Subtlety of the spin-crossover phenomenon observed with dipyridylamino-substituted triazine ligands

Article abstract of DOI:10.1002/ejic.201201085

Reactions of the new, closely related ligands 4,6-dichloro- N,N-di(pyridine-2-yl)-1,3,5-triazine-amine (Cldpat) and 6- chloro-N'-phenyl-N,N- di(pyridin-2-yl)-1,3,5-triazine-2,4-diamine (Cladpat) with iron(II) thiocyanate produced coordination compounds with drastically distinct magnetic properties. The compound trans-[Fe(Cldpat)₂(NCS)₂](H₂O) (1) is a highspin complex from room temperature down to 5 K whereas the analogous compound trans-[Fe(Cladpat) ₂ (NCS) ₂] (2) exhibits spin-crossover (SCO) properties with T_1/2 = 178K. Compounds 1 and 2 (both in its low-spin and high-spin states) have been structurally characterized by X-ray diffraction studies, which revealed identical metal coordination spheres. The SCO properties of 2 have been thoroughly investigated by temperature-dependent magnetic susceptibility measurements and differential scanning calorimetry (DSC), and a LIESST process with rapid relaxation of the trapped HS species has been observed. The equivalent coordination compound with selenocyanate anions, namely [Fe(Cladpat) ₂ (NCSe) ₂] (3) also displays SCO properties, although more gradual and with a lower T _1/2 value of 166 K.

Full text of DOI:10.1002/ejic.201201085

FULL PAPER
CLUSTER
ISSUE
DOI:10.1002/ejic.201201085
Subtlety of the Spin-Crossover Phenomenon Observed with
Dipyridylamino-Substituted Triazine Ligands
Nanthawat Wannarit,[a,b] Olivier Roubeau,*[c] Sujittra Youngme,*[b]
and Patrick Gamez*[a,d]
FRONT COVER
Keywords: Iron / LIESST effect / Calorimetry / Supramolecular interactions / Spin crossover
Reactions of the new, closely related ligands 4,6-dichloro-
N,N-di(pyridine-2-yl)-1,3,5-triazine-amine (Cldpat) and 6-
states) have been structurally characterized by X-ray diffrac-
tion studies, which revealed identical metal coordination
chloro-NЈ-phenyl-N,N-di(pyridin-2-yl)-1,3,5-triazine-2,4-di- spheres. The SCO properties of 2 have been thoroughly in-
amine (Cladpat) with iron(II) thiocyanate produced coordina-
tion compounds with drastically distinct magnetic properties.
The compound trans-[Fe(Cldpat)2(NCS)2](H2O) (1) is a high-
spin complex from room temperature down to 5 K whereas
the analogous compound trans-[Fe(Cladpat)2(NCS)2] (2) ex-
hibits spin-crossover (SCO) properties with T1/2 = 178 K.
Compounds 1 and 2 (both in its low-spin and high-spin
vestigated by temperature-dependent magnetic suscep-
tibility measurements and differential scanning calorimetry
(DSC), and a LIESST process with rapid relaxation of the
trapped HS species has been observed. The equivalent coor-
dination compound with selenocyanate anions, namely
[Fe(Cladpat)2(NCSe)2] (3) also displays SCO properties, al-
though more gradual and with a lower T1/2 value of 166 K.
materials may find potential applications in molecular
switches, data storage devices and optical displays, espe-
cially in the case of FeII ions, for which the LS state is dia-
magnetic.[8–11] Therefore, SCO species have received a great
deal of attention from the scientific community for the past
decade.[12,13] A great number of SCO FeII complexes have
been synthesized that were principally obtained from li-
gands based on nitrogen-containing aromatic donor groups
(such as pyridine and azole rings).[14–19]
For the past seven years, we have been involved in the
design and preparation of SCO iron(II) compounds with
polypyridine ligands.[20–22] In particular, the use of a ligand
derived from the s-triazine ring and 2,2Ј-dipyridylamine
units, namely 2,4,6-tris(dipyridin-2-ylamino)-1,3,5-triazine
(dpyatriz),[23] has led to remarkable SCO systems.[22,24,25]
Subsequently, Murray and coworkers have developed a
variety of dipyridylamino-substituted-triazine ligands,[26–30]
which were easily prepared from the highly versatile build-
ing block 2,4,6-trichloro-1,3,5-triazine.[31–33] The utilization
of these ligands has allowed the preparation of SCO coordi-
nation compounds, hence corroborating the great potential
of (2,2Јdipyridylamine/triazine)-based ligands to generate
molecular switches.
Introduction
The phenomenon of spin crossover (SCO) is a particular
and very interesting illustration of the ligand-field
theory.[1,2] For octahedral coordination complexes of transi-
tion-metal ions, the d orbitals split into two sets, i.e. the t2g
and eg sets, whose energy difference is given by the crystal-
[3]
field splitting parameter Δoct
.
The size of Δoct determines
the electronic structure of the d4–d7 metal ions. Thus, for
iron(II) complexes, a small Δoct will favour the high-spin
2
state (HS, eg t2g4, S = 2) while a large Δoct will produce a
low-spin (LS, t2g6, S = 0) compound. With an appropriate
ligand-field strength (namely for an intermediate Δoct
value), the transition-metal compound may exhibit LS ↔
HS bistability through the application of an external stimu-
lus, like temperature, pressure or light.[4–7] Hence, such SCO
[a] Departament de Química Inorgànica, QBI,
Universitat de Barcelona,
Martí I Franquès 1–11, 08028 Barcelona, Spain
Fax: +34-93-490-7725
E-mail: patrick.gamez@qi.ub.es
Homepage: www.bio-inorganic-chemistry-icrea-ub.com
[b] Materials Chemistry Research Unit, Department of Chemistry
and Center of Excellence for Innovation in Chemistry, Faculty
of Science, Khon Kaen University,
In the present study, we have taken advantage of the
straightforward and highly selective substitution of the
chloride atoms of 2,4,6-trichloro-1,3,5-triazine to synthesize
two related dipyridylamino-substituted-triazine ligands,
Khon Kaen 40002, Thailand
E-mail: sujittra@kku.ac.th
[c] Instituto de Ciencia de Materiales de Aragón (ICMA),
CSIC and Universidad de Zaragoza,
Plaza San Francisco s/n, 50009 Zaragoza, Spain
E-mail: roubeau@unizar.es
namely
4,6-dichloro-N,N-di(pyridine-2-yl)-1,3,5-triazine-
[d] Institució Catalana de Recerca i Estudis Avançats (ICREA),
Passeig Lluís Companys 23, 08010 Barcelona, Spain
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201201085.
amine (Cldpat) and 6-chloro-NЈ-phenyl-N,N-di(pyridin-2-
yl)-1,3,5-triazine-2,4-diamine (Cladpat) (Scheme 1), whose
sole difference lies in the replacement of one of the two
Eur. J. Inorg. Chem. 2013, 730–737
730
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
chloride atoms of Cldpat by an aniline unit, producing tion of Fe(NCX)2 (X = S or Se) to a methanolic solution
Cladpat. Actually, the use of these similar ligands to bind containing 2 equiv. of the dipyridylamino-substituted-tri-
iron(II) ions leads to the formation of mononuclear coordi- azine ligand (Cldpat or Cladpat). The solution of Fe-
nation compounds with similar molecular structures, but (NCS)2 was obtained from iron(II) sulfate and potassium
dissimilar magnetic behaviours, thus revealing the drastic thiocyanate while Fe(NCSe)2 was made from iron(II) per-
effect of tiny structural changes on the physical properties chlorate and potassium selenocyanate.
of related molecules.
Description of the Crystal Structure of trans-[Fe(Cldpat)2-
(NCS)2](H2O) (1)
Reaction of 1 equiv. of iron(II) thiocyanate with 2 equiv.
of Cldpat generates compound 1 with a yield of 73%. Struc-
tural information on 1 has been obtained from X-ray dif-
fraction studies at 100 K. The high R value found (see
Table S2) is most likely due to unresolved crystal twinning.
Nevertheless, the data collected are satisfactory for a rea-
sonably accurate determination of the molecular structure
of 1. Actually, the iron(II) centre in 1 exhibits the antici-
pated octahedral coordination environment [typically ob-
served for iron(II) thiocyanate complexes with this family
of ligands[25,34]], which is formed by two Cldpat ligands in
the equatorial plane and two trans thiocyanate anions (Fig-
Scheme 1. Triazine-based ligands 4,6-dichloro-N,N-di(pyridine-2-
yl)-1,3,5-triazine-amine (Cldpat) and 6-chloro-NЈ-phenyl-N,N-di-
(pyridin-2-yl)-1,3,5-triazine-2,4-diamine (Cladpat).
Results and Discussion
Synthesis
As evidenced by earlier studies by Murray[26–29] and
some of us,[22,24,25] s-triazine-based ligands containing at
least one 2,2Ј-dipyridylamine unit allow the preparation of
SCO iron(II) compounds. Herein, it was decided to investi-
gate the potential ability of the simplest member of this
family of dipyridylamino-substituted-triazine ligands,
namely
4,6-dichloro-N,N-di(pyridine-2-yl)-1,3,5-triazine-
Figure 1. Representation of the molecular structure of compound
1 with partial atom numbering scheme. The thermal ellipsoids are
drawn at the 50% probability level. The hydrogen atoms and the
lattice water molecule are not shown for clarity. Symmetry opera-
tion: a, 1 – x, 1 – y, –z.
amine (Cldpat; Scheme 1), to generate SCO properties upon
coordination to an iron(II) ion in the presence of thio-
cyanates or selenocyanates. Hence, Cldpat was prepared in
THF by the reaction of 2,2Ј-dipyridylamine with 1 equiv. of
2,4,6-trichloro-1,3,5-triazine, in the presence of N,N-diiso-
propylethylamine (DIPEA).
Table 1. Coordination bond lengths [Å] and angles [°], and supra-
molecular interactions for compound 1.
Next, one of the chloride atoms of Cldpat was replaced
by an aniline group with the aim of examining the influence Distances
of this slight modification of the ligand on the magnetic
properties of the ensuing iron(II) complex. Thus, the ligand
6-chloro-NЈ-phenyl-N,N-di(pyridin-2-yl)-1,3,5-triazine-
2,4-diamine (Cladpat; Scheme 1) was synthesized from
Cldpat through substitution of one of its chloride atoms by
aniline, using sodium carbonate as a base in acetone/water.
The molecular structure of Cladpat could be determined by
single-crystal X-ray diffraction (Figure S1). The solid-state
Fe1–N1
Fe1–N2
2.090(13)
2.214(13)
Fe1–N3
Fe1···Fe1inter
2.264(15)
8.396(16)
[a]
Angles
N2–Fe1–N3
81.7(5)
98.3(5)
180
N3a–Fe1–N2[b]
N2a–Fe1–N3a
98.3(5)
81.7(5)
N2a–Fe1–N3
N1–Fe1–N1a[b]
ΣFe1°[c]
43
Φ°[d]
74
structure of Cladpat shows that the molecules are strongly Hydrogen bond
associated by strong hydrogen-bonding interactions
[Naniline–H···Ntriazine = 2.972(3) Å; Є Naniline–H–Ntriazine
Cl3···O2So
2.739(5)
=
Lone pair–π interactions
Cg10···S2s[e]
3.242(7)
176(3)°; Figure S1], giving rise to a 1D supramolecular
chain along the crystallographic c axis. Such anticipated H
bonds, for which the ligand Cladpat was actually designed,
may be crucial to favour the occurrence of cooperative
SCO, if maintained in complexes of Cladpat.
Cg10···S2
3.337(8)
[a] Closest inter-monomer Fe···Fe distance. [b] Symmetry opera-
tion: a 1 – x, 1 – y, –z. [c] Σ° = the sum of |90 – θ| for the 12 N–
Fe–N angles in the octahedron.[12,35,36] [d] Φ° = sum of |60 – θ| for
the 24 N–Fe–N angles describing the trigonal twist angle as de-
scribed by Marchivie and coworkers.[12,37] [e] Symmetry operation:
The coordination compounds 1–3 were synthesized by
the direct addition of a freshly prepared methanolic solu- o – x, 1 – y, 1 – z; s – x, 1/2 + y, 1/2 – z.
Eur. J. Inorg. Chem. 2013, 730–737
731
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
ure 1). Selected bond lengths and angles are listed in
Table 1. Surprisingly, The Fe–NPy bond lengths in the range
2.214(13)–2.264(15) Å and the Fe–NNCS distances of
2.090(13) Å characterize a high-spin iron(II) compound.
This fact is confirmed by temperature-dependent magnetic-
susceptibility measurements revealing that the FeN6 species
in 1, rather unexpectedly, does not present a SCO process
(see below).
The solid-state structure of 1 shows the occurrence of
intramolecular lone pair···π interactions [Cg10···S2 =
3.337(8) Å; Table 1]. In addition, the molecules are associ-
ated by means of intermolecular lone pair···π interactions
[Cg10···S2s = 3.242(7) Å; Table 1], which generate a supra-
molecular 1D chain. The chains interact with each other
through Slonepair···triazineπ contacts, producing a 2D layer
in the crystallographic bc plane (Figures 2 and S2). These
2D layers are further connected through hydrogen bonds
between lattice water molecules and the chloride atoms of
the triazine rings that give rise to a 3D network (Figure S3).
Figure 3. Representation of the molecular structure of compound
2 (LS state, determined at 100 K) with partial atom numbering
scheme. The thermal ellipsoids are drawn at the 50% probability
level, and only the hydrogen atoms involved in hydrogen-bonding
interactions are shown for clarity.
Table 2. Coordination bond lengths [Å] and angles [°], and supra-
molecular interactions for compound 2 (low-spin and high-spin
states).
2
LS[a]
HS[b]
Distances
Fe1–N1
Fe1–N2
Fe1–N3
Fe1–N4
Fe1–N10
Fe1–N11
Fe1···Fe1inter
1.948(4)
1.945(4)
1.984(3)
1.994(4)
1.965(3)
2.000(4)
8.424(4)
2.090(6)
2.087(6)
2.190(5)
2.201(6)
2.178(5)
2.214(6)
8.584(6)
[c]
Angles
Figure 2. View of the crystal packing of 1 showing the formation
of a supramolecular 2D network in the bc plane by means of lone
pair···π interactions (black dotted lines) between the triazine rings
and neighbouring thiocyanate anions [Cg10···S2 = 3.337(8) Å and
Cg10···S2s = 3.242(7) Å]. Symmetry operation: s – x, 1/2 + y,
1/2 – z.
N3–Fe1–N4
86.27(15)
93.68(15)
86.86(15)
93.19(15)
176.6(2)
82.7(2)
97.7(2)
82.3(2)
97.4(2)
176.6(2)
N4–Fe1–N10
N10–Fe1–N11
N11–Fe1–N3
N1–Fe1–N2
ΣFe1°[d]
Φ°[e]
32
43
47
70
Description of the Crystal Structure of trans-[Fe(Cladpat)2-
(NCS)2] (2)
Hydrogen bonds[f]
N9–H9···N15a
3.010(5)
3.035(8)
165(6)
3.023(8)
157(6)
Reaction of 1 equiv. of iron(II) thiocyanate with 2 equiv.
of Cladpat produces compound 2 with a yield of 89%.
Compound 2, which shows SCO properties (see magnetic
studies), crystallizes in the monoclinic space group P21/c,
both at 100 K and at 270 K (Table S2). The molecular
structure of low-spin 2 is depicted in Figure 3, and selected
bond lengths and angles are listed in Table 2. Compound 2
consists of an octahedral iron(II) ion coordinated by four
pyridine units from two Cladpat ligands and two trans N-
bonded thiocyanate anions.
Є N9–H9–N15a
N16–H16···N8b
Є N16–H16–N8b
177(6)
2.995(5)
157(5)
Lone pair–π interactions
Cg5···S2
Cg8···S1
3.459(2)
3.504(2)
3.589(4)
3.564(3)
π–π interactions
Cg9···Cg10
3.976(3)
4.236(5)
[a] At 100 K. [b] At 270 K. [c] Closest inter-monomer Fe···Fe dis-
tance. [d] Σ° = the sum of |90 – θ| for the 12 N–Fe–N angles in the
octahedron.[12,35,36] [e] Φ° = sum of |60 – θ| for the 24 N–Fe–N
angles describing the trigonal twist angle as described by Marchivie
Such a FeN6 coordination environment is known to po-
tentially produce SCO iron(II) species.[35] The Fe–NPy dis-
tances ranging from 1.965(3) to 2.000(4) Å are typical of a and coworkers.[12,37] [f] Symmetry operations: a 1 – x, –1/2 + y,
1/2 – z; b 1 + x, 1/2 – y, 1/2 + z.
LS iron(II) entity, as is the case for the Fe–NNCS distances
of 1.945(4) and 1.948(4) Å (Table 2). These bond lengths
increase by ca. 0.20 Å for Fe–NPy and ca. 0.15 Å for Fe– which is indicative of a full spin transition, as also observed
NNCS when the temperature is raised to 270 K (Table 2), during magnetic studies on 2 (see below).
Eur. J. Inorg. Chem. 2013, 730–737
732
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
The distortion parameters Σ° and Φ° reflect the devia- may favour cooperativity between the transiting iron(II)
tion from an ideal octahedral geometry.[38] Σ° directly mea- ions.
sures the distortion from a perfect octahedron while Φ° de-
It has to be noted that the Σ° and Φ° values for HS 2 are
fines the deformation of the octahedral coordination geom- comparable to those of 1 (see Table 1), therefore suggesting
etry towards a trigonal-prismatic environment.[12] For a per- analogous octahedral distortions for the two high-spin mo-
fect octahedron, Σ° = Φ° = 0. For LS 2, Σ° = 32 and Φ° = lecules, and corroborating the metal coordination environ-
43, whereas the respective values for HS 2 are 43 and 70 ment in 1.
(Table 2). The lower Σ° and Φ° values for the LS compound
are typical since LS complexes are less deformed than HS cyanate [instead of iron(II) thiocyanate] yields the analo-
ones.[35,37] Hence, ΔΣ° (Σ°HS – Σ°LHS) and ΔΦ° (Φ°HS
gous compound [Fe(Cladpat)2(NCSe)2] (3), as indicated by
Reaction of the ligand Cladpat with iron(II) seleno-
–
Φ°LS) quantify the magnitude of the structural changes oc- elemental analyses, which also exhibits SCO properties (see
curring during the SCO. For 2, ΔΣ° = 15 and ΔΦ° = 27; the the section Magnetic Studies). Although the molecular
high ΔΦ° value observed for 2 is indicative of a significant structure of 3 could not be determined by X-ray diffraction
alteration of the octahedral geometry upon the LS Ǟ HS studies, it is expected that the coordination environment of
transition. Such a large structural variation may be associ- the iron(II) centres in 3 is equivalent to those of 1 and 2,
ated with cooperativity between the iron(II) centres, poss- with two trans N-bonded selenocyanate ions. Actually, the
ibly through intermolecular interactions (actually, the mag- IR absorption band observed at 2061 cm–1 for compound 3
netic measurements show a relatively cooperative SCO; see is identical to that reported by Murray and coworkers for
below). In fact, the crystal packing reveals that the iron(II) an iron(II) SCO complex from a dipyridylamino-substi-
molecules are connected through double Naniline
–
tuted-triazine ligand with trans-coordinated NCSe ions.[30]
H···Ntriazine bonds (Table 2), which generate a 1D supra-
molecular chain (Figure 4). These hydrogen bonds are af-
fected by the spin transition (a variation of about 0.025 Å
is observed); in particular, the N9–H9···N15a bond experi-
ences a bending of ca. 7%, the angle varying from 177(6)°
to 165(6)° during the LS Ǟ HS transition (Table 2). In ad-
dition, the molecules of 2 exhibit intramolecular lone
pair···π interactions[39,40] (Cg5···S2 and Cg8···S1 contacts;
Figure 4), which are also altered by the SCO (Table 2). A
closer look at the solid-state structure of 2 reveals that the
1D supramolecular chains are linked into a 2D layer by
parallel-displaced π–π interactions,[41,42] involving neigh-
bouring aniline rings [centroid-to-centroid distance
Cg9···Cg10 = 3.976(3) Å in LS 2; Figure S4]. The shortest
arene–arene contact distances are C17···C39i = 3.257(7) Å
and C18···C40i = 3.335(7) Å (Figure S4B), and rise to
3.283(12) Å and 3.397(11) Å, respectively, upon LS Ǟ HS
transition [corresponding to an increase of the centroid-to-
centroid distance of ca. 7%, from 3.976(3) to 4.236(5) Å;
Table 2]. All these supramolecular bonding interactions
Magnetic Studies
The effect of the substituent (ligands Cldpat and Clad-
pat) and the replacement of thiocyanate by selenocyanate
anions have been investigated by magnetic measurements.
Hence, the temperature dependence of the χT product of
compounds 1–3, χ being the molar paramagnetic suscep-
tibility, were derived from magnetization measurements on
bulk samples in an applied field of 0.5 T for 1 and 3, and
1 T for 2 (see Experimental Section), and in the temperature
range 2–300 K. The data for compounds 1 and 2, shown in
Figure 5, corroborate the structural observations, evidenc-
ing the occurrence of a complete thermal SCO for com-
pound 2, and the – surprising – absence of such a process
in the case of compound 1. Indeed, the χT product of 1
remains practically constant at 3.11–3.09 cm3 mol–1 K from
300 K down to 50 K, the temperature below which a de-
Figure 4. View of the crystal packing of LS 2 showing Naniline
–
H···Ntriazine bonds [N9–H9···N15a 3.010(5) Å and N16a–
=
H16a···N8 = 2.995(5) Å; blue dotted lines] connecting the iron(II)
molecules to generate a 1D supramolecular chain. Intramolecular
lone pair···π interactions (black dotted lines) take place between
the thiocyanate sulfur atoms and the triazine rings [Cg5···S2 =
3.459(2) Å and Cg8···S1 = 3.504(2) Å]. Symmetry operations: a
1 – x, –1/2 + y, 1/2 – z.
Figure 5. χT vs. T plot for 1 (squares) and 2, showing the process
of SCO (full circles), the partial LS to HS photo-induced trapping
at 10 K (LIESST effect) and the relaxation back to the LS ground
state and normal behaviour upon warming (empty circles).
Eur. J. Inorg. Chem. 2013, 730–737
733
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
1
crease of χT is observed down to 2.18 cm3 mol–1 K. The and A1Ǟ3T1 (LS) bands, resulting in competitive LIESST
magnetic properties of 1 are thus in agreement with an S = and reverse-LIESST processes.[44] Another cause of such an
2 HS ground state for the FeII ion, with g close to 2, incomplete light-induced trapping, even in thin samples, is
throughout the whole temperature range considered, the de- the competition of the light-induced LSǞHS trapping with
crease at low temperatures being due to zero-field-splitting the relaxation of the trapped HS species back to the LS
effects of the S = 2 spins. Magnetization vs. field measure- ground state, at the origin of light-induced bistability in co-
ments at 2 K indeed show saturation at ca. 4.05 NAμB at operative SCO systems.[45,46] Measurements vs. time at 10 K
5 T (see Figure S5), in agreement with an S = 2 spin ground after 1 h of irradiation for 2 indeed show that χT rapidly
state and a g value slightly above 2. On the other hand, the drops in an exponential manner, already reaching values of
χT product of 2 decreases upon cooling, from similar values ca. 0.40 cm3 mol–1 K after only 1 h. It is therefore clear that
to those of 1, i.e. 3.13 cm3 mol–1 K at 280 K, reaching a pla- the HSǞLS relaxation is fast, even at 10 K (Figure 6 bot-
teau at 0.11 cm3 mol–1 K below 100 K, and down to tom). An estimation of the characteristic relaxation time at
0.08 cm3 mol–1 K at 5 K. Compound 2 thus undergoes a 10 K of 765 s is obtained by adjusting the experimental data
complete SCO centred at ca. 178 K and with a limited co- to a stretched exponential with β = 0.67 (red line in Figure 6
operative character, as indicated by a ΔT80 of 50 K (80% of bottom). Confirmation of the relative instability of the
the transition occurs within about 50 K). In agreement with trapped HS state is obtained from the measurements upon
the structural observations, the process of SCO is virtually warming after irradiation (Figure 5). χT rapidly decreases
completed at 100 K. These observations are perfectly repro- to reach values similar to those of a normal LS state at
ducible upon warming, thus with no detectable hysteresis, temperatures as low as 55 K, as could be expected from the
and over various cycles.
thermal activation of a relaxation process that is already
The possibility of trapping the HS metastable state at fast at 10 K.
low temperature through irradiation, the so-called LIESST
Magnetic properties of compound 3 are shown in Fig-
effect,[43] was examined on a thin sample of 2. At 10 K, a ure 7 as a χT vs. T plot. Similar to 2, a decrease of χT is
fast increase of χT is indeed observed when irradiating with observed, from 3.43 cm3 mol–1 K at 280 K, down to a pla-
green light, clearly demonstrating the efficiency of the teau at ca. 0.15–0.11 cm3 mol–1 K below 90 K and down to
LIESST effect in this compound (Figures 5 and 6 top). 0.11 cm3 mol–1 K at 5 K, thus confirming a likely compar-
Nevertheless, the rate of increase rapidly drops and virtually able molecular structure to that of 2. Compound 3 thus
stable values are reached after ca. 1 h (Figure 6 top). After also presents a complete SCO, although centred at a lower
turning the irradiation OFF and thermalization, the re- temperature, e.g. 166 K, and more gradual than that of 2,
sulting χT value of 1.42 cm3 mol–1 K corresponds to ca. with a ΔT80 of ca. 80 K. Such a low cooperativity in a sele-
50% of trapped HS centres, based on the χT value of 1 at nocyanate compound with respect to its thiocyanate ana-
the same temperature. An incomplete LIESST effect in thin logue is documented, and is possibly related to the partici-
samples with an apparent steady state similar to that ob- pation of the more diffuse Se atom in intermolecular inter-
served herein may be due to an overlap of the 5T2Ǟ5E (HS) actions.[47,48] This is likely the case here since the S atoms
in the structure of 2 do participate in the network of inter-
molecular interactions (see above). The lower SCO tem-
perature in 3 with respect to 2 is more surprising since the
replacement of S by Se in NCX–-based SCO compounds
usually results in an increase in the SCO temperature.[47–51]
Figure 6. Top: χT vs. time evolution for 2 at 10 K showing the
increase due to LIESST during irradiation with a green light. Bot-
tom: χT vs. time evolution for 2 at 10 K in obscurity after the
irradiation with a green light, showing an exponential-like behav-
iour. The full red line is a fit to a stretched exponential with β =
0.67 and critical time of 765 s.
Figure 7. χT vs. T plot for 3 (rhombus), showing a complete SCO
centred at around 166 K. The data of compound 2 is recalled as
light circles for comparison.
Eur. J. Inorg. Chem. 2013, 730–737
734
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Thermal Properties
the chloride atoms of Cldpat by an aniline group (ligand
Cladpat) gives rise to a drastic change of the magnetic prop-
erties of the corresponding iron(II) complexes. Indeed, the
coordination compounds 1 and 2, which exhibit identical
coordination spheres for their metal centre, display distinct
magnetic behaviours; 1 is a high-spin species whereas 2 is a
SCO complex with a (weakly cooperative) transition. These
opposite comportments are obviously due to the different
ligands. Since the coordination bond lengths and angles are
comparable for 1 and HS 2, the distinct magnetic properties
observed are most likely caused by different ligand-field
strengths (i.e. Cldpat has a smaller Δoct value). Theoretical
studies and the preparation of dipyridylamino-substituted
triazine ligands bearing electron-withdrawing and/or elec-
tron-donating substituents may help to understand these
disparities. Therefore, these investigations will be carried
out and the results will be reported in a future paper.
The molar heat capacity at constant pressure, Cp, of 2
was derived from differential scanning calorimetry (DSC)
measurements over the temperature range 110–290 K (see
DSC traces in Figure S6). A strong heat capacity anomaly
is detected between 135 and 245 K, culminating at ca.
180 K, which can be associated with the SCO phenomenon
in 2. Indeed, both the temperature range and maxima are
in excellent agreement with the magnetic data. A lattice heat
capacity was estimated from data below 135 K and above
245 K (dashed line in Figure 8), allowing the determination
of the excess heat capacity associated with the SCO phe-
nomenon in 2 (inset in Figure 8). The related excess en-
thalpy and entropy were derived by integration of the excess
heat capacity and amount to 8.08 kJmol–1 and 44.6
Jmol–1 K–1, respectively (see Figures S7 and S8). Both fig-
ures are relatively large, which is usually taken as a conse-
quence of a cooperative character of the SCO. In particular,
the excess entropy is well above the purely electronic com-
ponent, RLn5, thus containing a significant content arising
from the coupling of the electronic transition with lattice
phonons. The so-called domain model (developed by Sorai
and widely used in SCO studies when calorimetric data are
available[52]) allows such a cooperative character to be
quantified through the number n of like-spin SCO centres
within an interacting domain; the larger the domains are,
the more cooperative the transition is. Values of n close to
1 are typical of solution-like gradual SCO,[53–55] while val-
ues of ca. 10 to 95[47,56,57] have been derived for cooperative
to very cooperative systems. Here, the excess heat capacity
of 2 is very nicely reproduced by this model with n = 3.33
(red line in the inset of Figure 8), indicative of a relatively
weakly cooperative SCO, and in agreement with the mag-
netic studies.
Experimental Section
General: All reagents and solvents were used as received from com-
mercial sources. The reactions were typically carried out in air. IR
spectra (as KBr pellets) were recorded with a Nicolet 5700 FTIR
spectrometer. Elemental analyses were performed by the Servei de
Microanalisi, Consejo Superior de Investigaciones Cientificas
(CSIC) of Barcelona. MS spectra were recorded with a MALDI-
TOF Voyager DE-RP mass spectrometer equipped with a nitrogen
laser (337 nm, 3 ns pulse) and an accelerating voltage of 20–25 kV
or an LC/MSD-TOF ESI-mass spectrometer from Agilent Tech-
nologies, at the Serveis Cientificotècnics of the University of Barce-
1
lona. H NMR spectra were recorded at room temperature with a
Varian Unity 300 MHz spectrometer; chemical shifts are reported
in ppm relative to the residual solvent signal of CDCl3 (δ =
7.26 ppm). TLC was performed on Alugram® SIL G/UV/254 silica-
gel precoated sheets (Macherey–Nagel, Germany).
Single-Crystal X-ray Diffraction Studies: Data for Cladpat, 1 and
2 (LS and HS states) were collected using Mo-Kα radiation (λ =
0.7107 Å) with a Bruker APEX II QUAZAR diffractometer
equipped with a microfocus multilayer monochromator at T =
100(2) K and at T = 270(2) K for HS 2. The structures were solved
by direct methods and refined on F2 using the SHELX-TL suite.[58]
In the case of 1, although the data collected appeared to be of good
quality and the processing took place normally, the refinement
could only converge to high final R values, and required displace-
ment parameter restraints for most of the organic ligands. This
situation is likely due to the presence of twinning; unfortunately,
we were not able to settle it. Therefore, even though the molecular
structure and the intermolecular interactions are likely accurate,
the X-ray structure described herein for 1 should be considered as
a preliminary report. Crystal data and refinement parameters are
given in Tables S1 (Cladpat) and S2 (1, LS 2 and HS 2). CCDC-
901348 (Cladpat), -901349 (1), -901350 (2, 100 K) and -901351 (2,
270 K) contain the supplementary crystallographic data for this pa-
per. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Figure 8. Molar heat capacities of 2 showing a broad hump associ-
ated with the SCO. The dashed line is the estimated lattice compo-
nent. Inset: excess heat capacities associated with the SCO in 2.
The full line is a fit to the domain model of Sorai (see text and SI)
with n = 3.3.
Magnetic Measurements: Variable-temperature magnetic-suscep-
tibility data for compound 1–3 were obtained on microcrystalline
samples with a Quantum Design MPMS-XL SQUID magnetome-
Conclusions
The present study has shown that an apparent minor
modification of a ligand, namely the substitution of one of ter housed at the SAI Physical Measurements of the University of
Eur. J. Inorg. Chem. 2013, 730–737
735
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Zaragoza. Variable-temperature measurements were performed in trans-[Fe(Cldpat)2(NCS)2](H2O) (1): A methanolic solution (5 mL)
a 0.5 T applied field, still in the linear range of M vs. H. Pascal’s
constants were used to estimate diamagnetic corrections to the mo-
lar paramagnetic susceptibility, and a correction was applied for
the sample holder. Warming and cooling rates were of the order
of 0.3 Kmin–1. Irradiation experiments were performed using the
Quantum Design fibre optics setup (FOSH). Using this setup, mea-
surements at low fields were noisy, and the applied field was 1 T
throughout the whole study. This field was found to be the best
compromise to remain close to the linear range of M vs. H curves
and obtain good quality data. The light source was a Xenon arc
lamp equipped with sets of short-pass and long-pass filters (SPF
or LPF). Specifically for the present study, a SPF 650 nm and a
LPF at 500 nm were used. Data were corrected for the empty
FOSH signal, determined beforehand.
of KNCS (0.019 g; 0.2 mmol) was added to an aqueous solution
(2 mL) of FeSO4·7H2O (0.028 g; 0.1 mmol). After 15 min of stir-
ring, the precipitate of K2SO4 was removed by filtration. Ascorbic
acid (in a small quantity) was added to the filtrate to prevent oxi-
dation to iron(III). Subsequently, this iron(II) solution was added
to a solution of Cldpat (0.064 g; 0.2 mmol) in dichloromethane
(15 mL). The resulting yellow reaction mixture was filtered and the
filtrate was left unperturbed for the slow evaporation of the solvent.
After two days, small yellow crystals of 1 were obtained (0.605 g,
0.730 mmol, 73%). C28H16Cl4FeN14S2 (1–H2O): calcd. C 41.50, H
1.99, N 24.20; found C 40.94, H 1.78, N 24.23. IR (KBr): ν = 3239
˜
(w), 3079 (w), 1608 (m), 1550 (s), 1464 (s), 1382 (s), 1226 (m), 988
(m), 804 (w), 767 (w), 662 (w) cm–1.
trans-[Fe(Cladpat)2(NCS)2] (2): A methanolic solution (5 mL) of
KNCS (0.019 g; 0.2 mmol) was added to an aqueous solution
(2 mL) of FeSO4·7H2O (0.028 g; 0.1 mmol). After 15 min of stir-
ring, the precipitate of K2SO4 was removed by filtration. Ascorbic
acid (in a small quantity) was added to the filtrate to prevent oxi-
dation to iron(III). Next, a solution of Cldpat (0.075 g, 0.2 mmol)
in dichloromethane (5 mL) was added to the iron(II) solution. The
resulting yellow reaction mixture was filtered and the filtrate was
left unperturbed for the slow evaporation of the solvent. After 2 d,
small yellowish-green single crystals of 2, suitable for X-ray diffrac-
tion studies, were obtained (0.082 g; 0.089 mmol, 89%; based on
iron). C40H28Cl2FeN16S2 (923.64): calcd. C 52.01, H 3.06, N 24.26;
Differential Scanning Calorimetry Measurements: DSC experiments
were performed with a differential scanning calorimeter Q1000
with the LNCS accessory from TA Instruments. The temperature
and enthalpy scales were calibrated with a standard sample of in-
dium, using its melting transition (156.6 °C, 3296 Jmol–1). Mea-
surements were carried out using aluminium pans with a mechani-
cal crimp, with an empty pan as reference. The zero-heat flow pro-
cedure described by TA Instruments was followed to derive heat
capacities, using synthetic sapphire as the reference compound. An
overall accuracy of ca. 0.2 K in temperature and up to 5 to 10%
in the heat capacity was estimated over the whole temperature
range, by comparison with the synthetic sapphire.
found C 52.11, H 3.02, N 24.34. IR (KBr): ν = 3433 (br.), 3244
˜
(w), 2061 (s), 1603 (m), 1558 (s), 1462 (s), 1417 (s), 1383 (m), 1226
4,6-Dichloro-N,N-di(pyridine-2-yl)-1,3,5-triazin-2-amine (Cldpat):
2,4,6-Trichloro-1,3,5-triazine (cyanuric chloride) (2.7 g, 14.6 mmol)
was dissolved in THF. N,N-Diisopropylethylamine (DIPEA)
(2.55 mL, 1.89 g; 14.6 mmol) was added whilst stirring. The re-
sulting yellow solution was cooled down to 0 °C using an ice bath.
Subsequently, a solution of 2,2Ј-dipyridylamine (5.10 g, 14.6 mmol)
in THF (50 mL) was added dropwise, and the reaction mixture was
stirred at this temperature for 1 h. The yellow precipitate obtained
(i.e. hydrochloride salt of DIPEA) was separated by filtration, and
the filtrate was concentrated under reduced pressure using a rotary
evaporator at 35 °C. The resulting crude product was purified by
column chromatography on silica gel with ethyl acetate as the elu-
ent (Rf = 0.54) to give pure Cldpat as a white powder (4.44 g,
13.9 mmol, 95%). C13H8Cl2N6 (319.15): calcd. C 48.92, H 2.53, N
26.33; found C 48.90, H 2.48, N 26.21. 1H NMR (300 MHz,
CDCl3, room temp.): δ = 7.24–7.33 (m, 2 H), 7.54 (dt, J = 8.1, J
= 0.9 Hz, 2 H), 7.80–7.89 (m, 2 H), 8.50 (ddd, J = 4.9, J = 1.9, J
= 0.8 Hz, 2 H) ppm. MS (ESI+): m/z = 320.17 [M + H]+. IR (KBr):
(m), 996 (w), 800 (w) cm–1.
[Fe(Cladpat)2(NCSe)2] (3): A small quantity of ascorbic acid was
added to a methanolic solution (2 mL) of Fe(ClO4)2·6H2O (0.064 g,
0.25 mmol). A methanolic solution (10 mL) of Cldpat (0.188 g,
0.5 mmol) was subsequently added. Finally, a solution of KNCSe
(0.072 g, 0.5 mmol) in methanol (10 mL) was added and the re-
sulting reaction mixture was filtered. The filtrate was left unper-
turbed for the slow evaporation of the solvent. After 2 d, small
yellow single crystals of 3 were obtained (0.207 g, 0.203 mmol,
81%). C40H28Cl2FeN16Se2 (1017.44): calcd. C 47.22, H 2.77, N
22.03; found C 47.19, H 2.51, N 21.83. IR (KBr): ν = 3447 (br.),
˜
3245 (w), 2061 (s), 1605 (s), 1560 (s), 1461 (m), 1416 (s), 1383 (m),
1225 (m), 996 (w), 799 (w) cm–1.
Supporting Information (see footnote on the first page of this arti-
cle): Crystallographic data for the ligand Cladpat (Table S1), com-
pounds 1, LS 2 and HS 2 (Table S2); representation of the molecu-
lar structure of Cladpat (Figure S1); views of the crystal packing of
1 (Figures S2 and S3); view of the crystal packing of 2 (Figure S4);
representation of the magnetization vs. applied field at 2 K for com-
pound 1 (Figure S5); DSC traces of compound 2 for warming and
cooling modes (Figure S6); Excess enthalpy and entropy involved
in the SCO process in compound 2 (Figure S7); Details of modelli-
zation of ΔCp data with the so-called domain model.
ν = 1585 (s), 1548 (br.), 1491 (s), 1461 (s), 1430 (s), 1324 (m), 1220
˜
(s), 1182 (s), 842 (s), 796 (m), 778 (m), 746 (m), 667 (m) cm–1.
6-Chloro-NЈ-phenyl-N,N-di(pyridin-2-yl)-1,3,5-triazine-2,4-diamine
(Cladpat): Aniline (0.57 mL, (0.58 g, 6.27 mmol) was added to a
solution of Cldpat (2.00 g, 6.27 mmol) in acetone (35 mL). Sub-
sequently, a solution of Na2CO3 (0.30 g, 3.14 mmol) in water
(10 mL) was added dropwise, giving rise to a white precipitate. Af-
ter a reaction time of 6 h, the content was poured into crushed ice
and the resulting mixture was stirred until all of the ice had melted.
The white precipitate of pure Cladpat was isolated by filtration,
washed with water and dried overnight under vacuum (2.33 g,
6.21 mmol, 99%). C19H14ClN7 (375.82): calcd. C 60.72, H 3.75, N
Acknowledgments
P. G. acknowledges ICREA (Institució Catalana de Recerca i Es-
tudis Avançats) and the Ministerio de Economía y Competitividad
26.09; found C 60.64, H 3.67, N 25.96. 1H NMR (300 MHz, of Spain (Project CTQ2011-27929-C02-01). The Royal Golden Ju-
CDCl3, room temp.): δ = 6.97–7.33 (m, 8 H), 7.60 (t, J = 16.9, J = bilee Ph. D Program (PHD/0234/2550), the Thailand Research
1.8 Hz, 2 H), 7.82 (td, J = 7.9, J = 1.8 Hz, 2 H), 8.49 (d, J = 3.8 Hz, Fund and Khon Kaen University, the Higher Education Research
2 H) ppm. MS (ESI+): m/z = 376 [M + H]+. IR (KBr): ν = 3239
Promotion and National Research University Project of Thailand
and the Center of Excellence for Innovation in Chemistry
(PERCH-CIC), Office of the Higher Education Commission, Min-
˜
(w), 3079 (w), 1608 (m), 1550 (s), 1464 (s), 1382 (s), 1226 (m), 988
(m), 804 (w), 767 (w), 662 (w) cm–1.
Eur. J. Inorg. Chem. 2013, 730–737
736
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[28] T. M. Ross, B. Moubaraki, D. R. Turner, G. J. Halder, G. Chas-
tanet, S. M. Neville, J. D. Cashion, J. F. Létard, S. R. Batten,
K. S. Murray, Eur. J. Inorg. Chem. 2011, 1395–1417.
[29] T. M. Ross, B. Moubaraki, K. S. Wallwork, S. R. Batten, K. S.
Murray, Dalton Trans. 2011, 40, 10147–10155.
istry of Education are also acknowledged. O. R. acknowledges
funding from the Ministerio de Economía y Competitividad of
Spain (Project MAT2011-24284).
[1] B. N. Figgis, M. A. Hitchman, in Ligand Field Theory and Its
Applications, Wiley-VCH, New York, 2000.
[30] S. M. Neville, B. A. Leita, D. A. Offermann, M. B. Duriska, B.
Moubaraki, K. W. Chapman, G. J. Halder, K. S. Murray, Eur.
J. Inorg. Chem. 2007, 1073–1085.
[2] A. B. Koudriavtsev, W. Linert, J. Struct. Chem. 2010, 51, 335–
365.
[31] T. J. Mooibroek, P. Gamez, Inorg. Chim. Acta 2007, 360, 381–
[3] A. B. P. Lever, in Werner Centennial (Ed.: G. B. Kauffman),
ACS Publications, Washington, USA, 1967, pp. 430–451.
[4] P. Gütlich, A. Hauser, H. Spiering, Angew. Chem. 1994, 106,
2109; Angew. Chem. Int. Ed. Engl. 1994, 33, 2024–2054.
[5] P. Gütlich, H. A. Goodwin, in Spin Crossover in Transition
Metal Compounds I, Springer-Verlag Berlin, Berlin, 2004, vol.
233, pp. 1–47.
404.
[32] P. Gamez, J. Reedijk, Eur. J. Inorg. Chem. 2006, 29–42.
[33] P. de Hoog, P. Gamez, W. L. Driessen, J. Reedijk, Tetrahedron
Lett. 2002, 43, 6783–6786.
[34] H. S. Scott, T. M. Ross, S. R. Batten, I. A. Gass, B. Moubaraki,
S. M. Neville, K. S. Murray, Aust. J. Chem. 2012, 65, 874–882.
[35] P. Guionneau, M. Marchivie, G. Bravic, J. F. Létard, D. Chas-
seau, Top. Curr. Chem. 2004, 234, 97–128.
[36] M. Quesada, F. Prins, E. Bill, H. Kooijman, P. Gamez, O. Rou-
beau, A. L. Spek, J. G. Haasnoot, J. Reedijk, Chem. Eur. J.
2008, 14, 8486–8499.
[6] J. A. Real, A. B. Gaspar, M. C. Muñoz, Dalton Trans. 2005,
2062–2079.
[7] A. Bousseksou, G. Molnar, J. A. Real, K. Tanaka, Coord.
Chem. Rev. 2007, 251, 1822–1833.
[8] A. B. Gaspar, M. Seredyuk, R. Gütlich, Coord. Chem. Rev.
2009, 253, 2399–2413.
[37] M. Marchivie, P. Guionneau, J. F. Létard, D. Chasseau, Acta
Crystallogr., Sect. B-Struct. Sci. 2005, 61, 25–28.
[9] P. Gamez, J. S. Costa, M. Quesada, G. Aromí, Dalton Trans. [38] J. K. McCusker, A. L. Rheingold, D. N. Hendrickson, Inorg.
2009, 7845–7853.
Chem. 1996, 35, 2100–2112.
[39] T. J. Mooibroek, P. Gamez, CrystEngComm 2012, 14, 1027–
1030.
[10] A. Bousseksou, G. Molnar, L. Salmon, W. Nicolazzi, Chem.
Soc. Rev. 2011, 40, 3313–3335.
[11] P. Gütlich, Y. Garcia, H. A. Goodwin, Chem. Soc. Rev. 2000,
[40] T. J. Mooibroek, P. Gamez, CrystEngComm 2012, 14, 3902–
29, 419–427.
3906.
[12] M. A. Halcrow, Chem. Soc. Rev. 2011, 40, 4119–4142.
[13] J. Tao, R. J. Wei, R. B. Huang, L. S. Zheng, Chem. Soc. Rev.
2012, 41, 703–737.
[14] G. Aromí, L. A. Barrios, O. Roubeau, P. Gamez, Coord. Chem.
Rev. 2011, 255, 485–546.
[15] J. Olguin, S. Brooker, Coord. Chem. Rev. 2011, 255, 203–240.
[16] M. Boca, R. F. Jameson, W. Linert, Coord. Chem. Rev. 2011,
255, 290–317.
[17] M. A. Halcrow, Coord. Chem. Rev. 2009, 253, 2493–2514.
[18] J. A. Kitchen, S. Brooker, Coord. Chem. Rev. 2008, 252, 2072–
2092.
[19] K. S. Murray, Eur. J. Inorg. Chem. 2008, 3101–3121.
[20] S. Bonnet, M. A. Siegler, J. S. Costa, G. Molnar, A. Boussek-
sou, A. L. Spek, P. Gamez, J. Reedijk, Chem. Commun. 2008,
5619–5621.
[21] J. S. Costa, K. Lappalainen, G. de Ruiter, M. Quesada, J. K.
Tang, I. Mutikainen, U. Turpeinen, C. M. Grunert, P. Gütlich,
H. Z. Lazar, J. F. Létard, P. Gamez, J. Reedijk, Inorg. Chem.
2007, 46, 4079–4089.
[41] C. A. Hunter, J. K. M. Sanders, J. Am. Chem. Soc. 1990, 112,
5525–5534.
[42] S. A. Arnstein, C. D. Sherrill, Phys. Chem. Chem. Phys. 2008,
10, 2646–2655.
[43] A. Hauser, Top. Curr. Chem. 2004, 234, 155–198.
[44] A. Hauser, J. Chem. Phys. 1991, 94, 2741–2748.
[45] A. Desaix, O. Roubeau, J. Jeftic, J. G. Haasnoot, K. Boukhed-
daden, E. Codjovi, J. Linarès, M. Noguès, F. Varret, Eur. Phys.
J. B 1998, 6, 183–193.
[46] F. Varret, K. Boukheddaden, E. Codjovi, C. Enachescu, J. Lin-
arès, Top. Curr. Chem. 2004, 234, 199–229.
[47] M. Sorai, S. Seki, J. Phys. Chem. Solids 1974, 35, 555–570.
[48] L. Capes, J.-F. Létard, O. Kahn, Chem. Eur. J. 2000, 6, 2246–
2255.
[49] T. Buchen, H. Toftlund, P. Gütlich, Chem. Eur. J. 1996, 2,
1129–1133.
[50] J. A. Real, I. Castro, A. Bousseksou, M. Verdaguer, R. Burriel,
M. Castro, J. Linarès, F. Varret, Inorg. Chem. 1997, 36, 455–
464.
[22] M. Quesada, V. A. de la Pena-O’Shea, G. Aromí, S. Geremia,
C. Massera, O. Roubeau, P. Gamez, J. Reedijk, Adv. Mater.
2007, 19, 1397–1402.
[23] P. Gamez, P. de Hoog, M. Lutz, W. L. Driessen, A. L. Spek, J.
Reedijk, Polyhedron 2003, 22, 205–210.
[24] M. Quesada, P. de Hoog, P. Gamez, O. Roubeau, G. Aromí, B.
Donnadieu, C. Massera, M. Lutz, A. L. Spek, J. Reedijk, Eur.
J. Inorg. Chem. 2006, 1353–1361.
[25] M. Quesada, M. Monrabal, G. Aromí, V. A. de la Pena-
O’Shea, M. Gich, E. Molins, O. Roubeau, S. J. Teat, E. J. Ma-
cLean, P. Gamez, J. Reedijk, J. Mater. Chem. 2006, 16, 2669–
2676.
[51] N. Moliner, M. C. Muñoz, J.-F. Létard, X. Solans, R. Burriel,
M. Castro, O. Kahn, J. A. Real, Inorg. Chim. Acta 1999, 291,
279–288.
[52] M. Sorai, Chem. Rev. 1996, 106, 976–1031.
[53] T. Nakamoto, Z. C. Tan, M. Sorai, Inorg. Chem. 2001, 40,
3805–3809.
[54] O. Roubeau, M. de Vos, A. F. Stassen, R. Burriel, J. G. Haas-
noot, J. Reedijk, J. Phys. Chem. Solids 2003, 64, 1003–1013.
[55] O. Roubeau, M. Evangelisti, E. Natividad, Chem. Commun.
2012, 48, 7604–7606.
[56] M. Sorai, Top. Curr. Chem. 2004, 235, 153–170.
[57] Z. Arcis-Castíllo, S. Zheng, M. A. Siegler, O. Roubeau, S. Be-
doui, S. Bonnet, Chem. Eur. J. 2011, 17, 14826–14836.
[58] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
Received: September 14, 2012
[26] T. M. Ross, B. Moubaraki, S. R. Batten, K. S. Murray, Dalton
Trans. 2012, 41, 2571–2581.
[27] T. M. Ross, B. Moubaraki, S. M. Neville, S. R. Batten, K. S.
Murray, Dalton Trans. 2012, 41, 1512–1523.
Published Online: November 27, 2012
Eur. J. Inorg. Chem. 2013, 730–737
737
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim