEMCOMP), the
Spanish Ministerio de Econom´a y Competitividad (MINECO) through Severo Ochoa Excelı
1

lence Accreditation 2014-2018 (SEV-2013-0319) and the ICIQ Foundation.

Abstract: Reaction of the polysulfonated triazole ligand L = 4-(1,2,4-triazol-4-yl)ethanedisulfonate) with iron(II) salts in water yields the trimeric species [Fe3 (µ-L)6 (H2 O)6 ]6− . This
polyanion can be isolated from solution as the dimethylammonium salt yielding a complex
hydrogen bonded network connecting anions, cations and solvent molecules in the solid state.
This material shows a thermally induced spin transition for the central Fe position in the trimer
above room temperature with a large hysteresis cycle (> 85 K) and remarkably slow dynamics.
This allows easy quenching of the metastable high spin (HS) state via slow cooling (5 K min−1 ).
Once it is trapped, the HS state remains metastable. Thermal energy is not able to promote
relaxation into the low spin (LS) ground state up to 215 K, with a characteristic TT IESST = 250
K, the highest temperature ever observed for thermal trapping in a spin crossover compound.

Iron(II) complexes displaying spin crossover (SCO), a phenomenon in which the spin state
of the metal center switches between low-spin (LS) and high-spin (HS) under external stimuli, have been the subject of many studies during the last decades. [1–5] SCO is particularly
appealing in the octahedral FeII case, because the spin transition from diamagnetic S = 0 to
paramagnetic S = 2 provokes a dramatic change also in color (from pink/red to white/yellow)
and size (increment of 10-15% in the metal to ligand bonding distances). Besides, if this transition occurs with hysteresis a memory effect is conferred. In very recent years there is a
renaissance of SCO materials in molecular magnetism [6–12] because, in contrast with other
bistable species such as single molecule magnets, [13,14] SCO may occur at room temperature.
Thus SCO appears specially suited for applications as stimuli-responsive molecular switches
for multifunctional materials, memories, electrical circuits or display devices. [15–27]
Most reported complexes with hysteretic SCO are mononuclear or polymeric FeII com2

pounds, typically with neutral or cationic overall charge. Our group has focused in the search
for anionic SCO complexes to increase their processing capabilities when incorporated into
hybrid materials. With this in mind, we have prepared sulfonated 1,2,4-triazole derivatives as
building blocks. [28, 29] Here, we report the SCO behavior of the polyanionic FeII SCO complex [Fe3 (µ-L)6 (H2 O)6 ]6− (L2− = 4-(1,2,4-triazol-4-yl)ethanedisulfonate, Figure S1). As the
corresponding (Me2 NH2 )+ salt, this complex exhibits a spin transition above room temperature
with a very large thermal hysteresis loop. Furthermore, the HS state can be easily trapped when
cooling down at slow rates (> 5 K min−1 ), remaining in its metastable state up to the highest
temperatures ever observed, close to room temperature.
Reaction of L2− with FeII in a 3:1 ratio in water yielded a pale pink solution. Single crystals of (Me2 NH2 )6 [Fe3 (µ-L)6 (H2 O)6 ](1) were isolated by layering the reaction mixture with
ethanol. 1 contains the trinuclear polyanion [Fe3 (µ-L)6 (H2 O)6 ]6− , formed by a linear array of
octahedral FeII ions (Fe1-Fe2-Fe1) connected by two triple µ-triazole bridges (Figure 1 and
Figure S2). The terminal Fe1 position completes its N3 O3 hexacoordination with three H2 O
molecules in f ac conformation. The metal to ligand distances at 100 K (Table S2) indicate HS
˚
conﬁguration for Fe1 (average Fe1−L = 2.13 A) and LS conﬁguration for Fe2 (average Fe2−N
˚
= 1.99 A). The dangling sulfonate moieties from the ethane groups adopt multiple orientations
participating in a complex weakly hydrogen bonded network involving the dimethylammonium
˚
cations and water molecules, with short intermolecular O−O distances (2.6−2.9 A, Figure S3).
Magnetic susceptibility measurements were carried out in the 2−300 K range for grained
single crystals of 1 (Figure 2). At room temperature, the χm T product is 6.37 cm3 mol−1 K,
consistent with a HS-LS-HS conﬁguration (spin-only χm T = 6.00 cm3 mol−1 K). On lowering
the temperature, χm T remains constant down to 50 K, when decreases down to 0.37 cm3 mol−1 K
at 2 K. We assign this decrease to zero-ﬁeld splitting (usually very small) and antiferromagnetic
interactions between paramagnetic centers.
3

Above room temperature (Figure 2, inset), χm T shows an abrupt increase at 360 K, suggesting a spin transition to the HS-HS-HS state, reaching 7.92 cm3 mol−1 K at 400 K. This
corresponds approximately to half of the trimers undergoing the spin transition. Upon cooling
again, the sample goes slowly back to the HS-LS-HS state, with a hysteresis loop of over 65
K, reaching the initial χm T values below 300 K. When the sample is maintained at 400 K,
χm T keeps increasing until it saturates at 8.89 cm3 mol−1 K (Figure S4), corresponding to ≈
84% completeness. The cooling behavior from magnetic saturation at 400 K in settle mode
(Figure 2inset), exhibits a hysteresis loop of 90 K (T1/2 (↑) = 400 K and T 1/2 (↓) = 310 K). This
is among the widest robust thermal hysteresis found for SCO materials. [31–35] This thermal
hysteresis is reproducible as conﬁrmed by the consistency of successive cycles at 1 K min−1
(Figure S5). The possibility of a hydration/dehydration process affecting the hysteretic behavior can be ruled out given that all measurements were done in pinched capsules to allow proper
purging. In addition, analogous behavior was observed for samples previously dehydrated in an
oven for 12 hours at 135◦ C (Figure S5) and thermal gravimetry analysis under humid air (Figure
S6) conﬁrms that 1 barely rehydrates during the cooling branch, and that water content is identical during successive cycles in the cooling or heating bench, depending only on temperature.
This conﬁrms that hydration/dehydration cannot promote appearance of a memory effect.
Since SCO can be very sensitive to scan rate, [30] we studied its effect in the hysteresis
cycle (Figure 3a). The heating branch does not change signiﬁcantly with scan rate. The cooling
branch deviates from ”settle mode” at rates above 2 K min−1 . Remarkably, the spin transition
becomes signiﬁcantly quenched at scan rates as slow as 5 K min−1 . This phenomenon, the
temperature-induced excited spin-state trapping (TIESST), typically requires very fast cooling
(liquid nitrogen immersion). [36–40] At a 10 K min−1 cooling rate, 57% of the complexes are
trapped as HS-HS-HS. We extracted a characteristic TT IESST = 250 K (Figure 3b) following the
method deﬁned by Letard et al. through a heating run at 0.3 K min−1 . [37] This is the highest
4

TT IESST reported, the previous record being 156 K. [41] At faster heating rates the HS-HS-HS
state is not completely depopulated until reaching room temperature. For example, 285 K are
needed at 1 K min−1 (Figure S7). It is important to note that the X-ray diffraction pattern does
not change after multiple thermal cycles, conﬁrming the stability of the material and also the
absence of additional crystallographic phases (Figure S8).
We measured isothermal relaxation curves after keeping 1 at 400 K for 6.5 hours, and then
cooling at 10 K min−1 down to the desired temperature (Figure 4). Relaxation exhibits sigmoidal character below 245 K, but it deviates from this behavior at higher temperatures suggesting a decrease in cooperativity. However, the plots cannot be satisfactorily modeled with
a unique relaxation process (sigmoidal or stretch exponential), which suggests participation of
multiple microscopic contributions. From the relaxation rate at very long times, as extracted
from the linear decay of the plot tails (Figure S9), we can estimate relaxation times as long as
over 8 hours at 265 K, and over 3 days at 235 K, for example. This showcases the remarkable
slow dynamics characteristic of this SCO material.
The

57

Fe M¨ ssbauer data are in good agreement with the magnetic measurements (Figure
o

5a,b). Both spectra at 295 and 4 K show two quadrupole doublets. One with higher quadrupole
splitting (QS) due to HS FeII and the other with lower QS typical of LS FeII . [1] At 4 K the
recoil-free fractions of LS and HS are usually similar and their relative areas may be considered
a good approximation of the LS and HS Fe fractions. The LS FeII relative area is slightly
higher at 295 K which may be attributed to its higher recoil-free fraction when compared to
HS FeII at 295 K. The ratio of the relative areas at 4 K, 2:1 within experimental accuracy,
is consistent with the HS-LS-HS ground state in the 295 - 4 K temperature range. To study
the thermally trapped metastable HS-HS-HS state, we heated a sample at 403 K during ﬁve
hours, droped it quickly into liquid nitrogen, and brought it to the cryostat. The spectrum
obtained at 4 K (Figure 5c) showed that the relative area of the LS FeII doublet was reduced
5

approximately 10% relative to the sample before heat treatment (Table 1). In addition, the HS
FeII contribution could only be ﬁtted if three doublets were considered: a ﬁrst doublet (d1)
with parameters similar to those of terminal HS FeII in the untreated sample; a second doublet
(d2) with a relative area of ≈11% that could be assigned to central HS FeII (notice that 11%
is within experimental accuracy equal to the relative area reduction of the central LS FeII );
and a third doublet (d3), with the highest QS, which may be attributed to terminal HS FeII
with a modiﬁed environment. The sum of d1 and d3 relative areas is equal to the area of the
HS doublet in the untreated sample. Thus, the three HS contributions to the spectra arise from
the conﬁgurations HS(d1)-LS-HS(d1) and HS(d3)-HS(d2)-HS(d3). Considering that the central
FeII cations are 33% of the total, these results indicate quenching of one third of the molecules in
this experiment. The spectrum taken after warming the quenched sample to room temperature is
identical, within experimental error, to the spectrum of the untreated sample (Figure 5, Table 1),
showing that the thermally quenched state decays at room temperature towards the as-prepared
state. This means that all molecules are back to the HS-LS-HS ground state. The presence of
two different environments for the terminal FeII positions is only observed when HS-HS-HS
species are present. The spectrum taken at 4 K of the treated sample after warming to room
temperature for 48 hours is indistinguishable from the 4 K spectrum of the untreated sample
conﬁrming the conclusions deduced from the 295 K spectrum, and the perfect stability (and
bistability) of the sample through multiple temperature cycles.
In summary, we have isolated a novel polyanionic FeII trimer as its dimethylamonnium salt,
which exhibits spin crossover behavior for its central FeII position above room temperature
with a large thermal hysteresis. The spin transition dynamics are remarkably slow, allowing for
an easy trapping of the excited spin state when the material is cooled down at reasonably slow
rates (> 5 K min−1 ). Remarkably, this material exhibits bistability in the high temperature range
(280−400 K), but also in the low temperature range (2−250 K), since the metastable HS state
6

Table 1: Parameters estimated from the M¨ ssbauer spectra of the pristine and the heat treated
o
samples of compound 1 taken at 295 and 4 K.
sample T (K)
295
pristine 1
4

1 heated at 403 K
and quenched
in liquid N2
1 heated at 403 K,
quenched in liquid
N2 and warmed
up to 295 K

4

295
4

IS(mm s−1 )a
0.45
1.14
0.52
1.27
0.52
1.29
1.30
1.25
0.45
1.15
0.52
1.28

QS(mm s−1 )b
0.14
1.98
0.19
2.91
0.24
2.91
2.30
3.38
0.12
1.97
0.20
2.88

a

Γ(mm s−1 )c
0.27
0.41
0.28
0.42
0.33
0.43
0.37
0.35
0.28
0.40
0.30
0.43

I(%)d
39
61
35
65
25
42
11
22
38
62
37
63

assignment
LS
HS (d1)
LS
HS(d1)
LS
HS (d1)
HS (d2)
HS (d3)
LS
HS (d1)
LS
HS (d1)

Isomer shift relative to metallic α-Fe at 295 K; b Quadrupole splitting; c Full width at half height; d Relative areas.
d1: terminal HS FeII in HS-LS-HS conﬁguration; d2: central HS FeII in HS-HS-HS conﬁguration; d3: terminal
HS FeII in HS-HS-HS conﬁguration (see text). Estimated errors are ≤ 0.02 mm s−1 for IS and QS and ≤ 2%
for I. The areas and widths of both peaks in a quadrupole doublet were constrained to remain equal during the
reﬁnement procedure.

7

cannot relax back to the ground state up to very high temperatures (TT IESST = 250 K). This is
the highest temperature at which memory effect remains in a thermally quenched SCO system,
100 K higher than that of the previous record. It is even higher than the values displayed by
Prussian blue analogues, where electron transfer kinetics play a role in the stabilization of the
metastable effect. [42] We assign the origin of such novel features to the presence of multiple
sulfate groups in the ligands backbone. The size changes in the central Fe position imposes the
movement of highly charged moieties. This is unique, since in previous SCO materials (mostly
neutral or cationic) the ligands are either neutral, or mono anionic with the charged groups
directly bound to the Fe centers. The effect of such electrostatic issues is particularly important
during the HS → LS process, when the shrinking to the ground state forces the direct approach
between multiple sulfate groups on opposite triazoles bridges. Thus we assign the appearance
of such a high activation energy to this electrostatic repulsion, making the process slow scan
rate dependent, and resulting in easy quenching of the metastable state. For instance, such scan
rate dependence does not appear during the opposite LS → HS transition, when the distances
between sulfate groups increase. We would like to also highlight that this molecular material
is also soluble and stable in water allowing for easy and convenient processing techniques to
develop bistable hybrid materials at the nanoscale.
Keywords: bistability; hysteresis; iron; spin crossover; triazole

References
[1] P. G¨ tlich, Y. Garcia, H. A. Goodwin. Chem. Soc. Rev. 2000, 29, 419−427.
u
[2] A. Bousseksou, G. Molnar, L. Salmon, W. Nicolazzi.
3313−3335.

8

Chem. Soc. Rev. 2011, 40

[3] M. Carmen Munoz, J. A. Real. Coord. Chem. Rev. 2011, 255, 2068−2093.
[4] G. Arom´, L. A. Barriosa, O. Roubeau, P. Gamez. Coord. Chem. Rev. 2010, 255, 485−546.
ı
[5] M. A. Halcrow. Chem. Soc. Rev. 2011, 40, 4119−4142.
[6] P. G¨ tlich. Eur. J. Inorg. Chem. 2013, 581−591.
u
[7] C. M. Quintero, G. F´ lix, I. Suleimanov, J. S. Costa, G. Molnar, L. Salmon, W. Nicolazzi,
e
A. Bousseksou. Beilstein J. Nanotechnol. 2014, 5, 2230−2239.
[8] J. S. Costa, S. Rodr´guez-Jim´ nez, G. A. Craig, B. Barth, C. M. Beavers, S. J. Teat,
ı
e
G. Arom´. J. Am. Chem. Soc. 2014, 136, 3869−3874.
ı
[9] F. Prins, M. Monrabal-Capilla, E. A. Osorio, E. Coronado, H. S. J. van der Zant. Adv.
Mater. 2011, 23, 1545−1549.
[10] T. Miyamachi, M. Gruber, V. Davesne, M. Bowen, S. Boukari, L. Joly, F. Scheurer, G. Rogez, T. K. Yamada, P. Ohresser, E. Beaurepaire, W. Wulfhekel. Nat Commun 2012, 3,
938.
[11] S. Ohkoshi, K. Imoto, Y. Tsunobuchi, S. Takano, H. Tokoro. Nat. Chem. 2011, 3,
564−569.
[12] S. Samanta, S. Demesko, S. Dechert, F. Meyer. Angew. Chem. Int. Ed. 2015, 54, 583−587.
[13] R. Sessoli, A. Powell. Coord. Chem. Rev. 2009, 253, 2328−2341.
[14] L. Bogani, W. Wernsdorfer. Nat. Mater. 2008, 7, 179−186.
[15] T. Matsumoto, G. N. Newton, T. Shiga, S. Hayami, Y. Matsui, H. Okamoto, R. Kumai,
Y. Murakami, H. Oshio. Nat. Commun. 2014, 5, 3865.
9

[16] D. Aravena, E. Ruiz. J. Am. Chem. Soc. 2012, 134, 777−779.
[17] G. A. Craig, O. Roubeau, G. Arom´. Coord. Chem. Rev. 2015, 269, 13−31.
ı
[18] B. Warner, J. C. Oberg, T. G. Gill, F. El Hallak, C. F. Hirjibehedin, M. Serri, S. Heutz,
M.-A. Arrio, P. Sainctavit, M. Mannini, G. Poneti, R. Sessoli, P. Rosa. J. Phys. Chem.
Lett. 2013, 4, 1546−1552.
[19] J. Larionova, L. Salmon, Y. Guari, A. Tokarev, K. Molvinger, G. Molnar, A. Bousseksou.
Angew. Chem. Int. Ed. 2008, 47, 8236−824.
[20] C.-F. Wang, R.-F. Li, X.-Y. Chen, R.-J. Wei, L.-S. Zheng, J. Tao. Angew. Chem. Int. Ed.
2015, 54, 1574−1577.
[21] Y.-S. Koo, J. R. Galan-Mascaros. Adv. Mater. 2014, 26, 6785−6789.
[22] M. B. Duriska, S. M. Neville, B. Moubaraki, J. A. Cashion, G. J. Halder, K. W. Chapman,
C. Balde, J. F. Letard, K. S. Murray, C. J. Kepert, S. R. Batten. Angew. Chem. Int. Ed.
2009, 48, 2549−2552.
[23] E. M. Hern´ ndez, C. M. Quintero, O. Kraieva, C. Thibault, C. Bergaud, L. Salmon,
a
G. Molnar, A. Bousseksou. Adv. Mater. 2014, 26, 2889−2893.
[24] H. Phan, S. M. Benjamin, E. Steven, J. S. Brooks, M. Shatruk. Angew. Chem. Int. Ed.
2015, 54, 823−827.
[25] E. Ludwig, H. Naggert, M. Kall¨ ne, S. Rohlf, E. Kr¨ ger, A. Bannwarth, A. Quer, K. Rossa
o
nagel, L. Kipp, F. Tuczek. Angew. Chem. Int. Ed. 2014, 53, 3019−3023.
[26] M. Seredyuk, M. C. Mu˜ oz, M. Castro, T. Romero-Morcillo, A. B. Gaspar, J. A. Real.
n
Chem. Eur. J. 2013, 19, 6591−6596.
10

[27] V. A. Money, C. Carbonera, J. Elha¨k, M. A. Halcrow, J. A. K. Howard, J.-F. Letard.
ı
Chem. Eur. J. 2007, 13, 5503−5514.
[28] V. Gomez, J. Benet-Buchholz, E. Martin, J. R. Galan-Mascaros. Chem. Eur. J. 2014, 20,
5369−537.
[29] V. Gomez, V. Lillo, E. C. Escudero-Adan, E. Martin, J. R. Galan-Mascaros. Dalton Trans.
2013, 42, 6374−6380.
[30] R. Kulmaczewski, J. Olgu´n, J. A. Kitchen, H. L. C. Feltham, G. N. L. Jameson, J. L.
ı
Tallon, S. Brooker. J. Am. Chem. Soc. 2014, 136, 878−881.
[31] M. A. Halcrow. Chem. Lett. 2014, 43, 1178−1188.
[32] C. Lochenie, W. Bauer, A. P. Railliet, S. Schlamp, Y. Garcia, B. Weber. Inorg. Chem.
2014, 43, 11563−11572.
[33] B. Weber, W. Bauer, T. Pfaffeneder, M. M. Dirtu, A. D. Naik, A. Rotaru, Y. Garcia. Eur.
J. Inorg. Chem. 2011, 3193−3206.
[34] B. Weber, W. Bauer, J. Obel. Angew. Chem. Int. Ed. 2008, 47, 10098−10101.
[35] M. B. Bushuev, V. A. Daletsky, D. P. Pishchur, Y. V. Gatilov, I. V. Korolkov, E. B. Nikolaenkova, V. P. Krivopalov. Dalton Trans. 2014, 43, 3906−3910.
[36] K. D. Murnaghan, C. Carbonera, L. Toupet, M. Grifﬁn, M. M. Dirtu, C. Desplanches,
Y. Garcia, E. Collet, J.-F. Letard, G. G. Morgan. Chem. Eur. J. 2014, 20, 5613−5618.
[37] N. Paradis, G. Chastanet, J.-F. Letard. Eur. J. Inorg. Chem. 2012, 2012, 3618−3624.
[38] Y. Garcia, V. Ksenofontov, S. Mentior, M. Dirtu, C. Gieck, A. Bhatthacharjee, P. G¨ tlich.
u
Chem. Eur. J. 2008, 14, 3745−3758.
11

[39] G. A. Craig, J. S. Costa, O. Roubeau, S. J. Teat, G. Arom´. Chem. Eur. J. 2011, 17,
ı
3120−3127.
[40] G. A. Craig, J. S. Costa, O. Roubeau, S. J. Teat, H. J. Shepherd, M. Lopes, G. Molnar,
A. Bousseksou, G. Arom´. Dalton Trans. 2013, 43, 729−737.
ı
[41] H. Wang, C. Sinito, A. Kaiba, J. S. Costa, C. Desplanches, P. Dagault, P. Guionneau, J.-F.
Letard, P. N´ grier, D. Mondieig. Eur. J. Inorg. Chem. 2014, 4927−4933.
e
[42] D. Li, R. Clerac, O. Roubeau, E. Hart´ , C. Mathoniere, R. Le Bris, S. M. Holmes. J. Am.
e
Chem. Soc. 2008, 130, 252−258.

12

Entry for the Table of Contents:
High temperature metastability is exhibited by the polyanion [Fe3 (µ-L)6 (H2 O)6 ]6− as the
dimehtylammonium salt. This complex shows a thermally induced spin transition for the central Fe position in the trimer above room temperature with a large hysteresis cycle (> 85 K)
and remarkably slow dynamics. This allows easy quenching of the metastable high spin (HS)
state via slow cooling (5 K min−1 ). Once it is trapped, the HS state remains metastable with a
characteristic TT IESST = 250 K, the highest temperature ever observed for thermal trapping in
a spin crossover material.

13

Figure 1: Molecular structure of the trimer [Fe3 (µ-L)6 (H2 O)6 ]6− .

Figure 2: χm T vs T plots for 1 in the 2-300 K range and in the 270-400 K range (inset). Empty
squares show the magnetic behavior when the sample is maintained at 400 K until saturation.
These measurements were performed stabilizing the temperature for each data point (scan rate
≈ 0.1 K min−1 ).
14

Figure 3: χm T vs T plots for 1 at: (top) different heating/cooling rates, allowing for saturation at
400 K; (bottom) heating up to 400 K until saturation (red circles), cooling down from saturation
at 10 K min−1 rate (blue circles), and warming up again at 0.3 K min−1 (black circles). (inset)
δχm T /δT vs T curve for the latter heating branch, whose minimum deﬁnes TT IESST .

15

Figure 4: Isothermal relaxation curves for the HS fraction (γHS ) for 1 at 265, 260, 255, 250,
240 and 235 K.

16

Figure 5: 57 Fe M¨ ssbauer spectra of compound 1. Spectra of pristine sample taken at (a) 295
o
and (b) 4 K. Spectra of sample heated at 403 K and quenched in liquid N2 , taken at (c) 4 K, (d)
295 K after warming up for 24 hours and (e) at 4 K after cooling down from 295 K. The blue
and red lines refer to the LS and HS FeII , respectively.

17