This document is the accepted manuscript version of a published work that appeared in final form in Chem. Commun., 2018,
54, 5526, DOI: 10.1039/c8cc01561a
View Article Online
View Journal | View Issue

A robust and unique iron(II) mosaic-like MOF
Estefania Fernandez-Bartolome, Jose´ Santos,
Saeed Khodabakhshi,
Laura J. McCormick, Simon J. Teat, Cristina Saenz de Pipaon,
Jose R. Galan-Mascaro´s, Nazario Mart´ın and Jose Sanchez Costa

A novel extended triazole-based ligand (PM-Tria) has been synthesized and an unprecedented MOF 3D has serendipitously been
formed by assembling iron(II), PM-Tria ligand and fluoride anions.
This MOF contains a perfectly linear one-dimensional {Fe(II)–F}n
bridging chain that shows an antiferromagnetic behaviour. Furthermore, the structure is compared with a 14th century mosaic found
in the Alhambra Palace in Granada showing a surprising symmetry
resemblance.

Coordination polymers, or most often called metal–organic
frameworks (MOF),1 are currently a hot topic in science. As an
illustrative example of their relevance, in 2017 it was estimated
that 6000 novel MOFs are published each year.2 This vast interest
derives from the fact that MOFs show a wide variety of potential
applications, which include gas uptake, catalysis, luminescence,
electrical conductivity or biotechnology among many others.
All these properties arise from their intrinsic porosity and
versatility.3–6 In fact, MOFs are extended molecular materials
formed by metal ions or metal ion clusters bridged by relatively
long organic ligands, thus creating size-tunable structural voids
to absorb guest molecules and act as highly specific molecular
vessels with different absorption capabilities.7,8 MOFs can also
display different reactivities towards targeted molecules based
on the different electronic environments created within the
cavities. In addition, the modular nature of MOFs (via the
combination of inorganic and organic components) makes them
ideally suited to chemical manipulation, aiming at fine-tuning
a

IMDEA Nanociencia, C/ Faraday 9, Ciudad Universitaria de Cantoblanco, 28049,
Madrid, Spain. E-mail: jose.sanchezcosta@imdea.org
b
Advanced Light Source, Berkeley Laboratory, 1, Cyclotron Road, CA 94720,
Berkeley, USA
Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of
Science and Technology (BIST), Avda. Pa¨sos Catalans, 16, 43007, Tarragona,
ı
Spain
d
ICREA, Pg. Llu´ Companys, 23, 08010, Barcelona, Spain
ıs
e
Facultad de Ciencias Qu´
ımicas, Universidad Complutense de Madrid, Madrid,
28040, Spain
† Electronic supplementary information (ESI) available. CCDC 1825247. For ESI and
crystallographic data in CIF or other electronic format see DOI: 10.1039/c8cc01561a

their structure and function.9 Nevertheless, obtaining novel
compositions or novel synthetic methods,10,11 with the aim of
obtaining targeted specific properties for their ultimate technological application, is still a real challenge. Despite the fact that the
colossal database of novel MOFs has enabled the development of
novel synthetic tools and progress of the rational design vs. the
serendipitous assembly, this second method’s randomness is still
offering us beautiful surprises every now and then (Scheme 1).
In this manuscript, a novel Fe(II) MOF is described by
assembling Fe(II), PM-Tria ligands and fluoride anions. The
1D Fe(II)–F chain exhibits a perfect linearity on its growingmode, which allows the expansion of the structure in the third
dimension by means of the PM-Tria ligands. The magnetic
behaviour of this material is also reported. Furthermore, this
molecular architecture astonishingly resembles a 14th Century
Islamic mosaic located at the Alhambra Palace in Granada (Spain)
that shows beautiful symmetric patterns.12 With the aim of obtaining large cavities coupled to metal centers that may be susceptible to
achieve one of the most fascinating molecular switchable processes,
the so-called spin crossover,13,14 we have synthesized a novel bridging triazole-based ligand, namely [1,4-phenylene(methanylylidene)]4H-1,2,4-triazol-4-amine (PM-Tria). The pure PM-Tria ligand is
prepared from terephthalaldehyde and 4-amino-1,2,4-triazole
through a straightforward Schiff-base condensation reaction,
with nearly quantitative yield (see the ESI†).
Reaction of 1 equivalent of Fe(BF4)2 with 1.33 equivalents of
PM-Tria in ethanol at 140 1C in a pressure vessel provides, after
four days, red cubic crystalline materials with {[Fe(PM-Tria)2(m2F)](BF4)}n (1) formulation in moderate yield (20%).

c

Scheme 1 The PM-Tria ligand [1,4-phenylenebis(methanylylidene)]bis(4H-1,2,4-triazol-4-amine).

The fluoride ligand is in situ generated by the decomposition
of the BF4- counterion of the starting Fe(II) salt (see the ESI,†
Section 3, for more details).
Decomposition of the tetrafluoroborate anion, through
hydrolysis or fluoride abstraction, has been long known as a
potential fluoride source in coordination compounds.15,16 After
this observation, an alternative reaction has been performed by
adding FeF2 salt in combination with Fe(BF4)2 and PM-Tria in a
1 : 1 : 2.5 molar ratio. After four days, red crystalline cubic materials
similar to those previously described are observed, but in a higher
reaction yield of 38% (see the ESI† for further details).
Single-crystal X-ray diffraction of 1 was examined at 100 K.
Compound 1 crystallizes in a tetragonal space group P4/ncc.
The asymmetric unit contains three PM-Tria ligands and three
crystallographic inequivalent iron(II) metal centers that are
compensated by three BF4- and three F- counterions. Details
of the structure solution and refinement are summarized in
Section 7 of the ESI,† and selected bond lengths and angles are
given in Tables S2–S4 and Fig. S10 and S11 (ESI†). As expected
for an iron(II) cation, the metal center shows an octahedral
environment.17,18 Fe(II) is coordinated by four nitrogen atoms
(belonging to four triazole-based PM-Tria ligands) placed at the
equatorial sites and two fluoride ions acting as coordinating
m2 ligands at the axial positions.
Every Fe(II) center is connected to four neighbouring Fe(II)
via four bridging PM-Tria ligands (Fig. 1c). Surprisingly, triazole
acts uniformly with a monodentate m1 bridging character instead
of the well-known bidentate nature of this kind of ligand with Fe(II)
metal centers.19 The three-dimensional structure is expanded
along the quasi-perfect one-dimensional {Fe(II)–F}n polymeric
unit (Fig. 1a, 2a and b). The average Fe–X = 2.1 Å bond length,
X being N and F donor atoms, is an effective indication that the
metal center is in its high-spin (HS) electronic configuration at
the given temperature (Table S2, ESI†). This observation is also
confirmed by the magnetic measurements (see below in this
communication).
The crystal packing of 1 along the c-axis shows that the
polymeric one-dimensional chains are packed parallel to the

Fig. 1 MERCURY view of 1 (hydrogen atoms have been removed for
clarity). (a) c view of a trimeric unit. (b) b view of an Fe(II)· · ·F· · ·Fe(II) trimeric
unit. (c) A segment of 1 illustrating the metallic connection along the c axis
via PM-Tria ligands.

Fig. 2 (a) The c axis view of 1 revealing the cavities contained by the
structure. (b) The representation along the b axis evidencing the collapsed
character of the structure.

c-axis with an interchain distance between iron centers of
19.94 Å. Interestingly, the PM-Tria ligands are tilted 35.711 with
relative to the Fe–F–Fe chain (Fig. 2b).
These ligands display strong supramolecular interactions
between them, evidenced by short ring to ring contacts (the
shortest being 3.337 Å, see Fig. S10, ESI†), which allows in turn
building a robust extended MOF. Moreover, it is worth mentioning
that the tetrafluoroborate counterions required for a charge
balance weakly interact with the PM-Tria ligands. At the same
time, these interactions reinforce the structural stability of the
crystals in the three dimensions (Fig. 2a, 3 and Fig. S10, ESI†).
As expected for a collapsed framework (see Fig. 2b) the
Brunauer–Emmett–Teller (BET) surface area is low, concretely,
6.7 m2 g-1. According to the structural analysis, this accounts
for a solvent-accessible volume of 640 Å3 and a molecular
surface area of 755 Å2 (see Fig. S13, ESI,† for further details).
The infrared spectrum of 1 shows the characteristic vibrational
modes of the three components, i.e. the PM-Tria ligand and the BF 4
and the F- counterions (Fig. S3–S6, ESI†). X-ray powder diffraction of
the different crystalline samples synthetized by both strategies has
been compared with the theoretically obtained single-crystal X-ray
diffraction of 1, affording a superimposed spectrum.

Fig. 3 iRASPA20 representation of 1 along the a axis. This illustration
shows how perfect the one-dimensional chains are, and the way the
counterions fill in the empty space in the network allowing the growth of
the crystalline material.

The thermogravimetric study of 1 confirms the presence of
solvent molecules from the synthesis (approximately 0.5 molecules of ethanol and 1.5 water molecules per unit cell). In
addition to this information, at around 225 1C, an irreversible
weight loss corresponding to the decomposition process of the
MOF (Fig. S7, ESI†) is observed.
The magnetic properties of 1 were determined in the
2–300 K temperature range under a constant field of 1 T and
the data are summarized in Fig. 4. The value of wMT at room
temperature is 2.5 cm3 K mol-1, slightly lower than the spinonly value expected for a magnetically diluted S = 2 sample
(of 3 cm3 K mol-1). This confirms the HS state of the Fe(II)
centers and suggests the presence of antiferromagnetic coupling
between spin carriers. Indeed, the wMT product decreases when
the temperature is decreased (Fig. 4, square dots). The wM vs. T
plot (Fig. 4, triangle dots) shows a maximum around 50 K which
is also a signature of dominant antiferromagnetic interactions.
The high temperature regime can be fitted to a Curie–Weiss law
(Fig. S15, ESI†), with parameters C = 3.85 cm3 K mol-1 (g = 2.24)
and y = -10.6 K. However, the low temperature data deviate
significantly from this model, with the appearance of a Curie tail
that indicates the presence of a paramagnetic contribution. Since
interchain distances are one order of magnitude longer than the
intrachain ones, we can consider in a first approximation that
only the fluoride bridges are effective magnetic super-exchange
pathways. Thus, we modeled the magnetic data to the analytical
expression for an infinite chain of classical spins, as derived by
Fisher from the exchange Hamiltonian (see the ESI† for equation
details).21 This model satisfactorily reproduces the data in all the
temperature range, with parameters S = 2, g = 2.12, J = -16 cm-1
and CP = 0.03 cm3 K mol-1 (see red and blue lines in Fig. 4). Such
small paramagnetic contribution corresponds approximately to
1% of the total Fe(II) content. It is probably related to the terminal
metal centres, since the actual chains (and crystals) are not
infinite as the model supposes. The antiferromagnetic nature of
the fluoride-bridged chains is also consistent with their geometry.
The Fe–F–Fe bridges of 1801 favour an orbital overlap between
semi-occupied metal orbitals carrying unpaired spins, stabilizing
a ground state with unparalleled spin alignment.16
A mosaic is defined as a design made with small pieces of
ceramic, glass, stones or other objects. These can either be

Fig. 5 On the left we see the Islamic mosaic found at the Alhambra
Palace of Granada. On the right is an illustration of 1 along its c-axis. In the
middle is a superposition of the two pictures.

organized in exact geometric shapes or created from more
haphazard and broken pieces. There has been an enormous
amount of research on mosaic tilings worldwide, and they are a
source of many teaching subjects in mathematics and art.
Probably, the Dutch M. C. Escher is one of the most recognized
artists inspired by these pieces of arts.
He was greatly surprised by Islamic art after visiting the
Alhambra Palace in Granada (Spain) in 1922 and 1936. Later, in
1958, he published a book titled ‘Regular Division of the Plane’,
in which he described the systematic build-up of mathematical
bridges. Today the Escher techniques are well known to create
tiles that, by applying geometric movements, may be used to
produce a variety of mosaics. What we have found for 1 is that
the view generated along the c axis almost perfectly fits with an
Islamic mosaic found at the Alhambra Palace (see Fig. 5 and
Fig. S15, ESI†). This genuine tessellation found for 1 mainly
arises from an unusual rotation of the central metal equatorial
coordination plane along the one-dimensional direction. In
particular, the rotation from Fe(1) to Fe(2) planes is 16.031
and from Fe(2) to Fe(3) is 9.231. The total rotation from Fe(1) to
Fe(3) is equal to 25.261 (see Fig. S11, ESI†). Remarkably, the
tetrafluoroborate counterions fill the space that in the mosaic is
occupied by an octagram.
In summary, a novel extended triazole-based ligand (PM-Tria)
has provided a very promising entry into a MOF 3D architecture with Fe(II)–F chains which exhibits a perfect linearity on its
growing-mode. The magnetic study of this material confirms an
antiferromagnetic 1D behaviour. Finally, this molecular architecture is compared with an XIV century Islamic mosaic found in the
Alhambra Palace in Granada (Spain) showing a surprising and
beautiful resemblance.
JSC is grateful to the Spanish MINECO for financial support
through the National Research Project (CTQ2016-80635-P) and
the Ramon y Cajal Research program (RYC-2014-16866) for
funding support. IMDEA Nanociencia acknowledges support
from the ‘Severo Ochoa’ Programme for Centres of Excellence
in R&D (MINECO, Grant SEV-2016-0686). This research used
resources of the Advanced Light Source, which is a DOE Office of
Fig. 4 wM vs. T (triangle dots for experimental data and the red line for a
Science User Facility under contract no. DE-AC02-05CH11231.
fitting curve) and wMT vs. T (square dots and blue line) plots.

7
JS and NM thank the MINECO of Spain (project CTQ201452045-R) and the CAM (FOTOCARBON project S2013/MIT-2841).
JRGM is grateful for the financial support of the European Union
(ERC StG grant CHEMCOMP no 279313); the Spanish MINECO 8
through project CTQ2015-71287-R and the Severo Ochoa Excellence
9
Accreditation 2014-2018 SEV-2013-0319; and the CERCA
10
Programme of the Generalitat de Catalunya.
11

Conflicts of interest
There are no conflicts to declare.

Notes and references
1 H. Furukawa, K. E. Cordova, M. O’Keeffe and O. M. Yaghi, Science,
2013, 341, 1230444.
2 MOFs find a use | Feature | Chemistry World, https://www.chemistry
world.com/feature/mofs-find-a-use/2500508.article, accessed 5 February
2018.
3 H. Deng, S. Grunder, K. E. Cordova, C. Valente, H. Furukawa,
M. Hmadeh, F. Gandara, A. C. Whalley, Z. Liu, S. Asahina,
H. Kazumori, M. O’Keeffe, O. Terasaki, J. F. Stoddart and O. M. Yaghi,
Science, 2012, 336, 1018–1023.
4 N. Zhang, B. Zhu, F. Peng, X. Yu, Y. Jia, J. Wang, L. Kong, Z. Jin,
T. Luo and J. Liu, Chem. Commun., 2014, 50, 7686.
5 S. Qiu, M. Xue and G. Zhu, Chem. Soc. Rev., 2014, 43, 6116–6140.
6 M. Zhao, K. Deng, L. He, Y. Liu, G. Li, H. Zhao and Z. Tang, J. Am.
Chem. Soc., 2014, 136, 1738–1741.

12
13
14
15
16
17
18
19
20
21

J. A. Mason, J. Oktawiec, M. K. Taylor, M. R. Hudson, J. Rodriguez,
J. E. Bachman, M. I. Gonzalez, A. Cervellino, A. Guagliardi,
C. M. Brown, P. L. Llewellyn, N. Masciocchi and J. R. Long, Nature,
2015, 527, 357–361.
E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown
and J. R. Long, Science, 2012, 335, 1606–1610.
J. Lee, O. K. Farha, J. Roberts, K. A. Scheidt, S. T. Nguyen and
J. T. Hupp, Chem. Soc. Rev., 2009, 38, 1450.
P. J. Waller, F. G´ndara and O. M. Yaghi, Acc. Chem. Res., 2015, 48,
a
3053–3063.
J. Jiang, Y. Zhao and O. M. Yaghi, J. Am. Chem. Soc., 2016, 138,
3255–3265.
B. L. Bodner, Proc. Bridg. Math. Music. Art, Archit. Cult., 2013, 225–232.
G. Moln´r, S. Rat, L. Salmon, W. Nicolazzi and A. Bousseksou, Adv.
a
Mater., 2017, 17003862.
J. S. Costa, S. Rodr´
ıguez-Jim´nez, G. A. Craig, B. Barth, C. M.
e
Beavers, S. J. Teat and G. Arom´ J. Am. Chem. Soc., 2014, 136,
ı,
3869–3874.
¨
S. Dammers, T. P. Zimmermann, S. Walleck, A. Stammler, H. Bogge,
E. Bill and T. Glaser, Inorg. Chem., 2017, 56, 1779–1782.
K. S. Pedersen, M. A. Sørensen and J. Bendix, Coord. Chem. Rev.,
2015, 299, 1–21.
G. A. Craig, J. S. Costa, O. Roubeau, S. J. Teat and G. Arom´
ı,
Chem. – Eur. J., 2012, 18, 11703–11715.
˜
Y. Garcia, V. Niel, M. C. Munoz and J. A. Real, in Spin Crossover
in Transition Metal Compounds I, ed. P. Gutlich and H. A. Goodwin,
Springer, 2004, vol. 233, pp. 229–257.
O. Roubeau, Chem. – Eur. J., 2012, 18, 15230–15244.
D. Dubbeldam, S. Calero and T. J. H. Vlugt, Mol. Simul., 2018, 7022,
1–24.
O. Kahn, Berichte der Bunsengesellschaft fur Phys. Chemie, 1994,
¨
98, 1208.