Direct conversion of iron stearate into magnetic Fe and Fe3C
nanocrystals encapsulated in polyhedral graphite cages
Junfeng Geng, David A. Jefferson and Brian F. G. Johnson*
Department of Chemistry, Cambridge University, Lensfield Road, Cambridge, UK CB2 1EW.
E-mail: bfgj1@cam.ac.uk; Fax: 144 (0)1223 336362; Tel: 144 (0)1223 336337
Received (in Cambridge, UK) 27th April 2004, Accepted 12th July 2004
First published as an Advance Article on the web 21st September 2004
We report a direct salt-conversion approach for large-scale
synthesis of carbon-encapsulated magnetic Fe and Fe3C
nanoparticles.
Carbon-encapsulated magnetic metal nanoparticles, such as Fe, Ni
and Co, have a core–shell structure.1 The excellent stability towards
environmental degradation makes these small particles ideal
candidates in many applications such as magnetic data storage,
magnetic toners in xerography and magnetic inks or ferrofluids,2 as
well as in bio-engineering applications such as drug delivery and
magnetic resonance imaging.3 These materials can be synthesised
by the electrical arc-discharge technique1–4 and pyrolysis of non-
graphitising carbon materials,5 however, the yield is low, which
makes large-scale synthesis impractical. Recently, catalytically
assisted chemical vapour deposition (CCVD) has been employed,6
but some obstacles still limit the method for large-scale synthesis.
These include problems caused either by the relatively low
productivity, or the poor product quality, or difficulty in separating
the product from the catalyst supporting material. In this paper we
report a new method for bulk synthesis of the carbon-encapsulated
magnetic iron nanoparticles. We find that the salt of iron stearate
can be directly converted into Fe and Fe3C nanocrystals
encapsulated in polyhedral carbon cages on pyrolysis under
appropriate conditions, and the direct salt-conversion phenomenon
can be used as a single-step approach to make the material in
large-scale.
Fig. 1 SEM image of the as-converted product shows the nanoparticles and
their size distribution. The product was prepared by placing ca. 3.0 g iron
stearate powder (Strem Chemicals Ltd., dark brown pills, needs to be
ground before use) in a quartz reactor that was then placed at the centre of
a tube-shaped furnace. A flow of argon (1.5–2.0 l min21) was employed to
get rid of gases produced during the thermal decomposition. The reaction
temperature was 900 uC, reaction time was 10 min.
contrast to the arc-discharge technique where the naked metal
particles are the main side-product. Thus, in regard to the
stoichiometric conversion of Fe, our yield approximates to 100%.
X-ray diffraction tests on the as-prepared products confirm that
the encapsulated species are either single-phase metallic Fe or Fe3C
single crystals (Fig. 3).{ The narrow peak at 2h ~ 26.3u results from
The chemical process of the salt conversion may be understood
from the following stoichiometric equation:
2
n(CH3(CH2)16COO)2 Fe21 A nFe 1 (36n)C 1
(4n)O 1 (70n)H
˚
well-crystallised graphite (002) fringes with a d-spacing of 3.39 A.
A laser Raman test shows the normal D and G peaks of graphitic
structureat1353cm21and1582cm21, respectively, withtheintensity
of G w D. At present it is still not possible to determine the ratio
between Fe and Fe3C.7 However, it is reasonable to believe, based on
our electron microscopic observations and the X-ray result, that
neither Fe nor Fe3C dominates in the product. To investigate the
magnetic properties, the magnetization of the encapsulated particles
versus the applied field at room temperature (300 K) was measured
with a maximum applied field of 50 kOe. The observed M–H loop
(Fig. 4) indicates the intrinsic magnetic properties of the nano-
particles. The saturation magnetization (Ms), remanent magnetiza-
Formation of the metal nanoparticles by atomic aggregation was
followed by a series of processes including segregation, diffusion
and precipitation of the carbon on the particle surfaces, as well as
nucleation and growth of the graphite cages through continuous
carbon deposition. Some of the carbon may be lost as CO and
CO2. The H and O will be eventually released as H2O. The
conversion experiment was carried out at 900 uC under an argon
atmosphere. To successfully realise the conversion process, we find
that it is necessary to restrict the movement of the atomic carbon
generated in the gas phase and thereby to let them be substantially
maintained within the reactor. A comparative experiment shows
that without this confinement, no carbon-encapsulated nano-
particles are produced under the same conditions.
Examination of the product by scanning electron microscopy
(SEM) and transmission electron microscopy (TEM) indicates that
the conversion has produced metal nanoparticles encapsulated in
polyhedral graphite cages (Fig. 1 and 2). The nanoparticles are
typically 20 to 200 nm in diameter, with 20 to 80 graphene layers on
the wall. The metal cores are single crystals, as evidenced by the
lack of grain boundaries in the HRTEM images. The carbon shells
are also highly crystalline. Evidence of 3-dimensional crystallinity in
the graphitic structure comes from the presence of (10l) fringes
tion (Mr) and coercive field (Hc) are measured as 58.8 emu g21
,
5.0 emu g21 and 240 Oe, respectively.{
The stability of the encapsulated products was tested by heating
the sample in air to 400 uC, followed by a slow cooling procedure to
room temperature over 12 h. The heated sample shows no change
in either sample weight or colour, suggesting good thermal stability
together with strong resistance to oxidation. This excellent property
obviously arises from the completely closed carbon structures. The
high thermal stability is also confirmed by the observation that no
degradation occurs at room temperature for over one year, and no
structural change in the heated sample can be seen in TEM
investigations.
In contrast to the electrical arc-discharge technique, this direct
salt-conversion route gives far higher yields at a much lower
temperature. Similarly, in comparison with the chemical vapour
deposition (CVD) method, our new technique does not require a
˚
observed crossing the 3.4 A layer fringes. The encapsulation occurs
independently of the particle size. No naked Fe nanoparticles have
been observed under intensive microscope observations. This is in
2 4 4 2
C h e m . C o m m u n . , 2 0 0 4 , 2 4 4 2 – 2 4 4 3
T h i s j o u r n a l i s ß T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
Geng, Junfeng
