Matrix-dependent cooperativity in spin crossover Fe(pyrazine)Pt(CN) 4 nanoparticles

Article abstract of DOI:10.1039/c1cc14463d

Anisotropic nanoparticles of the Fe(pyrazine)Pt(CN)₄ network were prepared embedded in various matrices that revealed to have a dramatic effect on the cooperative spin crossover phenomena. By a judicious choice of the nature of the matrix and t

Full text of DOI:10.1039/c1cc14463d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 11501–11503
www.rsc.org/chemcomm
COMMUNICATION
Matrix-dependent cooperativity in spin crossover Fe(pyrazine)Pt(CN)4
nanoparticlesw
Yousuf Raza,a Florence Volatron,a Simona Moldovan,b Ovidiu Ersen,b Vincent Huc,a
a
c
c
d
Cyril Martini,a Franc
¸ ois Brisset, Alexandre Gloter, Odile Stephan, Azzedine Bousseksou,
´
Laure Catala*a and Talal Mallah*a
Received 22nd July 2011, Accepted 9th September 2011
DOI: 10.1039/c1cc14463d
Anisotropic nanoparticles of the Fe(pyrazine)Pt(CN)4 network
were prepared embedded in various matrices that revealed to have a
dramatic effect on the cooperative spin crossover phenomena. By
a judicious choice of the nature of the matrix and the control of
interparticle distances, a hysteresis of 15 K was achieved close to
room temperature for such nano-objects.
The Fe(pz)Pt(CN)4 nanoparticles were synthesized according
to the previously reported method (see ESIw).5 Transmission
Electronic Microscopy (TEM) imaging performed on the
nanoparticles in microemulsion (after step 1, see ESIw) shows
mainly square-like particles (10 Â 10 nm) as well as some
rectangular ones that may be due to different relative orientations
of the platelet-like particles as against the observation direction
(Fig. 1, top left). To assess the exact shape and morphology of
the particles, an electron tomography study was performed in
TEM mode. A section of the reconstructed volume is plotted
(Fig. 1, bottom right and S1, ESIw), it shows anisotropic
nanoparticles, with in-plane widths L of 10 Â 10 nm and a
thickness h of 5 nm (mean L/h ratio slightly inferior to 2, see
ESIw). This provides an estimation of the average number of
atoms of 800 Fe atoms (and 800 atoms of Pt) included in a
10 Â 10 Â 5 nm3 particle, with a high proportion of 55% Fe
atoms located at the surface.
Spin crossover (SCO) systems are among the most fascinating
materials in the field of molecular magnetism because they
may show hysteresis at room temperature conferring a memory
effect to the material. The switching between the high spin and
the low spin states may be triggered by different external
stimuli such as light, temperature, pressure, magnetic field,
etc. . ..1 This opens the possibility for many useful applications
as it has already been suggested and recently demonstrated.2
One challenge in this area concerns the integration of SCO
objects into nanoscopic devices for miniaturization issues,
which implies the preparation and the investigation of the
behaviour of nanosized objects. Such objects have been prepared
very recently mainly for the triazole and the Hofmann-like
clathrate families.3–5,7 Size effect has been evidenced in one
system5 and hysteresis has been observed for sub-10 nm
nanoparticles in other two systems.4
Preliminary studies of the environment effect on the SCO
behaviour were evidenced on the particles coated with the
AOT (bis(2-ethylhexyl) sulfosuccinate) surfactant. Progressively
removing the organic matter by successive washing led to a
complete change in the magnetic behaviour (Fig. S2, ESIw).
In this communication, we demonstrate that the environment
of the nanoparticles may play a crucial role in the crossover
process and that the cooperativity may be tuned by the matrix
surrounding the nanoparticles within the nanocomposite. A
hysteresis loop with an aperture of 15 K is evidenced for
anisotropic (10 Â 10 Â 5 nm3) nanoparticles (NPs) of the
Fe(pz)Pt(CN)4 network when coated with a thin shell of SiO2,
while softer matrices destroy the cooperativity.
a
´
´
´
Institut de Chimie Moleculaire et des Materiaux d’Orsay, Universite
Paris-Sud 11, F-91405 Orsay, France. E-mail: laure.catala@u-psud.fr,
talal.mallah@u-psud.fr; Fax: +33 169154754
b IPCMS, UMR CNRS 7504, 23, rue du Loess, B.P. 43, F-67034
Strasbourg Cedex 2, France
c Laboratoire de Physique des Solides, Universite´ Paris-Sud 11, F-91405
Orsay, France
d Laboratoire de Chimie de Coordination, CNRS UPR-8241, 205,
route de Narbonne, 31077 Toulouse Cedex, France
Fig. 1 TEM image and distribution of 10 nm NPs after step 1 (top),
TEM image of 1 dispersed in chloroform (bottom left) and slices redrawn
from the reconstructed volume in the XZ direction (bottom right).
w Electronic supplementary information (ESI) available: Experimental
section, tomography study, IR data, XRPD data, additional SQUID
measurements. See DOI: 10.1039/c1cc14463d
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 11501–11503 11501

View Article Online
Scheme 1 Calix8, R = C8H17
.
However, when all the organic matter was removed, X-Ray
Powder Diffraction (XRPD) showed an increase in the mean
size of the objects, indicating a partial connection between the
nanoparticles (Fig. S3, ESIw). To avoid the coalescence and
investigate the environment effect on the well defined original
10 nm nanoparticles, two coating agents were used: (i) a
Fig. 2 STEM image of NPs of 2 after dispersion in methanol (top)
and EELS map of one particle showing the presence of Si and Fe in the
particle’s core (left and right).
calixarene based ligand bearing
8 pyridine functions
1 (calix8 = C192H264N8O16S8, Scheme 1 and ESIw for the
synthetic procedure) that may replace the peripheral pyrazine or
water molecules and thus avoid coalescence, (ii) the inorganic
polymer silica with thin 2 and thick 3 SiO2 shells.
particles are isolated and, when present in aggregates, are
separated by the thin silica shell (Fig. S8, ESIw). For sample 3,
the thicker silica shell can be observed in the TEM images as a
mean size of 18.4 Æ 3.4 nm is observed in the microemulsion
(Fig. S9, ESIw). The XRPD diagram still gives a size of 10 nm
for the core and the silica shell is estimated to be around 4.5 nm
(Fig. S7, ESIw), leading to an average separation of
9 nm between the nano-objects within the composite, while this
separation is less than 5 nm for 2. The infra-red spectrum
(Fig. S10, ESIw) and EDS analysis (Si : Fe : Pt = 60 : 1 : 1 for 3)
confirm the much larger amount of silica formed for
3 compared to 2.
For 1, TEM imaging of the microemulsion shows relatively
monodispersed objects with a size of 10.5 Æ 1.8 nm (Fig. 1, top).
After addition of calix8 (step 2), the powder is washed with
acetone and water several times in order to remove excess
surfactant as confirmed by the IR analysis (Fig. S4, ESIw). The
XRPD diagram is consistent with 10 nm objects within the
nanocomposite as expected (Fig. S5, ESIw). Elemental Electron
Dispersive Spectroscopy X-ray (EDS) analysis performed on
1 gives the following unit formula FePt(CN)4(C4H4N2)0.4
Á
Magnetic studies, performed using a SQUID susceptometer
on all three compounds, were repeated in several batches for
each one. A strict protocol was applied consisting of heating
the sample at 130 1C for 1 hour within the SQUID in order to
remove solvent molecules that may have an influence on the
magnetic behaviour. The cooling rate was 0.5 K minÀ1 for all
samples. For 3 (thick SiO2 shell) a gradual transition with the
TC close to 220 K is observed; the residual Fe(II) high spin
(HS) fraction was found to be around 50% at T = 50 K
(Fig. 3). In the case of more concentrated samples 1 and 2, a
steeper transition occurs at higher temperature (around 260 K),
and a smaller HS Fe(II) residual fraction remains (30% for
both at T = 50 K). In all cases, the decrease of the hysteresis
width and the transition temperature is observed together with
an increase of the residual HS fraction. It was always observed
(H2O)0.5Á(C192H264N8O16S8)0.07 (see Experimental section).
This leads to a mean value of 0.56 pyridine/Fe(II), which
corresponds to a full coverage of the 55% surface iron sites
by calix8 that replace pyrazine on the particles’ surface.
Nanocomposite 1 is thus made of 10 nm NPs coated by a thin
layer of calix8 that has a thickness of around 2 nm. The
average distance between the nano-objects is thus around 4 nm.
This organic shell prevents the NPs from coalescence and leads
to dispersible particles. Indeed, 1 can be dispersed again in
CHCl3; the TEM image (Fig. 1, bottom left) confirms the size
and the integrity of the nanoparticles.
The preparation of the silica-coated NPs 2 and 3 was carried
out by slightly modifying already published procedures (see ESIw).6
An average size of 14 nm (14.0 Æ 2.4 nm) was determined for
2 by TEM imaging on the NPs after the silica shell formation
prior to their isolation from the microemulsion (Fig. S6, ESIw).
After recovering, the XRPD diagram confirms the integrity of the
clathrate nanocrystals when coated by SiO2 (Fig. S7, ESIw).
The presence of the SiO2 shell was first confirmed by EDS
analysis; a Si : Fe : Pt ratio of 6 : 1 : 1 was found. A dispersion
of the NPs in MeOH allowed imaging the SiO2 coated NPs by
Scanning Transmission Electronic Microscopy in dark field
mode (HAADF-STEM) and their composition was probed by
Electron Energy Loss Spectroscopy (EELS). Fig. 2 shows a
typical profile of one of the largest particles; it reveals that Fe
is present on a width of 11–13 nm that corresponds to the
Fe(pz)Pt(CN)4 cores, surrounded by a shell of around 2 nm of
silica, leading to an overall size close to 17 nm. Most of the
Fig. 3 wT = f(T) for 1(green), 2 (black) and 3 (red).
c
11502 Chem. Commun., 2011, 47, 11501–11503
This journal is The Royal Society of Chemistry 2011

View Article Online
upon increasing the average distance between the nanoparticles
within the composite independently from the nature of the
coating agent. Indeed, diluting the NPs in the organic polymer
polyvinylpyrrolidone (PVP) leads to a magnetic behaviour
very close to that of 3 (Fig. S11, ESIw).
The main result that highlights the role of the matrix is the
presence of a large hysteresis of 15 K for 2 where the particles
are coated with a thin silica shell, while only a very small
aperture of 2 K occurs for 1 (the calix-covered particles) and
no hysteresis is observed for the diluted composites whether in
silica or in PVP. This is consistent with results reported on
particles of related 2D networks embedded in PVP.7
electron–phonon couplings within the crystal network. These
experimental results thus open new perspectives in the area of
spin crossover nanomaterials where cooperativity can be
triggered and/or tuned by changing, for instance, the mechanical
properties of the media surrounding the nano-objects.
The authors gratefully acknowledge for their financial support
from the Higher Education Commission of Pakistan, the
French Ministry of Research (ANR grant contract number
JCJC N1 ANR-08-JCJC-0085-01) and the METSA network.
We thank the Centre Commun de Microscopie Electronique
d’Orsay for the instrumentation and Dr Gabor Molnar for
fruitful discussion.
These results demonstrate that the relative separation
between the NPs plays a major role for the cooperativity.
The diluted compounds (3 and the PVP coated particles) have
weak cooperativity and the amount of the residual HS Fe(II)
fraction corresponds to the percentage of Fe(II) ions present on
the particles surface. While for the concentrated ones (1 and 2),
the cooperativity is stronger. However, the degree of cooperativity
depends not only on the average distance between the objects
but on the nature of the matrix also. When the particles are
coated with the organic ligand (calix8) that imposes a distance
between the particles of around 4 nm, only a weak hysteresis is
present while for the SiO2 coated NPs a hysteresis of 15 K
occurs despite a separation of 5 nm. The observed results lead
to the conclusion that the key-point of the magnetic behaviour
stems from a conjugated effect of the nature of the matrix and
the average distance between the nanoparticles within the
nanocomposites. The nature of the interface plays a minor
role since compounds 2 and 3 are made of nanoparticles
coated with SiO2 and have a completely different magnetic
behaviour, while 3 and the PVP nanocomposite have the same
magnetic behaviour.
Notes and references
1 P. Gutlich and A. Hauser, Coord. Chem. Rev., 1990, 97, 1–22;
J. Krober, E. Codjovi, O. Kahn, F. Groliere and C. Jay, J. Am.
Chem. Soc., 1993, 115, 9810–9811; A. Hauser, J. Jeftic,
H. Romstedt, R. Hinek and H. Spiering, Coord. Chem. Rev.,
1999, 192, 471–491; V. Niel, J. M. Martinez-Agudo,
M. C. Munoz, A. B. Gaspar and J. A. Real, Inorg. Chem., 2001,
40, 3838; A. Bousseksou, G. Molnar, L. Salmon and W. Nicolazzi,
Chem. Soc. Rev., 2011, 40, 3313–3335; S. Bonhommeau, T. Guillon,
L. M. L. Daku, P. Demont, J. S. Costa, J. F. Letard, G. Molnar and
A. Bousseksou, Angew. Chem., Int. Ed., 2006, 45, 1625–1629;
S. Bonhommeau, G. Molnar, A. Galet, A. Zwick, J. A. Real,
J. J. McGarvey and A. Bousseksou, Angew. Chem., Int. Ed., 2005,
44, 4069–4073.
2 O. Kahn and J. P. Launay, Chemtronics, 1988, 3, 140; O. Kahn and
C. J. Martinez, Science, 1998, 279, 44–48; L. Salmon, G. Molnar,
D. Zitouni, C. Quintero, C. Bergaud, J. C. Micheau and
A. Bousseksou, J. Mater. Chem., 2010, 20, 5499–5503.
3 S. Cobo, G. Molnar, J. A. Real and A. Bousseksou, Angew. Chem.,
Int. Ed., 2006, 45, 5786–5789; G. Molnar, S. Cobo, J. A. Real,
F. Carcenac, E. Daran, C. Vieu and A. Bousseksou, Adv. Mater.,
2007, 19, 2163–2167; T. Forestier, S. Mornet, N. Daro,
T. Nishihara, S. Mouri, K. Tanaka, O. Fouche, E. Freysz and
J. F. Letard, Chem. Commun., 2008, 4327–4329; T. Forestier,
A. Kaiba, S. Pechev, D. Denux, P. Guionneau, C. Etrillard,
N. Daro, E. Freysz and J. F. Letard, Chem.–Eur. J., 2009, 15,
6122–6130; J. R. Galan-Mascaros, E. Coronado, A. Forment-Aliaga,
M. Monrabal-Capilla, E. Pinilla-Cienfuegos and M. Ceolin, Inorg.
Chem., 2010, 49, 5706–5714; A. Tissot, J. F. Bardeau, E. Riviere,
F. Brisset and M. L. Boillot, Dalton Trans., 2010, 39(33), 7806–7812;
C. Faulmann, J. Chahine, I. Malfant, D. de Caro, B. Cormary and
L. Valade, Dalton Trans., 2011, 40(11), 2480; J. M. Herrera, S. T.-P.
Titos-Padilla, S. X. W. Chen, J. J. Delgado and E. Colacio, Angew.
Chem., Int. Ed., 2011, 50, 3290.
4 E. Coronado, J. R. Galan-Mascaros, M. Monrabal-Capilla,
J. Garcia-Martinez and P. Pardo-Ibanez, Adv. Mater., 2007, 19,
1359–1361; J. Larionova, L. Salmon, Y. Guari, A. Tokarev,
K. Molvinger, G. Molnar and A. Bousseksou, Angew. Chem., Int.
Ed., 2008, 47, 8236–8240.
5 F. Volatron, L. Catala, E. Riviere, A. Gloter, O. Stephan and
T. Mallah, Inorg. Chem., 2008, 47, 6584–6586; A. B. Gaspar,
I. Boldog, V. Martinez, P. Pardo-Ibanez, V. Ksenofontov,
A. Bhattacharjee, P. Gutlich and J. A. Real, Angew. Chem., Int.
Ed., 2008, 47, 6433.
These conclusions may be rationalized by considering the
ability of the matrix to transmit the elastic vibrations due to
the switching of each nanoparticle from LS to HS. The
propagation of the phonons via elastic interactions is invoked
in many models to explain the cooperativity of the SCO
phenomena, and parameterized in molecular complexes by
an empirical interaction parameter G.8 In the Spiering model,
G is related to the bulk modulus of the crystal that quantifies
the compressibility of the crystal and the ability to transmit the
vibrations.9 For the case of the embedded nanoparticles,
cooperativity is due to the transmission within the nanocomposite
of the elastic forces between the single nanoparticles through
the matrix. The less compressible the matrix (high bulk
modulus) the better the propagation of the vibrations is and
the larger the hysteresis is. For a defined matrix (samples 2 and
3), the shorter the distance the better the propagation of the
elastic interactions is.
6 F. J. Arriagada and K. Osseoasare, Adv. Chem. Ser., 1994, 234,
113–128; F. J. Arriagada and K. Osseoasare, J. Colloid Interface
Sci., 1995, 170, 8–17.
7 A. B. Gaspar, V. Martinez, I. Boldog, V. Ksenofontov,
A. Bhattacharjee, P. Gutlich and J. A. Real, Chem. Mater., 2010,
22, 4271–4281.
This communication highlights the crucial influence of the
matrix surrounding the particles that propagates elastic inter-
actions: a ‘‘compressible’’ matrix leads to a lower cooperativity of
the SCO phenomena in the present case, while more rigid media,
such as silica, restores the hysteresis loop due to a stronger
cooperativity. In a first stage, this may be understood as the
variation, through the mechanical/elastic properties of the matrix
surrounding the SCO nanoparticles, of the long range elastic
interactions which, as we know, are mainly attributed to the
8 C. P. Slitcher and H. G. Drickamer, J. Chem. Phys., 1972, 56,
2142–2148.
9 V. Ksenofontov, H. Spiering, A. Schreiner, G. Levchenko,
H. A. Goodwin and P. Gutlich, J. Phys. Chem. Solids, 1999, 60,
393–399; H. Spiering, T. Kohlhaas, N. Romstedt, A. Hauser,
C. Bruns-Yilmaz, J. Kusz and P. Gutlich, Coord. Chem. Rev.,
1999, 192, 629–647.
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 11501–11503 11503