Enhancing the magnetic anisotropy of maghemite
nanoparticles via the surface coordination of molecular
complexes
Yoann Prado, Nielie Daffe„ Aude Michel, Thomas Georgelin, Nader Yaacoub,
Jean-Marc Greneche, Fadi Choueikani, Edwige Otero, Philippe Ohresser,
Marie-Anne Arrio, et al.
To cite this version:
Yoann Prado, Nielie Daffe„ Aude Michel, Thomas Georgelin, Nader Yaacoub, et al.. Enhancing the
magnetic anisotropy of maghemite nanoparticles via the surface coordination of molecular complexes.
Nature Communications, Nature Publishing Group, 2015, 6, pp.10139. 10.1038/ncomms10139. hal01390543
HAL Id: hal-01390543
https://hal.sorbonne-universite.fr/hal-01390543
Submitted on 6 Jan 2017
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
ARTICLE
Received 10 Mar 2015 | Accepted 8 Nov 2015 | Published 4 Dec 2015
DOI: 10.1038/ncomms10139
OPEN
Enhancing the magnetic anisotropy of maghemite
nanoparticles via the surface coordination of
molecular complexes
Yoann Prado1, Niéli Daffé1,2,3, Aude Michel1, Thomas Georgelin4,5, Nader Yaacoub6, Jean-Marc Grenèche6,
Fadi Choueikani3, Edwige Otero3, Philippe Ohresser3, Marie-Anne Arrio2, Christophe Cartier-dit-Moulin7,8,
Philippe Sainctavit2,3, Benoit Fleury7,8, Vincent Dupuis1, Laurent Lisnard7,8 & Jérôme Fresnais1
Superparamagnetic nanoparticles are promising objects for data storage or medical
applications. In the smallest—and more attractive—systems, the properties are governed by
the magnetic anisotropy. Here we report a molecule-based synthetic strategy to enhance this
anisotropy in sub-10-nm nanoparticles. It consists of the fabrication of composite materials
where anisotropic molecular complexes are coordinated to the surface of the nanoparticles.
Reacting 5 nm g-Fe2O3 nanoparticles with the [CoII(TPMA)Cl2] complex (TPMA: tris(2pyridylmethyl)amine) leads to the desired composite materials and the characterization of
the functionalized nanoparticles evidences the successful coordination—without nanoparticle
aggregation and without complex dissociation—of the molecular complexes to the nanoparticles surface. Magnetic measurements indicate the significant enhancement of the anisotropy in the final objects. Indeed, the functionalized nanoparticles show a threefold increase
of the blocking temperature and a coercive field increased by one order of magnitude.
1 Sorbonne Universités, UPMC Univ Paris 06, UMR 8234, PHENIX, CNRS, F-75005 Paris, France. 2 Institut de Minéralogie, de Physique des Matériaux et de
Cosmochimie, UMR 7590, CNRS, UPMC, IRD, MNHN, F-75005 Paris, France. 3 Synchrotron SOLEIL, L’Orme des Merisiers, Saint-Aubin—BP 48,
91192 Gif-sur-Yvette, France. 4 Sorbonne Universités, UPMC Univ Paris 06, UMR 7197, LRS, F-94200 Ivry-sur-Seine, France. 5 CNRS, UMR 7197, Laboratoire
de Réactivité de Surface, F-94200 Ivry-sur-Seine, France. 6 Institut des Molécules et Matériaux du Mans CNRS UMR-6283, Université du Maine,
F-72085 Le Mans, France. 7 Sorbonne Universités, UPMC Univ Paris 06, UMR 8232, IPCM, F-75005 Paris, France. 8 CNRS, UMR 8232, Institut Parisien de
Chimie Moléculaire, F-75005 Paris, France. Correspondence and requests for materials should be addressed to Y.P. (email: yoann.prado@upmc.fr) or to L.L.
(email: laurent.lisnard@upmc.fr) or to J.F. (email: jerome.fresnais@upmc.fr).
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
1
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
n single-domain superparamagnetic nanoparticles, magnetic
anisotropy has a direct impact on the magnetization, its
remanence, reversal and relaxation. Magnetic anisotropy is
therefore a key parameter in the preparation of magnetic
nanocrystals designed for high-density data storage applications
or for medical applications1–7. For such applications, achieving a
controlled modulation of the magnetic anisotropy for a given
size of crystals represents thus one of the most efficient ways
to improve and tune their magnetic properties. For example,
the optimization of the specific loss power of nanocrystals
for magnetic hyperthermia and the comprehension of the
interplay between magnetic anisotropy and magnetization is of
crucial importance for applications in nanomedicine8,9. On the
other hand, the fabrication of small magnetic nanocrystals
displaying high blocking temperature while maintaining
magnetic bistability—that is, large coercive fields—remains a
considerable challenge aimed at overtaking the so-called
superparamagnetic limit and increasing data storage densities10.
Different chemical approaches have been used so far to
alter the magnetic anisotropy of single-domain magnetic
nanoparticles: particle doping or formation of alloys11–14,
coordination of solvent or organic molecules to the
nanoparticles surface15–17, preparation of core-shell systems
incorporating highly anisotropic components8,9,12,18–23 and
insertion of the nanoparticles into magnetic host matrices10,24.
Herein, we report a strategy for the preparation of anisotropically enhanced magnetic nanoparticles. Our synthetic strategy is
based on the direct coordination of magnetic molecular
complexes to the surface of the nanoparticles. Using molecular
complexes as the enhancing unit is a straightforward approach to
improve the nanoparticles properties and it also represents a very
advantageous functionalization tool. Indeed, coordination
chemistry with its great versatility allows the design and the
use of specific complexes where both the local ion anisotropy
and the nature of the coordination sphere are controlled. The use
of coordination complexes to achieve surface functionalization
will also prevent diffusion of the ions into the particles.
Decorating magnetic nanoparticles with molecular complexes is
thus equivalent to building a new surface with ions whose
environment and hence local anisotropy is predetermined. This
represents an efficient way to not only modify but also tune the
nanoparticles surface anisotropy, which is known to contribute
greatly to the whole magnetic anisotropy and actually be one of
its most influential components in small systems25,26.
Furthermore, investigations motivated by the study of the
possible interactions between two magnetic components in
hybrid materials made from inorganic nanomaterials and
coordination compounds remain scarce27. In another type of
multi-scale hybrid system, it has been shown that the
magnetization relaxation of deposited mononuclear complexes
can be influenced by a magnetic substrate28. Our work, however,
is motivated by the opposite phenomena, that is the influence of
the complexes on the magnetic behaviour of the nanoparticles.
We demonstrate here that the grafting of an adequate magnetic
complex—even for a low quantity—on a superparamagnetic
nanoparticle leads to a massive improvement of the magnetic
properties.
I
Results
Strategy. In combining small superparamagnetic nanoparticles
and magnetic molecular complexes we wished to efficiently
transmit the anisotropy of the complex to the particles and thus
achieve the modulation of the particles magnetic anisotropy. As
the magnetic moment originating from mononuclear complexes
(few Bohr magnetons) is weak compared to that of a particle
2
(thousands of Bohr magnetons), weak interactions or electrostatic
interactions between the complexes and the particles were not
suitable for the occurrence of a significant magnetic effect.
Therefore, we have targeted the formation of a chemical bond
between the complexes and the particles as the best way to
promote a strong exchange pathway, namely, a coordination
bridge between the metal ions of the complexes and the ones
located at the surface of the particles. To successfully coordinate
complexes at the surface of nanoparticles and yield an anisotropically enhanced system, we have followed three prerequisites.
First, we have selected a magnetic complex made with a
polydentate ligand and bearing labile groups. Such a ligand
should warrant the stability and the geometry of the complex
while the labile groups will promote the coordination of the
complex to the particle surface. In this work, we have chosen
the complex [Co(TPMA)Cl2] (ref. 29; TPMA, tris(2-pyridylmethyl)amine, Fig. 1a). With its intrinsic magnetocrystalline
anisotropy, cobalt(II) was an obvious choice of metal ion30–32.
The tetradentate TPMA ligand occupies four positions in the
cobalt(II) coordination sphere leaving two cis-coordinated
chloride groups that are easily substitutable. As the geometry of
the mononuclear cobalt(II) complex allows no more than one
layer of complex at the surface of the particles, this approach
guarantees a low increase of the nanoparticle size.
Second, we have chosen magnetic nanoparticles possessing
coordinating atoms at their surface that could be synthesized
without any stabilizing organic species. The use of bare
nanoparticles is indeed necessary to promote the approach of
the complexes to the particles surface and then the coordination
reaction. The well-known Massart procedure along with a
thorough size sorting procedure allows the synthesis of
maghemite nanoparticles g-Fe2O3 with a relatively low
polydispersity and a small size33,34. The colloidal solutions
of the particles are stabilized by the pH-dependent surface
charge (positive or negative under acidic or basic conditions,
respectively) that prevents the use of any stabilizing organic
species.
Finally, it was crucial to maintain colloidal stability during and
after the coordination reaction of the complexes at the
nanoparticles surface. We took special care to ensure that no
aggregation of the particles was occurring in the solution, since it
would have led to undesirable magnetic dipolar interactions.
Indeed, such inter-particle interactions would have masked the
actual impact of the complexes on the nanoparticles magnetic
behaviour.
Synthesis and characterization. We have used small maghemite
nanoparticles in acidic colloidal solution (sample 0a, D0 ¼ 5.1 nm,
s ¼ 0.12; refs 33,34). In a first step, the [Co(TPMA)Cl2] complex
is added at room temperature to the nanoparticles acidic solution.
The number of added complexes has been varied from B3 to 210
per nanoparticle. At this stage, no iono-covalent bond between
the complexes and the particles surface is expected, albeit
supramolecular interactions cannot be ruled out. There are
furthermore no sign of increase of the hydrodynamic diameter in
dynamic light scattering (DLS). A single peak is detected and it
remains close to the value observed for the bare acidic particles 0a
(Zav ¼ 7.3 nm). In the second step, the condensation of the
complexes at the surface of the particles takes place by a brutal
modification of the pH: from 2.4 to 11 with the addition of a
concentrated solution of tetramethylammonium hydroxide
(TMAOH), to give the functionalized nanoparticles (Fig. 1c).
The very quick crossing of the zero point charge (pH 7–8) allows
the conservation of the colloidal stability. An increase of the
hydrodynamic diameter is observed after the condensation
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
Count
30
20
10
Co
N
Cl
C
0
3
4 5 6 7
Size (nm)
TMA+
+ TMAOH
NO3
0b
pH 11
Zav = 9.5 nm
D = 5.1 nm = 0.12
–
0a
pH 2.4,
Zav = 7.3 nm
+ TMAOH
+ [Co(TPMA)Cl2]
pH 2.4
Zav = 7.3 nm
1
pH 11
Zav = 10.2 nm
D = 5.0 nm
= 0.09
Figure 1 | Enhancing molecular complex and functionalized maghemite nanoparticles. (a) Representation of the [Co(TPMA)Cl2] complex used to
enhance the magnetic anisotropy of the g-Fe2O3 nanoparticles. (b) TEM image of the g-Fe2O3 nanoparticles functionalized with the cobalt(II) complex: 1
(5.0 nm, s ¼ 0.09) and (c) schematic view of the coordination of the complex with the iron ions. (d) Synthesis scheme with measured pH values,
hydrodynamic diameters (Zav), sizes (D) and distributions (s).
reaction, up to Zav ¼ 10.2 nm when B85 complexes were added
per nanoparticle in the first step (Supplementary Fig. 1 and
Supplementary Table 1). Sample 1, which corresponds to the
nanoparticles functionalized by the addition of ca. 60 complexes
per particle in the first synthesis step, shows a similar increase.
This can be ascribed to the coordination of the complexes and to
the presence around the particles of TMA þ counterions, which
accompany the modification of the nature of the surface charge
(from positively to negatively charged). Indeed, for bare nanoparticles, DLS shows a similar increase of the hydrodynamic
diameter (from 7.3 to 9.5 nm) when performing the brutal pH
change in the absence of complexes: passing from 0a (pH 2.4) to
the basified colloidal solution (sample 0b, pH 11) with the
addition of TMAOH. For the functionalized nanoparticles, the
absence of any additional peaks in DLS indicates that there are
neither aggregation of particles nor side nucleation of cobalt
oxide—that could have occurred were the complexes unstable. No
evolution of the single peak has been observed over weeks.
The addition of 485 complexes per particle induces a
dramatic increase of the hydrodynamic diameter, followed by the
flocculation of the particles. The latter is probably caused by
the loss of the electrostatic repulsion-induced stabilization
that should accompany the increase of the grafting rate. In the
following we will focus on 0b and 1.
Transmission electron microscopy (TEM) indicates that very
similar sizes and distributions are observed for 0b and 1 (5.1 nm,
s ¼ 0.12 and 5.0 nm, s ¼ 0.09, respectively; Fig. 1b and
Supplementary Fig. 2). Along with the DLS experiments, this
supports the absence of aggregation or of higher size particles.
It also indicates that the functionalization has a negligible effect
on the size of the objects. X-ray powder pattern analysis shows
that 0b and 1 both display the cubic structure of the maghemite
(Fd-3m) while the estimated crystallite sizes are in agreement
with TEM imaging (Supplementary Fig. 3 and Supplementary
Table 2). High-resolution TEM also confirms the cubic structure
for the particles and indicates that no structural evolution has
occurred during the functionalization reaction (Supplementary
Fig. 4 and Supplementary Table 3). In addition, X-ray photoelectron spectroscopy (XPS) measurement at the Fe 2p edges
shows an energy gap between 2p1/2 and 2p3/2 (13.7 eV), in
agreement with the g-Fe2O3 structure35 (Supplementary Fig. 5).
XPS measurements at the N 1s edge show two peaks at 404 and
399 eV for 1 (Fig. 2). The spectra of the bare nanoparticles 0b and
of the complex display only one peak at 403 and 398 eV,
respectively. As the presence of nitrogen atoms can originate
from the TMA þ counterions in 0b and in 1, and from the TPMA
ligand in 1 and in [Co(TPMA)Cl2], the low energy contribution
at 399 eV can be assigned to the nitrogen atoms from the TPMA
ligand and the high energy peak at 404 eV to the contribution
from the TMA þ counterion. The experimental Fe/Nlig atomic
ratio of the peaks has been found equal to 1.00 for sample 1,
which differs from the calculated one (10 accounting for the
TPMA ligands only). Nevertheless, the experimental Fe/Nlig ratio
agrees well with the calculated one if only surface iron ions
(B10%) are taken into account. Atomic absorption spectroscopy
(AAS) measurements made on a precipitated sample of 1 confirm
the presence of cobalt(II) ions. The found 46±4 Fe/Co ratio
corresponds to 52 complexes per particle (considering 2418
Fe(III) ions for a spherical 5 nm g-Fe2O3 nanoparticle). This
would indicate an 86% grafting rate corresponding to a surface
density of 0.66 complex per nm2.
The presence of complexes coordinated to the nanoparticles
surface has been further evidenced by X-ray absorption spectroscopy (XAS) measurements at the L2,3 edges of the iron and
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
3
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
35
30
FC (0b)
M (emu.g–1)
25
ZFC (0b)
FC (1)
ZFC (1)
20
15
10
5
0
10
20
30
40
50
60
70
80
T (K)
80
M (emu.g–1)
40
0
(0b)
(1)
–40
408
406
404
402
400
398
396
394
Binding energy (eV)
–80
Figure 2 | Presence of the enhancing unit in the functionalized
maghemite nanoparticles. XPS spectra at the N1s edge of samples 0b
(a), 1 (b) and of the [Co(TPMA)Cl2] complex (c).
cobalt ions for sample 1 and at the cobalt L2,3 edges for the
[Co(TPMA)Cl2] complex. For 1, the spectra at the cobalt edges
confirm the presence of octahedral Co(II) and the absence of
Co(III) (Supplementary Fig. 6). Moreover, differences are
observed between the spectra of 1 and of the ‘ungrafted’ complex
[Co(TPMA)Cl2]. They can be attributed to a change in the first
coordination sphere of the cobalt ion, since we expect the
replacement of chloride ions by oxo ligands through the
condensation at the nanoparticle surface. Indeed chloride ligands
are expected to induce a weaker ligand field than the oxo
groups from the particle surface36. This is confirmed by ligand
field multiplet calculations of Co L2,3 edges that indicate a
ligand field in 1 stronger than in the [Co(TPMA)Cl2] complex
(Supplementary Fig. 7 and Supplementary Methods).
In summary, the combination of XPS, AAS and XAS
measurements clearly confirms the presence of the
{CoII(TPMA)}2 þ complex at the surface of the nanoparticles
and the formation of an oxo-bridge between Co(II) and Fe(III)
ions.
Magnetic characterization. To assess the influence of the
molecular complex and investigate the magnetic properties of
the functionalized nanoparticles we have performed d.c.
magnetization measurements, as well as Mössbauer and X-ray
magnetic circular dichroism (XMCD) spectroscopies. Where
the former gives the macroscopic behaviour of the functionalized
nanoparticles, the latter two—as local probes—give
element-specific information.
In 1, the presence of the {CoII(TPMA)}2 þ complexes at the
nanoparticles surface increases considerably the temperature of
the maximum in the zero-field-cooled (ZFC) magnetization
curve, reaching 30 K (11 K for 0b, Fig. 3). The fit of the ZFC
4
–4,000 –2,000
0
2,000
4,000
Field (Oe)
Figure 3 | Enhanced anisotropy and improved magnetic properties.
(a) Field-cooled and zero-field-cooled (FC/ZFC) magnetization curves
measured in the 5–80 K temperature range under an applied field of 50 Oe
and (b) magnetization vs field curves measured at 5 K for 0b and 1 in
diluted solutions (%vo0.15). Lines in the ZFC plots represent the best fit
(see Methods for calculation details).
curves gives—using the same size distribution function—effective
anisotropy constants of 26 and 65 kJ m 3 for 0b and 1,
respectively, attesting thus the anisotropy enhancement (Fig. 3a,
see methods for calculation details). This enhancement is also
confirmed with the magnetization vs field curves. No break in the
hysteresis curve around the remnant magnetization is observed,
in agreement with a uniform reversal of the magnetization
(Fig. 3b and Supplementary Fig. 8). The presence of the
complexes impressively increases the coercive field of the
nanoparticles, multiplying the value by 13 (from 62 Oe for 0b
to 839 Oe for 1). In an attempt to differentiate the effect of a
surface modification due to the coordination of the complexes
from that of a magnetic coupling between the Co(II) complexes
and the nanoparticle, a Zn(II) analogue of 1 has been prepared
and measured (2). The same quantity of the diamagnetic
{ZnII(TPMA)}2 þ fragment grafted on the particle surface does
not induce a comparable effect on the temperature of the
maximum in the zero-field-cooled magnetization curve (from 11
to 14 K; Supplementary Fig. 9). In the magnetization vs field
curve, the presence of the {ZnII(TPMA)}2 þ complex has a slight
effect on the remnant magnetization but an almost negligible one
on the coercive field (from 62 to 73 Oe; Supplementary Fig. 10).
These results indicate that the effect of the {CoII(TPMA)}2 þ
units on the magnetic properties does not originate from a simple
modification of the environment of the iron ions located at the
nanoparticles surface. Moreover, since no aggregation of the
nano-objects occurs after the condensation of the complexes, the
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
observed effect necessarily results from the magnetic interaction
of the complexes with the particles, leading to an increase of the
magnetic anisotropy. The transmission of the anisotropy from the
complexes to the particles is possible only if there is an exchange
interaction between the Co(II) and the Fe(III) ions. As the
observed enhancement of the magnetic properties is important
and effective at relatively high temperature, electrostatic interactions must be ruled out. Only the occurrence of a chemical bond
such as an oxo-bridge between the Co(II) and the Fe(III) ions can
support the effective anisotropy enhancement, source of the
improved properties.
57Fe Mössbauer spectrometry has been performed at 77 K on
frozen solutions of 0b and 1 (Fig. 4 and Supplementary Fig. 11) to
discriminate the chemical environment and magnetic properties
of the different Fe species, through the analysis of the hyperfine
interactions37. Indeed, this local probe technique remains a
powerful tool for investigating Fe-containing nanoparticles and
the influence of the functionalization, thanks to its high sensitivity
to electron transfer38. The 77 K spectra result from a minor
central quadrupolar doublet and a prevailing broadened lines
magnetic sextet: they have exactly the same isomer shift and their
proportions are rather independent of the samples. These two
–5
–10
0
5
10
0
5
10
1.000
Relative transmission
0.996
(0b)
0.992
(1)
1.000
0.998
0.996
0.994
–10
–5
V (mm.s–1)
P (Bhf)
12
(0b)
(1)
8
4
contributions are unambiguously assigned to Fe species with fast
and weak superparamagnetic relaxation phenomena, due to size
distributions in the samples. The lack of resolution does not allow
the proportions of iron in tetrahedral and octahedral sites to be
estimated but they were accurately estimated from in-8 T field
Mössbauer spectra at 12 K (FeOh(III)/FeTd(III) ¼ 1.70 close to 5/3
as expected for maghemite; Supplementary Fig. 11). The mean
values of isomer shift (at 77 K 0.41(2) mm s 1), which probes the
electronic density at the 57Fe nuclei, that is the valence state, are
consistent with the presence of pure ferric species for both 0b and
1. This excludes the presence of a ferric impurity and the
occurrence of Fe2 þ species or intermediate valence state. It
further evidences that no electron transfer is induced by the
presence of the Co(II) complexes. The mean hyperfine field
distribution profiles, which correspond to the shape of the
magnetic lines, indicate clearly that the grafting of the complexes
gives rise to both a shrinkage of the distribution and a shift
towards larger hyperfine fields, that is a significant increase of the
mean hyperfine field (28.4(5) and 35.1(5) T, respectively). These
features distinctly attest a slowdown of the relaxation phenomena
of the magnetization in 1 because the attached Co(II) complexes
increase the magnetic anisotropy of the Fe(III) moments,
strengthening thus the magnetization of each nanoparticle, in
agreement with the ZFC measurements.
The shape and intensity of the XMCD signals at the Fe L2,3
edges for 1 are similar to those observed for previously reported
maghemite nanoparticles39. It bears the signature of
antiferromagnetic coupling between Fe(III) ions in tetrahedral
sites and Fe(III) ions in octahedral sites (Fig. 5). The magnetic
moment for Fe(III) ions in the sub-network of the octahedral
Fe(III) is parallel to the external magnetic field. The ratio between
the occupation of the tetrahedral and octahedral sites can be
determined from the ligand field multiplet analysis of the XMCD
shape and a FeOh(III)/FeTd(III) ratio close to 5/3 is found, as
expected for maghemite. Traces of Fe(II) have also been detected.
The latter are due to sample preparation (see methods). The
XMCD at Co L2,3 edges in 1 is mainly negative at the L3 edge
indicating that the Co(II) magnetic moment is, at 6 T, parallel to
the octahedral Fe(III) ions and antiparallel to the tetrahedral
Fe(III) ions. Element-specific magnetization curves for Fe and Co
were also obtained measuring the dependence of the XMCD
signal as a function of the applied magnetic field amplitude (see
methods). The Co-specific magnetization curve (Fig. 6 and
Supplementary Fig. 12) does not show any inversion in the sign of
the XMCD when varying the magnetic field, indicating that no
inversion of coupling can be expected at low magnetic field. All
three curves are superimposed demonstrating that the Co(II) is
magnetically coupled to the Fe(III) ions of the maghemite
nanoparticle. Moreover, the Co-specific magnetization curve of 1
differs drastically from the XMCD-detected magnetization curve
of the [Co(TPMA)Cl2] complex. The latter shows a slow increase
of the magnetization with no saturation reached at 6.5 T, as
expected for a non-interacting paramagnetic Co(II) ion. For 1, the
magnetization increases abruptly and saturates above 2 T. This
behaviour evidences and confirms that the Co(II) ions within the
grafted complexes are magnetically coupled to the iron(III) ions
at the nanoparticles surface.
0
0
10
20
30
40
50
Bhf (T)
Figure 4 | Slowdown of the relaxation of the magnetization. Zero field
57Fe Mössbauer spectra (circles: experimental; lines: calculated) measured
at 77 K for 0b (a) and 1 (b) and corresponding hyperfine field distributions
(P(Bhf)) vs hyperfine field (Bhf) plot (c).
Discussion
We have presented in this work a synthetic strategy, which, in
combining molecular and nano chemistry, offers a way towards
control and modulation of the magnetic anisotropy in nanoparticles. Magnetic measurements, Mössbauer spectrometry
and XMCD measurements show that {CoII(TPMA)}2 þ complexes grafted on the surface of maghemite nanoparticles
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
5
ARTICLE
0.04
0.03
0.2
0.1
0.02
0.01
0.0
0.00
700
xmcd signal (a.u.)
Fe L2,3
Co L2,3
710
720
730
740
770
Energy (eV)
0.1
780
790
Isotropic absorption (a.u.)
0.3
800
0.02
Fe(III) Td
0.00
0.0
–0.1
–0.02
Fe(III) Oh
xmcd signal (a.u.)
Isotropic absorption (a.u.)
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
Figure 5 | Element-specific characterization of the functionalized nanoparticles. XAS and XMCD signals measured on sample 1 at the Fe (a,c) and Co
(b,d) L2,3 edges at 5 K and 6 T.
Methods
Preparation of 0a. The solution was prepared according to literature
procedures33,34 (5.1 nm, s ¼ 0.12, [Fe] ¼ 0.87 M, %m ¼ 6.96%, %v ¼ 1.39%,
pH ¼ 1.8).
xmcd signal norm. (a.u.)
1.0
0.8
Preparation of 0b. A measure of 250 ml of 0a were diluted 10 times with a
H2O:MeOH 50% v/v mixture. Then, 500 ml of an aqueous TMAOH solution
(2.8 M) were brutally added to the solution under strong stirring leading to the
sample 0b.
0.6
0.4
(1) Fe (III) Oh L3
(1) Fe (III) Td L3
(1) Co (II) L3
(Co(TPMA)Cl2) Co(II) L3
0.2
0.0
0
1
2
3
4
5
6
7
Field (T)
Figure 6 | Fe-specific and Co-specific XMCD-detected magnetization
curves at 5 K. The XMCD curves for octahedral Fe(III) and for Co(II) were
multiplied by 1 before normalization. All the curves were normalized to
one at the highest field value, error bars are s.d.
massively enhance the magnetic properties of the nano-objects.
Our results also indicate that the strong influence of the
molecular component on the nanoparticle comes from
the covalent linking of the two species through oxo-bridges and
the resulting magnetic interaction.
This work may open tremendous prospects in the design of
nanomagnets and of multifunctional nano-platforms. Provided
that the choice of nanoparticle to functionalize allows the
formation of a coordination bridge able to promote magnetic
exchange, and that the particle size and the characteristics of
the molecule are adequately matched, it should be possible to
obtain composite nano-objects with desired blocking temperature
and coercive field. Objects prepared in soft conditions, in air, in
aqueous media, and in the lack of surfactant to stabilize
the colloidal solution represents an important advantage for
the preparation of applied materials (surface deposition,
biocompatible polymers). The possibility of performing
chemistry on the ligand born by the complex also represents an
asset and offers many additional possibilities. The adequate
choice of ligand could easily allow pre- or post- functionalization,
whether to add a property (organic chromophores) or to
structure the objects (polymerizable/‘clickable’ units) towards
polyfunctional devices.
6
Preparation of 1. A measure of 250 ml of 0a were diluted 10 times with a
H2O:MeOH 50% v/v mixture. A volume of 0.550 ml of a [Co(TPMA)Cl2] solution
(10 mM, H2O:MeOH 50% v/v) were added dropwise under stirring, followed by the
rapid addition of 500 ml of an aqueous TMAOH solution (2.8 M) under strong
stirring. Then, the solution was stirred for 4 h at 60 °C and for 24 h at room
temperature leading to sample 1.
Preparation of 2. The sample was prepared following the procedure described for
1 using [Zn(TPMA)Cl2] instead of [Co(TPMA)Cl2].
Precipitation of the particles. The addition of three volumes of acetone into the
solutions led to the precipitation of the particles. The suspension was placed on a
NdFeB magnet to settle the particles and the supernatant was removed. The
obtained paste-like solid was washed with an aliquot of ethanol and dried in an
oven at 40 °C for 48 h.
Atomic absorption spectroscopy. The total iron, cobalt and zinc concentration
(mol l 1) was determined by AAS with a Perkin–Elmer Analyst 100 apparatus
after degrading the precipitated particles in HCl (37%).
Transmission electron microscopy. Images have been performed on a JEOL
100CX2 microscope with 65 keV incident electrons focused on the specimen. Highresolution TEM has been achieved on a JEOL JEM 2011 microscope with an
acceleration voltage of 200 kV and a resolution of 0.18 nm.
Dynamic light scattering. The DLS measurements have been performed on a
Malvern Zetasizer nanoZS model equipped with a backscattering mode on the
solutions containing the particles using the intensity profile. The sizes given in the
article correspond to the Z average measurements.
X-ray powder diffraction. Patterns were collected on a Philips X’pert Pro
diffractometer using Co-Ka1 monochromatic radiation (l ¼ 1.78901 Å) and
equipped with a X’celerator linear detector.
Magnetic measurements. Magnetic measurements were carried out with
Quantum Design MPMS-XL and MPMS-5S magnetometers working in d.c. mode
on frozen solutions of the samples. The solution was diluted in a H2O:MeOH 50%
v/v mixture before measurements. The solution volume was 150 ml and the
weight concentration 0.4%. The solution is placed in a 0.2 ml eppendorf and
inserted in the cryostat of the superconducting quantum interference device
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
(SQUID) magnetometer and frozen directly from room temperature to 100 K in
zero magnetic field (lowering the rod takes a few seconds) before any measurement.
The temperature sweeping rate for ZFC/FC measurements was 2 K min 1.
Anisotropy constants calculations. Following Tamion et al.40 we have used their
semi-analytical model to describe the temperature dependence of the ZFC
magnetization and extract an estimate of the magnetic anisotropy energy
density Keff.
Having defined a switching-field frequency:
Keff Vmag
nðTÞ ¼ n0 exp
kB T
ð1Þ
with v0 ¼ 109 Hz the attempt frequency and a characteristic time dt(T), which
depends on the temperature sweeping rate (here 2 K min 1), the magnetic
moment measured during a ZFC protocol is given by
mZFC ðTÞ ¼ NT
Z1
0
Keff Vmag
M0 Vmag e uðTÞdtðTÞ þ
1 e uðTÞdtðTÞ PðDmag ÞdDmag
kB T
ð2Þ
where M0Vmag ¼ m0m2s H/(3KeffVmag) (equation 3) is the initial ZFC susceptibility in
the frozen low temperature state and P(Dmag) is the size distribution function taken
here as a lognormal with characteristic parameter taken from the TEM
characterization of the particles. In the equations above, m0 and kB denote the
magnetic permeability of vacuum and the Boltzmann constant, H is the magnetic
field strength, mS is the saturation magnetization of the maghemite and NT is the
number of magnetically active clusters.
Using this equation, it is quite straightforward to obtain a calculated ZFC curve
and optimize the value of Keff in order that make it best match the experimental
data.
Mössbauer spectrometry. 57Fe Mössbauer spectra were performed at 77 K
using a conventional constant acceleration transmission spectrometer with a
57Co source (Rh matrix) and a bath cryostat and at 12 K in a 8 T external field
applied parallel to the g-beam in a cryomagnetic device. The spectra were fitted by
means of the MOSFIT program and an a-Fe foil was used as the calibration
sample.
57Fe
X-ray photoemission spectroscopy analyses. At various times of adsorption,
t ¼ 30 s, 1, 3 and 30 min, the chamber was evacuated to some 10 10 torr, and the
sample was analysed by X-ray photoemission spectroscopy using an Omicron
NanoTechnology GmbH (Taunusstein, Germany) Argus hemispherical analyser
and a monochromatic AlKa X-ray source (1,486.6 eV). After recording a broad
range spectrum (pass energy 100 eV), high-resolution spectra were recorded for the
N 1s, C 1s, O 1s and Fe 2p core levels (pass energy 20 eV). High-resolution XPS
conditions have been fixed: ‘sweep’ analysis mode and an electron beam power of
280 W (14 kV and 20 mA). The spectra were fitted using the Casa XPS v.2.3.16
Software (Casa Software Ltd., UK) and applying a Gaussian/Lorentzian ratio G/L
equal to 70/30.
XAS and XMCD measurements. XAS and XMCD spectra at Fe and Co L2,3
edges were recorded on the soft X-ray beamline DEIMOS41 at synchrotron
SOLEIL (France). Circularly polarized photons delivered by an Apple II
undulator are monochromatised by a variable groove depth (VGD) grating
monochromator working in the inverse Petersen geometry. All reported spectra
were measured using total electron yield detection under a 10 10 mbar ultra-high
vacuum (UHV). The XMCD signals were recorded by both flipping the circular
polarization (either left or right helicity) and the applied magnetic field (either þ 6
or 6 T). The XMCD signal is obtained as the difference sXMCD ¼ s s þ
where s ¼ [sL(H ) þ sR(H þ )]/2, s þ ¼ [sL(H þ ) þ sR(H )]/2, sL (sR) is the
cross-section with left (right) polarized X-rays, and H þ (H ) the magnetic field
parallel (antiparallel) to the X-ray propagation vector. This procedure ensured a
high signal-to-noise ratio and allowed us to discard any spurious systematic signals.
XAS and XMCD spectra were measured for samples cooled to 5 K and in a 6 T
applied magnetic field.
The XMCD-detected magnetization curves are the field dependence of the
dichroic signal. The XMCD amplitude is recorded at the energy of its maximum
amplitude (707.56 eV for Td Fe, 708.19 eV for Oh Fe and 778.19 eV for Co) by
quickly switching the circular polarization thanks to the electromagnet/permanent
magnet helical undulator (EMPHU)41 available on DEIMOS beamline. Due to the
presence of TMA þ cations, drop-casts of the nanoparticles solution on goldcoated silicon plate yielded highly hydroscopic deposits. The solid samples were
thus prepared by precipitation in acetone. The solid was suspended in ethanol,
drop-casted on a slide, dried on a hot plate and fixed on carbon conductive tape to
the copper sample holder. Traces of Fe(II) have been detected on the XAS spectrum
of 1 and estimated at B2% of the iron signal. This cannot be the result of an
electron transfer from the Co(II) to the Fe(III) (Supplementary methods).
References
1. Magnetic recording media. Fuji Electric Review 57, 30–62 (2011).
2. Rosensweig, R. E. Heating magnetic fluid with alternating magnetic field.
J. Magn. Magn. Mater. 252, 370–374 (2002).
3. Jun, Y., Seo, J. & Cheon, J. Nanoscaling laws of magnetic nanoparticles and
their applicabilities in biomedical sciences. Acc. Chem. Res. 41, 179–189 (2008).
4. Frey, N. A., Peng, S., Cheng, K. & Sun, S. Magnetic nanoparticles: synthesis,
functionalization, and applications in bioimaging and magnetic energy storage.
Chem. Soc. Rev. 38, 2532–2542 (2009).
5. Fortin, J.-P. et al. Size-sorted anionic iron oxide nanomagnets as colloidal
mediators for magnetic hyperthermia. J. Am. Chem. Soc. 129, 2628–2635
(2007).
6. Mehdaoui, B. et al. Optimal size of nanoparticles for magnetic hyperthermia:
a combined theoretical and experimental study. Adv. Funct. Mater. 21,
4573–4581 (2011).
7. Georgelin, T., Bombard, S., Siaugue, J.-M. & Cabuil, V. Nanoparticle-mediated
delivery of bleomycin. Angew. Chem. Int. Ed. 49, 8897–8901 (2010).
8. Noh, S. et al. Nanoscale magnetism control via surface and exchange
anisotropy for optimized ferrimagnetic hysteresis. Nano Lett. 12, 3716–3721
(2012).
9. Lee, J.-H. et al. Exchange-coupled magnetic nanoparticles for efficient heat
induction. Nat. Nano 6, 418–422 (2011).
10. Skumryev, V. et al. Beating the superparamagnetic limit with exchange bias.
Nature 423, 850–853 (2003).
11. Salazar-Alvarez, G. et al. Reversible post-synthesis tuning of the
superparamagnetic blocking temperature of g-Fe2O3 nanoparticles by
adsorption and desorption of Co(II) ions. J. Mater. Chem. 17, 322–328 (2007).
12. Prado, Y. et al. Tuning the magnetic anisotropy in coordination nanoparticles.
Random distribution versus core-shell architecture. Chem. Commun. 48,
11455–11457 (2012).
13. Fantechi, E. et al. Exploring the effect of Co doping in fine maghemite
nanoparticles. J. Phys. Chem. C 116, 8261–8270 (2012).
14. Vichery, C. et al. Introduction of cobalt Ions in g-Fe2O3 nanoparticles by direct
coprecipitation or postsynthesis adsorption: dopant localization and magnetic
anisotropy. J. Phys. Chem. C 117, 19672–19683 (2013).
15. Vestal, C. R. & Zhang, Z. J. Effects of surface coordination chemistry on the
magnetic properties of MnFe2O4 spinel ferrite nanoparticles. J. Am. Chem. Soc.
125, 9828–9833 (2003).
16. Salafranca, J. et al. Surfactant organic molecules restore magnetism in
metal-oxide nanoparticle surfaces. Nano Lett. 12, 2499–2503 (2012).
17. Prado, Y. et al. Magnetization reversal in CsNiIICrIII(CN)6 coordination
nanoparticles: unravelling surface anisotropy and dipolar interaction effects.
Adv. Funct. Mater. 24, 5402–5411 (2014).
18. Zeng, H., Li, J., Wang, Z. L., Liu, J. P. & Sun, S. Bimagnetic core/shell FePt/
Fe3O4 nanoparticles. Nano Lett. 4, 187–190 (2003).
19. Nogués, J. et al. Exchange bias in nanostructures. Phys. Rep. 422, 65–117
(2005).
20. Catala, L. et al. Core-multishell magnetic coordination nanoparticles:
toward multifunctionality on the nanoscale. Angew. Chem. Int. Ed. 48, 183–187
(2009).
21. Salazar-Alvarez, G. et al. Two-, three-, and four-component magnetic
multilayer onion nanoparticles based on iron oxides and manganese oxides.
J. Am. Chem. Soc. 133, 16738–16741 (2011).
22. Dia, N. et al. Synergy in photomagnetic/ferromagnetic Sub-50 nm coremultishell nanoparticles. Inorg. Chem. 52, 10264–10274 (2013).
23. Estrader, M. et al. Robust antiferromagnetic coupling in hard-soft bi-magnetic
core/shell nanoparticles. Nat. Commun. 4, 2960 (2013).
24. Zeng, H., Li, J., Liu, J. P., Wang, Z. L. & Sun, S. Exchange-coupled
nanocomposite magnets by nanoparticle self-assembly. Nature 420, 395–398
(2002).
25. Poulopoulos, P. & Baberschke, K. Magnetism in thin films. J. Phys. Condens.
Matter 11, 9495 (1999).
26. Surface effects in magnetic nanoparticles (ed. Fiorani, D.) 1–298
(Springer US, 2005).
27. Zoppellaro, G., Tuček, J., Herchel, R., Šafářová, K. & Zbořil, R. Fe3O4
nanocrystals tune the magnetic regime of the Fe/Ni molecular magnet: a new
class of magnetic superstructures. Inorg. Chem. 52, 8144–8150 (2013).
28. Lodi Rizzini, A. et al. Coupling single molecule magnets to ferromagnetic
substrates. Phys. Rev. Lett. 107, 177205 (2011).
29. Davies, C. J., Solan, G. A. & Fawcett, J. Synthesis and structural characterisation
of cobalt(II) and iron(II) chloride complexes containing bis(2pyridylmethyl)amine and tris(2-pyridylmethyl)amine ligands. Polyhedron 23,
3105–3114 (2004).
30. Batchelor, L. J. et al. Pentanuclear cyanide-bridged complexes based on highly
anisotropic coii seven-coordinate building blocks: synthesis, structure, and
magnetic behavior. Inorg. Chem. 50, 12045–12052 (2011).
31. Mondal, A. et al. A cyanide and hydroxo-bridged nanocage: a new generation
of coordination clusters. Chem. Commun. 49, 1181–1183 (2013).
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications
7
ARTICLE
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms10139
32. Ruamps, R. et al. Ising-type magnetic anisotropy and single molecule magnet
behaviour in mononuclear trigonal bipyramidal Co(II) complexes. Chem. Sci. 5,
3418–3424 (2014).
33. Massart, R. Preparation of aqueous ferrofluids without using surfactant;
behavior as a function of pH and counterions. C. R. Seances Acad. Sci. Ser. C
291, 1–3 (1980).
34. Lefebure, S., Dubois, E., Cabuil, V., Neveu, S. & Massart, R. Monodisperse
magnetic nanoparticles: preparation and dispersion in water and oils. J. Mater.
Res. 13, 2975–2981 (1998).
35. Grosvenor, A. P., Kobe, B. A., Biesinger, M. C. & McIntyre, N. S. Investigation
of multiplet splitting of Fe 2p XPS spectra and bonding in iron compounds.
Surf. Interface Anal. 36, 1564–1574 (2004).
36. Lever, A. B. P. Inorganic electronic spectroscopy (Elsevier, 1984).
37. Greneche, J.-M. in Mössbauer Spectroscopy (eds. Yoshida, Y. & Langouche, G.)
187–241 (Springer, 2013).
38. Fouineau, J. et al. Synthesis, mössbauer characterization, and ab initio modeling
of iron oxide nanoparticles of medical interest functionalized by dopamine.
J. Phys. Chem. C 117, 14295–14302 (2013).
39. Brice-Profeta, S. et al. Magnetic order in g-Fe2O3 nanoparticles: a XMCD
study. J. Magn. Magn. Mater. 288, 354–365 (2005).
40. Tamion, A., Hillenkamp, M., Tournus, F., Bonet, E. & Dupuis, V. Accurate
determination of the magnetic anisotropy in cluster-assembled nanostructures.
Appl. Phys. Lett. 95, 062503 (2009).
41. Ohresser, P. et al. DEIMOS: a beamline dedicated to dichroism measurements
in the 350–2500 eV energy range. Rev. Sci. Instrum. 85, 013106 (2014).
Acknowledgements
This work was supported by the Centre National de la Recherche Scientifique (CNRS,
France), the Ministère de l’Enseignement Supérieur et de la Recherche (MESR, France),
the LabEx MATISSE and by the LabEx MiChem part of French state funds managed by
the ANR within the Investissements d’Avenir programme under reference ANR-11IDEX-0004-02. We acknowledge SOLEIL for provision of synchrotron radiation
facilities. We thank Sandra Casale and the UPMC Chemistry department microscopy
8
service for high-resolution TEM images. Y.P. and L.L. would like to thank Laure Catala
and Talal Mallah for their constant support and encouragements.
Author contributions
J.F., L.L., V.D. and B.F. conceived and supervised the project. Y.P., J.F., L.L. and B.F.
planned and implemented the synthetic and analytical experiments. A.M. and J.F.
performed the TEM imaging. N.D. and L.L. performed the powder X-ray diffraction
measurements, N.Y. and J.-M.G. analysed the X-ray data. T.G performed and analysed
the XPS measurements. Y.P., L.L. and V.D. performed and analysed the SQUID measurements. N.Y. and J.-M.G. performed and analysed the Mössbauer spectrometry
measurements. M.-A.A., N.D., C.C.-.d.-M., L.L., P.S., F.C., E.O. and P.O. performed and
analysed the XAS and XMCD measurements. Y.P. and L.L. wrote the manuscript with
the help of all authors.
Additional information
Supplementary Information accompanies this paper at http://www.nature.com/
naturecommunications
Competing financial interests: The authors declare no competing financial interests.
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
How to cite this article: Prado, Y. et al. Enhancing the magnetic anisotropy of
maghemite nanoparticles via the surface coordination of molecular complexes.
Nat. Commun. 6:10139 doi: 10.1038/ncomms10139 (2015).
This work is licensed under a Creative Commons Attribution 4.0
International License. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise
in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material.
To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
NATURE COMMUNICATIONS | 6:10139 | DOI: 10.1038/ncomms10139 | www.nature.com/naturecommunications