S2.0 S0014305712002303 Main [617311]

Macromolecular Nanotechnology
Superparamagnetic magnetite–divinylbenzene–maleic anhydride
copolymer nanocomposites obtained by dispersion polymerization
Dan Donescua, Valentin Raditoiua,⇑, Catalin Ilie Spatarua, Raluca Somoghia, Marius Ghiureaa,
Constantin Radovicia, Radu Claudiu Fierascua, Gabriel Schinteieb, Aurel Lecab,
Victor Kuncserb
aNational Research and Development Institute for Chemistry and Petrochemistry – ICECHIM, 202 Splaiul Independentei, P.O. Box 35-174,
Bucharest 060021, Romania
bNational Institute of Materials Physics, P.O. Box MG-7, Bucharest-Magurele 077125, Romania
article info
Article history:Received 22 February 2012Received in revised form 3 July 2012Accepted 18 July 2012Available online 27 July 2012
Keywords:
NanocompositeMagnetiteDivinylbenzene–Maleic Anhydridecopolymer
Dispersion polymerization
Superparamagneticabstract
Magnetite alternating copolymers divinylbenzene–maleic anhydride (DVB–MA) compos-
ites were prepared by dispersion polymerization. Because magnetite is used as a complexwith oleic acid (Fe
3O4OLA), the final hybrids show good dispersion of inorganic nanofillers
in the polymer matrix. The obtained composites were analyzed by infrared absorption
spectrometry, diffuse reflectance in visible light, thermogravimetry, X-ray fluorescence,X-ray diffraction, dynamic light scattering, scanning electron microscopy and vibratingsample magnetometry. The obtained results indicate the successful preparation of magne-
tite nanoparticles with an average size of about 23 nm dispersed in micrometer size
copolymer spherical particles, which relative content can be controlled via the processingparameters. A relationship between the relative content of magnetite nanoparticles and the
size of the polymer particles, with direct influence on the diffuse reflectance in the visible
domain, was observed. A superparamagnetic behavior was evidenced at room temperaturewith a blocking temperature lower than as expected from the bulk anisotropy constant and
the average size of the magnetite nanoparticles. Both the unexpected low blocking temper-
ature and the observed low specific magnetizations were explained by a defected and poorcrystalline structure of the magnetite nanoparticles, giving rise to spin disorder and dimin-ished crystalline anisotropy constant.
/C2112012 Elsevier Ltd. All rights reserved.
1. Introduction
Polymer nanocomposites containing magnetic particles
as fillers have shown a particular interest in the last three
decades [1,2] . Such materials that offer many possibilities
of modeling magnetic and physicochemical properties
were obtained for different targeted applications.
The main concerns were directed towards the possibil-
ity of more advanced dispersion of magnetic filler in thepolymer matrix [1,2] . One important direction is the poly-
merization of monomers in disperses media: emulsion,
miniemulsion, suspension, dispersion [2]. Depending on
the polarity of the monomers and the dispersion medium,
nanocomposite systems containing magnetic nanoparti-
cles of core–shell structure, homogeneously dispersed in
micrometer size polymeric particles/or fixed on the surface
of the polymeric particles, can be obtained [2].
A good compatibility of the magnetic nanoparticles
(magnetite, in this report), can be obtained by organophil-
ization with oleic acid (OLA). Using such organophilizated
magnetite (Fe 3O4OLA), various nanocomposites can be pre-
pared by emulsion polymerization [3], miniemulsion
0014-3057/$ – see front matter /C2112012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.eurpolymj.2012.07.012⇑Corresponding author. Fax: +40 0318115609.
E-mail address: vraditoiu@icechim.ro (V. Raditoiu).European Polymer Journal 48 (2012) 1709–1716
Contents lists available at SciVerse ScienceDirect
European Polymer Journal
journal homepage: www.elsevi er.com/locate/europolj
MACROMOLECULAR NANOTECHNOLOGY

[4–6] , microemulsion [7]and dispersion [8,9] . As already
proven in previous reports, the liophilic complex Fe 3O4OLA
is dispersible in monomer [6]or in organic solvents [3,10–
17]. In water–oil mixtures, depending on working condi-
tions (pH), it can be distributed in the oil ( o) phase, in
the aqueous ( w) phase or at the o/w interface [18].
Based on the above mentioned information, we con-
sider that a good dispersion of magnetic filler can be ob-
tained by dispersion polymerization [8]. The system is
initially homogeneous under these conditions and the ob-
tained polymeric particles can disperse more efficiently
Fe3O4OLA magnetic nanoparticles, due to the copolymeri-
zation ability of OLA with the existing monomers [7].I n
this work, the Fe 3O4OLA complex was prepared as in
[10]. A subsequent dispersion polymerization was aimed
to obtain alternanting copolymers divinylbenzene–maleic
anhydride [19]. Heptane-methyl ethyl ketone (C7-MEK)
mixture [19] was used as a dispersion medium, and deca-
ethylene glycol oleyl ether (Brij/C21096V) (OLAEO 10) as a reac-
tive surfactant. The influence of Fe 3O4OLA concentration
on particle size of alternanting copolymers DVB–MA, ther-
mal stability and optical and magnetic properties of the
composites was analyzed. The reported copolymers are
interesting because the hydrolysis in alkaline aqueous
medium [19] can transform them into polyelectrolyte dis-
persed in water, of importance in fixing some biologically
active compounds.
2. Experimental
2.1. Materials
Divinylbenzene (Fluka) (DVB) and maleic anhydride
(Fluka) (MA), were purified by distilation, respectively by
sublimation. Heptane (Fluka), methyl ethyl ketone (Chimo-
par) were purified by distilation. Decaethylene glycol oleyl
ether (Brij/C21096V) (OLAEO 10) (Fluka), azo iso butyronitrile
(AIBN) (Amos Organica) and oleic acid (Fluka) (OLA) were
used without further purification. Also FeCl 2/C14H2O and
FeCl 3/C16H2O (Merck) were used as they were received.
2.2. Samples preparation
2.2.1. Preparation of Fe 3O4OLA
The procedure previously published in [10] has been
used for preparation of magnetite nanoparticles stabilized
with OLA. Briefly, it was proceeded as follows: 240 cm3
25 wt.% NH 4OH and 260 cm3distilled water were intro-
duced in a three necked flask equipped with mechanical
stirring (350 rpm). After purging with nitrogen while stir-
ring at room temperature, a mixture of 4 g FeCl 2/C14H2O dis-
solved in 10 cm32 N HCl and 10.8 g FeCl 3/C16H2O dissolved
in 40 cm3distilled water was added. The mixture, which
immediately turns colored was kept stirring for one hour.
After stopping the stirring ferrite precipitate can be quickly
decant magnetically. After removing, the aqueous layer
precipitate was washed twice, with 100 cm3H2O. The
black-brown mixture separated by magnetic field was
decanting from the aqueous phase and over the wet pre-
cipitate were added 250 cm3CH2Cl2containing 1.5 g OLA.After an additional 0.5 h stirring, the mixture was left to
stand. A bottom highly colored organic layer and a superior
highly transparent and colorless aqueous phase, were ob-
tained. The aqueous layer was separated and then the or-
ganic layer was transferred in a polyethylene container
after removal of water and the CH 2Cl2was evaporated at
room temperature. After drying, good quality Fe 3O4OLA
nanoparticles (similar to the ones reported by [10]), with
24.5% OLA/1 g solid mixture (according to the thermo-
gravimetry measurements) were obtained.
2.2.2. Preparation of DVB–MA/Fe 3O4OLA hybrids in dispersion
A similar methodology as previously published in [19],
but adapted to our target, has been used. The obtained
samples (together with their labeling) as well as the corre-
sponding concentrations of reactants are listed in Table 1 .
An example of synthesis is provided in the following for
the case of sample D47: In a three necked flask equipped
with a mechanical stirrer (300 rpm) and a condenser,
40 cm3MEK and 60 cm3heptane as well as 4 g DVB and
4 g MA were successively introduced. After stirring, the
mixture was homogeneous and transparent. Under nitro-
gen and stirring, there were introduced 0.8 g OLAEO 10
and 0.4 g Fe 3O4OLA. After 15 min of stirring and 30 min
of ultrasonic disaggregation, the mixture was homoge-
neously colored. The whole mixture was heated at 70 /C176C
while stirring. After reaching this temperature, 0.2 g AIBN
were firstly added and another 0.2 g after 2 h. The mixture
was cooled after 4 h of reaction. A part of the mixture was
transferred to a glass vial and sealed, and another part in
polyethylene containers for solvent evaporation. After dry-
ing in vacuum at 40 /C176C, the resulting composite was
weighted ( Table 1 ).
2.3. Methods
Conversions were estimated gravimetrically. Thermal
analysis was conducted with a TA Q5000 IR instrument un-
der nitrogen at a heating rate of 10 /C176C/min. Infrared (IR)
spectra were obtained directly on the copolymer powders
by using a Specac Golden Gate ATR device (KRS5) with a
Jasco FT-IR 6300 instrument, in the range 400–4000 cm/C01
(30 scans at a resolution 4 cm/C01).
The particle size distribution was measured by Dynamic
Light Scattering (DLS) in the mixture of solvents used in
polymerization or in 0.1 N HCl aqueous solutions. The
polymer concentration in the dispersion medium was
0.01% by weight. Measurements were made with an
instrument Nanosizer ZS (Malvern) after ultrasonic disag-
gregation for 10 min. The Scanning Electron Microscopy
(SEM) was performed with a FEI Quanta 200 instrument.
Diffuse reflectance spectra were recorded on solid samples
(as powders) with Spectralon as a reference, on a Jasco UV–
Vis–NIR V 570 spectrometer equipped with a 472 ILN inte-
grating sphere (150 mm).
X-ray diffraction (XRD) analysis was performed using
an X-ray diffractometer DRON-UM with horizontal goni-
ometer in Bragg–Brentano geometry, at room temperature
using an iron filtered Co K aradiation ( k= 1.79021 Å). X-ray
fluorescence analysis was performed on a PW4025-Mini-
Pal-Panalytical EDXRF Spectrometer. Measurements were1710 D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716MACROMOLECULAR NANOTECHNOLOGY

carried out in helium atmosphere, for 300 s., without any
filter, at 20 kV and with automatically adjustment of the
current intensity.
Magnetic measurements were performed on a cryo-
genic free High Field Measurement System (Cryogenics
Ltd., Oxford, GB) working down to 1.6 K and with a maxi-
mum applied field of 9 T. The magnetic hysteresis loops
were obtained under the vibrating sample magnetometry
option, with a well confined small amount of powder, in
order to respect the point-like dipolar approximation for
the sample.
3. Results and discussion
3.1. Particle size distribution
The effect of magnetite particles stabilized with OLA in
copolymerization of DVB–MA was analyzed by changing
the concentration of inorganic filler, as shown in Table 1 .
A first observation is that the conversion of the copoly-
merization reaction is not affected by the concentration of
Fe3O4OLA. However, the diameters of DVB–MA copolymer
particles are heavily modified by increasing the concentra-
tion of the inorganic filler as it was observed from DLS
measurements ( Fig. 1 ). If just the average size of the
copolymer particles without filler is considered, it is found
that the surfactant chosen in this work (OLAEO 10), is less
effective as the maleic monoesters, previously used in
[19]. The presence of small amounts (0.2 and 0.4 g) of
Fe3O4OLA increases the tendency of copolymer particles
to agglomerate as well as their average size. By increasing
the concentration of magnetite above the previously men-
tioned values, a pronounced decrease of the average diam-
eter of the copolymer particles was observed from DLS
measurements. The particle sizes are situated in the
micrometer range, as proven by SEM, in good agreement
with those obtained for other systems [8,9] . A very impor-
tant observation can be drawn from Fig. 1 : after dispersion
in water, sizes are getting smaller, with the same tendency
to decrease at increased concentration of Fe 3O4OLA. In
good agreement with the results previously published in
[19], the SEM images shown in Fig. 2 , highlights the
agglomerated nature of the obtained particles. It is noticing
that for the hybrid with the highest concentration of mag-
netite (D 48) the relative content of smaller particles is
maximum ( Fig. 2 c).
The explanation of these results might be related to the
behavior of the radical copolymerization in differently cho-sen dispersion media [8,9] . In the CH
2Cl2solvent, used forsynthesis of inorganic filler, particles of Fe 3O4OLA are dis-
persed as entities with an average diameter of about
23 nm, as proven by the DLS results, presented in Fig. 3 .
In solvent mixture C7/MEK used for the synthesis of
copolymers, they remain in the nanometer range, but have
smaller sizes (98% are about 22 nm in size) ( Fig. 3 ).
At the beginning of the polymerization, initiation takes
place in the homogenous reaction mixture in which are dis-
persed Fe 3O4OLA nanometric particles. OLA double bonds
could be active in copolymerization with monomers in
the homogeneous initial system [7]. In a paper published
previously [19] was demonstrated that MA formed com-
plexes 1:1 with DVB and with styrene. The same phenom-
enon takes place with OLA double bonds. For systems in
which polymerization takes place in the presence of a small
quantity of Fe 3O4OLA nanoparticles, these act as agglomer-
ation centers for the DVB–MA copolymer precipitated par-
ticles. When the content of Fe 3O4OLA was raised, the
nucleation centers of precipitated particles become, more
and more numerous. The average diameters of aggregated
particles decreased while raising the content of Fe 3O4OLA.
InFig. 2 c it is seen this agglomeration of small particles.
3.2. Thermogravimetric results
InTable 1 are presented also the results obtained by
TGA, in order to observe the effect of the magnetic fillercontent on the thermal stability of the composite.Table 1
Reaction conditions and thermal properties of the obtained hybrids (4 g DVB, 4 g MA, 0.8 g OLAEO 10,4 0c m3MEC, 60 cm3C7, (0.2 + 0.2 g) AIBN, 70 /C176C, N 2).
Sample Quantity of filler
(Fe3O4OLA)(g)Quantity
of hybridobtained (g)Weight loss (%) / Temperature ( /C176C) of maximum decomposition rate
Temperature range ( /C176C) Residue
at 700 /C176C20–100 100–200 200–300 300–400 400–500 500–600 600–700
D44B – 9.33 1.442 4.456 140.17 2.470 11.56 53.17 431.79 4.176 1.005 21.62
D46 0.2 9.27 2.013 4.678 152.28 3.003 14.91 47.69 428.76 3.921 2.546 22.06D47 0.4 9.93 1.881 7.029 133.11 3.773 15.02 45.80 428.76 3.645 1.346 21.52D45 0.8 9.72 2.124 5.905 140.42 3.677 15.93 40.09 425.86 5.715 4.418 22.15D48 1.2 10.41 1.704 5.674 144.21 3.515 12.33 43.41 428.76 3.304 5.467 655.79 24.51
Fig. 1. The diameter of DVB–MA copolymer particles versus the concen-
tration of Fe 3O4OLA in water and C 7-MEK obtained from DLS
measurements.D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716 1711
MACROMOLECULAR NANOTECHNOLOGY

InFig. 4 are given the curves of weight loss for the
Fe3O4OLA complex, DVB–MA copolymers without filler
(D44B) and copolymers with a content of 1.2 g Fe 3O4OLA
(D48). It is obviously that the composite begins to decom-
pose faster due to the presence of magnetite and also the
temperature at which the loss of weight appears (e.g. at
about 20%, according to Fig. 4 ) is decreasing with increas-
ing the concentration of the filler. In the temperature rangeat which this loss occurs, oleic acid and OLAEO
10decom-poses. Hence, the presence of Fe 3O4induces a more rapid
decomposition both of the stabilizer (OLAEO 10) and of the
complexing agent (OLA). In good agreement with previous
results published in [18], it was found that the pro-
grammed heating in the range 400–500 /C176C leads to a rapid
decay of the copolymer chains ( Fig. 4 ).
A careful analysis of DTG curves shows that the rate of
decomposition (%/ /C176C) decreases with increasing the
amount of Fe 3O4OLA, as evidenced by the inset of Fig. 4 .
These results attest the increased thermal stability of poly-
mer matrix in the presence of inorganic magnetic filler, in
good agreement with results obtained for styrene-butyl
acrylate copolymers [5].
The copolymers analyzed are crosslinked due to the
polyfunctional DVB and to the OLA double bond. By per-
forming thermal analysis in nitrogen atmosphere, at the
end of the heating cycle it was found for D44B sample a
residue at 700 /C176C of about 20% by weight. This is due to car-
bonaceous char resulted by thermal decomposition of the
copolymer chains in an inert atmosphere. In air the ob-
tained residue is lowered because of oxidation reactions
which take place during the sample heating. Thus for the
same sample, in air, the residue at 700 /C176C is only about
8% by weight.
3.3. FTIR results
FTIR spectra were recorded for composites, filler and
copolymer, which were obtained via the mentioned syn-
thesis. The characteristic bands are presented in Table 3
and revealed that the filler consist of magnetite coated
with OLA and composites are formed by deposition of
DVB–MA copolymer on the Fe 3O4OLA nanoparticles. A
characteristic absorption peak of the copolymer is ob-
served at 586 cm/C01, namely in the same spectral range
where the characteristic absorption peak related to Fe–O
bond in magnetite is located. However, it is found that
the absorption peak shifts to 579 cm/C01for composites
and get a much broader shape, probably due to overlap-
Fig. 2. SEM images of DVB–MA copolymers depending on the concen-
tration of Fe 3O4OLA: (a) D44B = 0 g filler; (b) D45 = 0.8 g filler; (c)
D48 = 1.2 g filler.
Fig. 3. DLS curves for Fe 3O4OLA dispersed initially in CH 2Cl2(a) and then
in C 7-MEK mixture (b).1712 D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716MACROMOLECULAR NANOTECHNOLOGY

ping of the bands corresponding to the components of the
hybrids, as shown in Fig. 5 . After the FTIR spectra normal-
ization (in relation to the band located at 714 cm/C01) it was
also observed an increasing in the intensity of the absorp-
tion peak located at 579 cm/C01, reported for the composite,
comparatively to the band located at 586 cm/C01, reported
for the polymer. In our opinion, these behaviors are defi-
nitely related to the magnetite content. In terms of com-
paring peak area ratios, A 579/A714,in the case of hybrids
and A 586/A714, in case of copolymers, it results a relation-
ship of (A 579/A714)>>( A 586/A714), e.g. 3.3 > > 1.2, when
comparing samples D44B and D48. In conclusion, all the
mentioned FTIR results indicates the presence of magnetite
in the analyzed samples.
3.4. VIS diffuse reflectance results
The diffuse reflectance spectra in the visible wavelength
domain are shown in Fig. 6 for all the analyzed composites.
It can be observed that reflectance values of the polymeric
component are higher than 60% while the reflectance of
magnetite containing composites decreases with increas-
ing the amount of magnetite. Quantitatively, according to
the reflection curves, the decrease in reflectance is sharper
at lower concentration of magnetite in the hybrids,
whereas at higher concentration the decrease in reflec-tance becomes almost linearly. While both the polymerparticle diameter and the reflectance of the samples are
dependent on the magnetite content, a relationship be-
tween the reflectance and the polymer particle diameter
might be assumed, but this is not so obviously from the
mathematical point of view.
In the range of concentrations studied by us it could be
established a linear relationship between reflectance value
and quantity of Fe
3O4OLA (by weight) in the composite
materials. The reflectance value (%) is given by the equa-
tion R=/C017.5 m+ 58, where Ris the reflectance value
and mis the quantity (by weight) of Fe 3O4OLA in the com-
posite materials. This equation could be used to estimatethe content of Fe
3O4OLA in this type of composite material
from reflectance values measured at k= 725 nm (the point
in spectra, in which reflectance is maximum for all the ana-
lyzed samples), in the range 0.2 6m61.2, with a standard
deviation r2= 0.9983.
3.5. X-ray diffraction spectra and X-ray fluorescence results
The XRD patterns of the as-prepared D48 sample and
Fe3O4OLA sample are shown in Fig. 7 . All the diffraction
peaks can be indexed to face-centered cubic structure of
magnetite according to ASTM card No. 11-614. The broad-
ening of XRD peaks is attributed to small structural coher-
ence lengths, which might be related to sizes of the
involved nanocrystallites or nanoparticles. The specificbaseline shift of the XRD pattern of sample D48, as com-
pared to the XRD pattern of sample Fe
3O4OLA, is due to
the presence of the amorphous DVB–MA polymeric matrix.
The peak positions of magnetite in D48 and Fe 3O4OLA sam-
ples are unchanged, but the peak intensities are lower in
D48 as compared to Fe 3O4OLA, as expected due to the
incorporation of the magnetite nanoparticles in the DVB–
MA polymeric matrix.
Interplanar distances ( dhKl) were calculated via Bragg
equation, dhKl=k/(2sin h), while for estimation of the crys-
tallite size was used Scherrer formula, DhKl=Kk/(bccosh),
where: Dis the mean crystallite size, kis the X-ray wave-
length, his the Bragg angle, bc=(b2/C0bi2)1/2is the corrected
integrated peak width ( band biare the experimental and
the instrumental line-widths at angle h) and Kis a constant
which in our case is approximately equal to unit.
Differences between values obtained for interplanar
distances and ASTM card No. 11-614 data were situated
around ±0.03% which confirms that the structure of the fil-
ler correspond to magnetite. The calculated medium size of
the crystallite using Scherrer equation was situated in the
range of 7.2–8.3 nm as it is presented in Table 2 . Determi-
nation of the mean crystal size was made with an accuracy
of ±5%. It is worth to mention that the structural coherence
length obtained by XRD, of only 8 nm, is much lower than
the average particle size of about 23 nm, as obtained by
DLS. Two explanations might be consistent with such an
unexpected difference: (i) either one deals with agglomer-
ated nanocrystallites/nanoparticles of 8 nm each, rising up
an overall nanoentity of more than 20 nm average size or
(ii) one deals with a nanoentity with just an 8 nm better
crystallized core and an 8 nm thick shell of amorphous like
structure. The first explanation seems to be more realistic,while the XRD pattern of sample Fe
3O4OLA is not consis-
Fig. 4. Weight losses in programmed heating of Fe 3O4OLA, copolymers
DVB–MA without magnetite (D44B) and with 1.2 g magnetite (D48)
(conditions as in Table 1 ) (left hand axis – TGA curves; right hand axis –
DTG curves).The inset shows the evolution of the maximum rate ofdecomposition versus the concentration of Fe
3O4OLA.
Table 2Main crystallographic characteristics of nanocomposites.
Sample 2 h hKl d(Å) DhKl(nm)
Fe3O4OLA 35.095 220 2.967 8.3
41.422 311 2.529 7.7
50.490 400 2.097 7.2
D48 35.094 220 2.967 7.8
41.422 311 2.529 8.2
50.489 400 2.097 7.3
Fe3O4ASTM card No. 11-614 35.130 220 2.966 –
41.439 311 2.530 –
50.561 400 2.096 –D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716 1713
MACROMOLECULAR NANOTECHNOLOGY

tent with a large amount of amorphous phase and physi-
cally, larger nanoparticles involve better crystallization.Regarding the X-ray fluorescence spectra, they confirm
the expected good correspondence between the intensity
of the iron line and the magnetite content of composites.
However, this relationship is not linear over the whole
concentration range, as confirmed by the data presented
in the inset of Fig. 7 (e.g. the variation has the same
slope just in an intermediary domain of magnetite con-
centrations).
3.6. Magnetization measurements
The hysteresis loops for a sample containing just
Fe3O4OLA at room temperature ( RT) as well as for samples
D46 and D48, obtained both at RTand at 2 K are presented
inFig. 8 . One may observe that none of the samples shows
any coercive field at RT, whereas at 2 K they open hystere-
sis with coercive fields of about 260 Oe (e.g. 255(5) and
265(5) Oe for samples D46 and D48, respectively). While
the average diameters of the magnetite crystallites are of
about 8 nm, according to XRD measurements, they have
to be considered as magnetic monodomains [20]. Nanoen-
tities of about 23 nm, evidenced by DLS measurements canbe formed by many magnetic monodomain nanoparticles
Fig. 5. FTIR spectra in the absorption range of the Fe–O bond, for both the
copolymer DVB–MA and the hybrid sample D48.
Fig. 6. Diffuse reflectance spectra in the visible range for all analyzed
samples.
Fig. 7. XRD patterns of: (a) Fe 3O4OLA sample and (b) D48 nanocomposite.
The inset shows the intensity of the XRF iron line versus the Fe 3O4OLA
content of composites.Table 3FTIR peaks assignments for filler, nanocomposites and copolymer.
Peak position (cm/C01) and intensity Peak assignment (s-strong; m-medium; w-weak; sh-shoulder)
Fe3O4OLA D48 D44B
2913 (s) 2927 (s) 2926 (s) CH asymmetric stretching vibration
2845 (s) 2857 (s) 2856 (s) CH symmetric stretching vibration– 1854 (s) 1853 (s) C @O asymmetric stretching vibration (anhydride)
– 1774 (s) 1774 (s) C @O symmetric stretching vibration (anhydride)
1736 (s) 1729 (sh) 1729 (sh) C @O symmetric stretching vibration (COOH)
1519 (s) 1513 (w) 1513 (w) COO
/C0asymmetric stretching vibration and C–C aromatic stretching vibrations
1437 (sh) 1446 (m) 1447 (m) C @C stretching vibrations (OLA) and C–C aromatic stretching vibrations
1401 (s) – – COO/C0symmetric stretching vibration
1368 (sh) 1363 (w) 1363 (w) CH 2twisting and wagging vibrations
1223 (s) 1218 (s) 1219 (s) C–O–C stretching vibration (acid anhydride)1072 (s) 1081 (s) 1081 (s) C–O–C stretching vibration (acid anhydride)– 918 (s) 918 (s) C–O–C stretching vibration (cyclic anhydrides)
– 844 (w) 844 (w) CH out-of-plane bending vibration and C–C aromatic deformation (1,4-disubstituted benzene)
– 803 (w) 803 (w) CH out-of-plane bending vibration and C–C aromatic deformation (1,4-disubstituted benzene)– 714 (s) 714 (s) CH
2in plane rocking vibration
531 (s) 579 (s) 586 (m) Fe–O bending vibration1714 D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716MACROMOLECULAR NANOTECHNOLOGY

in close interaction. The superparamagnetic regime for
such nanoparticles is reached when the magnetic anisot-ropy energy per particle is over-passed by thermal fluctu-
ations. The fluctuation time between two potential
minima, s, is given by the equation, s=s0exp( KV/kBT)
[20,21] , where: Kis the anisotropy constant, Vis the parti-
cle volume, Tthe temperature and s0an attempt time scale
of the order of 10/C011–10/C010s. The blocking temperature, TB,
with the meaning of the temperature where the nanopar-
ticle behaves as superparamagnetic (has all the character-
istics of a paramagnetic phase, including null coercivity), is
defined as the temperature at which sequates the charac-
teristic time of the measuring method, sm. The magnetic
experimental data in Fig. 8 , showing a finite coercive field
at low temperature and a null coercive field at RT, indicate
a blocking temperature, TB, lower than RT. On the other
hand, if K,V,smands0are assumed, the blocking temper-
ature, TB, of the nanoparticles can be also estimated from
the above relation, by taking s=smatT=TB. In the present
case, with the measuring time window of the vibrating
sample magnetometer of /C240.05 s, the volume of the nano-
entity taken /C2423 nm as evidenced by DLS and a bulk
anisotropy constant for magnetite nanoparticles of
0.12/C2105J/m3[21], a blocking temperature much higher
than RTis estimated. Hence, the only explanation for an
experimental blocking temperature lower than RT(super-
paramagnetic behavior at RT), has to be related either to
a much lower anisotropy constant of the magnetite nano-
particles as compared to the bulk value (e.g. due to a de-
fected and poor crystalline structure in the nanosized
particles), or to a much lower crystallite size (e.g. of
/C248 nm, as evidenced by XRD). On the other side, the satu-
ration magnetization at RT is about 53 emu/g for sample
containing only Fe 3O4OLA, 1.1 emu/g for sample D46 and
5.8 emu/g for sample D48. By using the weight values gi-
ven in Table 1 for filler, stabilizer and polymer, sample
D46 contains just /C242% wt of magnetite whereas sample
D48 contains /C2410% wt. So a value of 55 emu/(g of magne-
tite) corresponds to sample D46 and 57 emu/(g of magne-
tite) corresponds to sample D48. Note that almost the
same specific saturation magnetization of 55(2) emu/(g of
magnetite) is provided for the analyzed samples. This is
only 60% from the typical value of bulk magnetite, which
is about 90 emu/g, according to [22]. This unexpected
reduction has to be related again to the defected structure
of the nanosized particles, involving increased spin disor-
der at the particle surface and the possible formation of amagnetic dead layer. Finally, it is worth to mention that
the observed saturation magnetization values of the
Fe
3O4OLA containing composites are consistent with the
ones presented by other works [9,23,24] .
At low temperatures, the saturation magnetization in-
creases by about 10% with respect to the room tempera-
ture corresponding values (at 1.3 emu/g in D46 and
6.4 emu/g in D48), which is a plausible thermal effect for
such systems.
4. Conclusions
Micrometer sized hybrid particles embedding magne-
tite nanoparticles have been obtained by the copolymeri-zation of DVB–MA in dispersion, in the presence of
Fig. 8. Hysteresis loops for: (a) just Fe 3O4OLA sample at room temper-
ature, (b) D46 at room temperature, (c) D46 at 2 K, (d) D48 at roomtemperature, (e) D48 at 2 K.D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716 1715
MACROMOLECULAR NANOTECHNOLOGY

Fe3O4OLA. Morphologic and dynamic light scattering char-
acterization provides information on both the dispersion of
the inorganic filler in DVB–MA copolymer particles as well
as on the average size of the magnetite nanoentities, which
is of about 23 nm. The increase of the magnetite content,
according qualitatively to the preparation expectations,
was proven by FTIR, VIS diffuse reflectance and X-ray fluo-
rescence results. A better thermal stability of the polymer
matrix was induced by the presence of the magnetic filler.
The magnetic results prove the presence of magnetite
nanoparticles in the blocked regime at low temperature
and a superparamagnetic behavior at room temperature,consistent with an agglomeration of nanocrystallites/nano-
particles of 8 nm average size in larger magnetic entities of
about 23 nm average size. The much lower saturation mag-
netization of magnetite as compared to the bulk values
indicates a poor crystallization with a defected surface
layer, giving rise to a randomly oriented spin structure
and hence, to a magnetic dead layer at the particle surface.
Acknowledgements
Work at NIMP was performed with the financial sup-
port of the Romanian Research Program Idei/contract 75/
2011.
References
[1] Millan A, Palacio F, Snoeck E, Serin V, Lecante P. Magnetic polymer
nanocomposites. In: Mai YW, Yu ZZ, editors. Polymer
nanocomposites. Woodhead Publishing Ltd, CRC Press LLC; 2006.
[2] Braconnot S, Lansalot M, Elaissari A. Les latex magnétiques:
méthodes d ´’élaboration et proprieties. In: Daniel JC, Pichot C,
editors. Les latex synthetiques. TEC&DOC Ed; 2006.
[3] Mouaziz H, Braconnot S, Ginot F, Elaissari A. Elaboration of
hydrophilic aminodextran containing submicron magnetic latex
particles. Coll Polym Sci 2009;287:287–97.
[4] Yang S, Liu H. A novel approach to hollow superparamagnetic
magnetite/polystyrene nanocomposite microspheres via interfacialpolymerization. J Mater Chem 2006;16:4480–7.
[5] Mahdavian AR, Ashjari M, Mobarakeh HS. Nanocomposite particles
with core–shell morphology I Preparation and characterization of
Fe
3O4–poly(butyl acrylate-styrene) particles via miniemulsion
polymerization. J Appl Polym Sci 2008;110:1242–9.
[6] Lu S, Ramos J, Forcada J. Self-stabilized magnetic polymeric
composite nanoparticles by emulsifier-free miniemulsion
polymerization. Langmuir 2007;23:12893–900.[7] Liu G, Liu P. Mono-dispersed functional polymeric nanocapsules
with multi-lacuna via soapless microemulsion polymerization with
spindle-like a-Fe 2O3nanoparticles as templates. Nanoscale Res Lett
2009;4:281–5.
[8] Horák D, Lednicky ´F, Petrovsky ´E, Kapic ˇka A. Magnetic characteristics
of ferrimagnetic microspheres prepared by dispersion
polymerization. Macromol Mater Eng 2004;289:341–8.
[9] Xu JH, Zhang CC. Synthesis and characterization of imidazole type
magnetic chelating resin. Iran Polym J 2007;16:787–94.
[10] Banert T, Peuker UA. Preparation of highly filled super-paramagnetic
PMMA-magnetite nano composites using the solution method. J
Mater Sci 2006;41:3051–6.
[11] Stolarczyk JK, Ghosh S, Brougham DF. Controlled growth of
nanoparticle clusters through competitive stabilizer desorption.
Angew Chem Int Ed 2008;48:175–8.
[12] Wu Y, Zuo F, Zheng Z, Ding X, Peng Y. A novel approach to molecular
recognition surface of magnetic nanoparticles based on host–guesteffect. Nanoscale Res Lett 2009;4:738–47.
[13] Caruntu D, Caruntu G, Chen Y, O’Connor CJ, Goloverda G,
Kolesnichenko V. Synthesis of variable-sized nanocrystals of Fe
3O4
with high surface reactivity. Chem Mater 2004;16:5527–34.
[14] Xie J, Xu C, Kohler N, Hou Y, Sun S. Controlled PEGylation of
monodisperse Fe 3O4nanoparticles for reduced non-specific uptake
by macrophage cells. Adv Mater 2007;19:3163–6.
[15] Altavilla C, Ciliberto E, Gatteschi D, Sangregorio C. A new route to
fabricate monolayers of magnetite nanoparticles on silicon. Adv
Mater 2005;17:1084–7.
[16] De Palma R, Peeters S, Van Bael MJ, Van den Rul H, Bonroy K, Laureyn
W, et al. Silane ligand exchange to make hydrophobicsuperparamagnetic nanoparticles water-dispersible. Chem Mater
2007;19:1821–31.
[17] Bourlinos AB, Bakandritsos A, Georgakilas V, Tzitzios V, Petridis D.
Facile synthesis of capped
c-Fe 2O3and Fe 3O4nanoparticles. J Mater
Sci 2006;41:5250–6.
[18] Lan Q, Liu C, Yang F, Liu S, Xu J, Sun D. Synthesis of bilayer oleic acid-
coated Fe 3O4nanoparticles and their application in pH-responsive
Pickering emulsions. J Coll Int Sci 2007;310:260–9.
[19] Somoghi R, Donescu D, Ghiurea M, Radovici C, Serban S, Petcu C,
et al. Copolymerization of DVB with MA in non aqueous dispersion. J
Optoelectron Adv Mater 2008;10:1457–62.
[20] Sorensen CM. Magnetism. In: Klabunde KJ, editor. Nanoscale
materials in chemistry. John Wiley and Sons; 2001.
[21] Kuncser V, Schinteie G, Sahoo B, Keune W, Bica D, Vekas L, et al.
Magnetic interactions in water based ferrofluids studied by
Mossbauer spectroscopy. J Phys: Condens Matter
2007;19:016205–20.
[22] Landfester K, Ramirez LP. Encapsulated magnetite particles for
biomedical application. J Phys: Condens Matter 2003;15:S1345–61.
[23] Xu ZZ, Wang CC, Yang WL, Deng YH, Fu SK. Encapsulation of
nanosized magnetic iron oxide by polyacrylamide via inverse
miniemulsion polymerization. J Magn Magn Mat 2004;277:136–43.
[24] Hong RY, Feng B, Cai X, Liu G, Li HZ, Ding J, et al. Double-
Miniemulsion preparation of Fe 3O4/poly(methyl methacrylate)
magnetic latex. J Appl Polym Sci 2009;112:89–98.1716 D. Donescu et al. / European Polymer Journal 48 (2012) 1709–1716MACROMOLECULAR NANOTECHNOLOGY