ArticlePDF Available

Enhancement of L10 phase formation in FePt nanoparticles by nitrogenization


Abstract and Figures

FePt particles 6 nm in size are produced by argon/nitrogen sputtering and gas-phase condensation. Marked changes in the atomic structure and morphology of the particles occur upon addition of nitrogen to the sputter gas. Electron energy loss spectroscopy and x-ray absorption spectroscopy show that nitrogen is incorporated in the particles in molecular and compound form. In situ sintering of the particles drives out the nitrogen causing enhanced diffusion leading to the preference of the L10 structure over the multiply twinned icosahedral structure, which forms in the absence of nitrogen in the sputter gas.
Content may be subject to copyright.
J. Phys. D: Appl. Phys. 39 (2006) 4741–4745 doi:10.1088/0022-3727/39/22/001
Enhancement of L10phase formation in
FePt nanoparticles by nitrogenization
O Dmitrieva1, M Acet1, G Dumpich1, J K¨
astner1, C Antoniak1,
M Farle1and K Fauth2
1Fachbereich Physik, Experimentalphysik–AG Farle, Universit¨
at Duisburg-Essen, D-47048
Duisburg, Germany
2Max Planck Institut f¨
ur Metallforschung, D-70569 Stuttgart, Germany
Received 29 May 2006, in final form 19 September 2006
Published 3 November 2006
Online at
FePt particles 6 nm in size are produced by argon/nitrogen sputtering and
gas-phase condensation. Marked changes in the atomic structure and
morphology of the particles occur upon addition of nitrogen to the sputter
gas. Electron energy loss spectroscopy and x-ray absorption spectroscopy
show that nitrogen is incorporated in the particles in molecular and
compound form. In situ sintering of the particles drives out the nitrogen
causing enhanced diffusion leading to the preference of the L10structure
over the multiply twinned icosahedral structure, which forms in the absence
of nitrogen in the sputter gas.
1. Introduction
The search for materials serving as high density magnetic data
storage media aims to produce nanomagnets with minimum
dimensions and with temporally stable magnetization beyond
the superparamagnetic limit. There are several materials
having high magnetic anisotropy, such as CoPt, FePd and FePt
[1]. Among these, Fe100xPtxthin film alloys with x50 in
the L10phase stand out as having a particularly high magneto-
crystalline anisotropy of (6–10)×106J m3[2,3]. Similar
values are aimed to be attained in nanometre sized grains and
The thermodynamic equilibrium L10phase of bulk FePt
consists of layers of Fe and Pt that are stacked along the [001]
direction (c-axis). The structure is tetragonally distorted along
this direction with (c/a) =0.966; aand cbeing the lattice
parameters. The formation of the L10structure in FePt occurs
below 1500 K, above which the structure is face centred cubic
(FCC). The L10phase forms by volume diffusion. On cooling
to room temperature from such high temperatures at moderate
cooling rates, sufficient diffusion allows for the formation of
this phase without the necessity of long-time annealing at
elevated temperatures. The same would intuitivelybe expected
for alloys at reduced dimensions such as in thin films and
nanoparticles. However, several works on such materials have
shown that the L10phase is kinetically suppressed, and post
preparation heat treatment is indeed necessary for forming
the L10phase [48]. The disadvantage of heat treatment is
that it leads to the unwanted effect of grain growth in the
case of films and particle coalescence in the case of wet
chemically prepared nanoparticles, both of which contravene
the minimum dimension condition for magnetic data storage.
An alternative to the wet chemical preparation method of
FePt nanoparticles is the gas phase preparation method, by
which coalescence can be avoided in the heat treatment phase.
This method employs a sequence of dc sputtering, inert gas
condensation and subsequent in situ in-flight sintering prior
to deposition on a substrate. The in-flight sintering is also
the heat treatment stage [9,10]. We have previously reported
that FePt nanoparticles prepared with this method show no
agglomeration and have a mean diameter of about 6 nm with
a narrow lognormal size distribution of geometrical standard
deviation 1.1 [11,12]. Electron microscopy investigationshave
shown that the particles have predominantly a multiply twinned
icosahedral morphology. No occurrence of the ordered L10
phase in these particles up to a sintering temperature of 1273 K
has been observed, and the particles are superparamagnetic at
room temperature [12,13]. The non-occurrence of the L10
phase is ascribed to inadequate diffusion in the icosahedral
particles, most probably resulting from the lack of a sufficient
amount of vacancies.
In order to produce magnetically hard nanoparticles,
it is therefore required to boost the formation of the
thermodynamically favourable L10-phase by providing for
diffusion with means other than post preparation or in situ
heat treatment. One method in which it is possible to avoid
0022-3727/06/224741+05$30.00 © 2006 IOP Publishing Ltd Printed in the UK 4741
O Dmitrieva et al
the formation of multiply twinned particles is to introduce
a supply of oxygen during particle preparation when using
the sputtering technique [14]. Although using this method
leads to the destabilization of the icosahedral structure and
to the formation of single-crystalline FCC nanoparticles, no
formation of the L10phase is observed. The surface free
energy of the {111}facets, which build up the multiply twinned
icosahedral particles, is minimum in nanoparticles of systems
that are FCC in their bulk material counterpart [15,16]. In
the presence of oxygen, which acts as a surfactant, the free
energy is favoured for {100}facets [17], and the icosahedral
structure is destabilized leading to the formation of FCC
crystalline particles [14]. However, the absence of the L10state
indicates that adequate volume diffusion is still insufficient in
this method.
Irradiation is another method that can enhance diffusion.
It has been demonstrated that post-growth He-ion irradiation
of sputter-grown FePt (001) films furthers the formation of
the L10phase by radiation-enhanced mobility [18,19]. These
results were also supported by Monte Carlo simulations [20].
A further way to boost Fe and Pt diffusion is to introduce
nitrogen during sample preparation and afterwards allow it to
effuse by annealing, thereby increasing the mobility of the
atoms. Nitrogen takes up octahedral interstitial positions in
the FCC lattice of Fe incorporating 3d systems and effuses at
temperatures above about 950 K [21]. In nitrogenizing steels
by sputtering, nitrogen can penetrate down to millimeters into
the surface of the metal, meaning that in thin films the complete
volume of the sample can be nitrogenized. Indeed, FePt thin
films grown with nitrogen added to the sputtering gas showed
enhanced coercivity when the films were annealed [22,23].
The increased coercivity has been discussed to be related to
the formation of the L10state during annealing. Annealing
causes nitrogen to escape and, thereby, increases the mobility
of Fe and Pt atoms. This process can be expected to occur in
gas-phase prepared FePt nanoparticles as well.
In this work, we prepare FePt nanoparticles by sputtering
in the presence of nitrogen. Nitrogen is introduced into the
particles during nucleation [24]. Subsequent in-flight sintering
of the particles should then drive out the nitrogen and cause
an increase in the mobility of the Fe and Pt atoms (as in
the case of thin films) and lead to the formation of the L10
state. Indeed, we find by x-ray absorption spectroscopy (XAS)
and electron energy loss spectroscopy (EELS) that nitrogen
incorporated into the lattice of the primary nanoparticles leaves
the particles after sintering. High-resolution transmission
electron microscopy (HRTEM)studies on the sintered particles
show that the L10state is formed in about 70% of the particles.
2. Experimental
FePt nanoparticles with and without nitrogen are prepared by
the inert gas condensation method based on a dc sputtering
process from alloy targets in a continuous gas flow of
argon, helium and nitrogen. The sputtering targets with
purity 99.99% were purchased from MaTecK-GmbH with
the composition Fe50Pt50 . The magnetron sputtering gun
was set to a power of 250 W. During particle preparation,
the total gas pressure is adjusted to an optimized value of
0.5 mbar. After nucleation and growth of primary particles
in the nucleation chamber, of which the walls are cooled
with liquid nitrogen, the particles travel through a sintering
furnace at temperatures adjustable between room temperature
and 1270 K. The particles traverse the furnace in about 0.1s
and subsequently deposit thermophoretically onto a liquid
nitrogen-cooled substrate holder. Further details of the particle
preparation are discussed in [12]. We use commercially
available copper grids covered with amorphous carbon as
substrates for structural and morphological investigations
and naturally oxidized silicon substrates for XAS. The
mean metallic composition of the individual particles was
determined by spatially resolved energy dispersive x-ray
analysis to be at 48 at% Fe with a standard deviation of 6%.
Thus, the atomic compositions of the particles are mostly
distributed within the stability range of the L10-ordered phase.
The observed standard deviation of the atomic composition of
the gas-phase prepared FePt particles is less than that for the
solution-phase prepared particles [25].
We measured the lattice parameter of the particles by
electron diffraction. A polycrystalline gold film was used as
a calibration standard. We obtain for the FePt nanoparticles
a lattice constant of 0.386 nm with a certainty of about 1%.
The lattice parameter given in the literature of similar sized
FePt nanoparticles varies from 0.376 to 0.387 nm [2628]. The
lattice parameter of bulk FePt is 0.381 nm [29]. In our particles
we observe a slight expansion of the lattice compared with the
The structure and morphology of the FePt nanoparticles
are characterized by HRTEM using a 200kV Philips Tecnai
F20 with field emission gun and Gatan imaging filter. EELS
and XAS studies were carried out to examine the presence of
nitrogen in the primary particles. XAS on the particles was
investigated using soft x-rays at the PM-3 bending magnet
beamline of the BESSY II synchrotron facility in Berlin,
Germany. Data were obtained in the total electron yield
(TEY) mode by measuring the sample drain current under
x-ray illumination. The intensity was normalized to the
simultaneously recorded TEY from a gold grid.
3. Results
3.1. Structure and morphology
In figure 1(a), we show a low magnification TEM micrograph
of FePt nanoparticles prepared in the absence of nitrogen at
a nucleation pressure of 0.5 mbar and sintering temperature
of 1270 K. The sputtering time is 10min. The particles have a
narrow lognormal size distribution with mean particle diameter
of 5.9 nm and geometric standard deviation of 1.07 as in [12],
where the same parameters were used. The upper part of
figure 1(b) shows a high resolution image of a typical particle
with multiply twinned icosahedral morphology. The observed
morphology is also supported by the results of structure
simulations [30,31]. The image displays the typical contrast
of an icosahedron observed along its three-fold symmetry axis.
The surface of the icosahedron, which consists of 20 tetrahedra,
is bounded by {111}crystal planes. The power spectrum of
the image obtained by Fourier transformation is shown in the
lower part of figure 1(b) and exhibits the characteristic pattern
with elongated {111}reflections. These multiply twinned
Enhancement of L10phase formation in FePt nanoparticles
a) b)
Figure 1. Particles prepared in the absence of nitrogen. (a) Low
magnification image. (b) Upper panel is a high resolution image
showing the multiply twinned morphology of the particles. The
lower panel is the power spectrum exhibiting the characteristic
elongated {111}reflections.
Table 1. Particle diameter dpwith geometric standard deviation σG
and amount of different structured particles in dependence of
nitrogen-to-argon ratio. In the brackets we show the percentage of
L10ordered particles found among those that are single-crystalline
and those that are polycrystalline.
nitrogen- multiply single- poly-
to-argon dptwinned crystalline crystalline
ratio (nm) σG(%) (%) (%)
0 : 1 5.9 1.07 87 0 13
1 : 9 5.9 1.06 29 32 (19 L10) 32 (9 L10)
1 : 4 5.6 1.14 8 70 (23 L10) 22 (21 L10)
1 : 2 6.0 1.20 6 70 (24 L10) 24 (20 L10)
FePt nanoparticles are found to be stable up to an annealing
temperature of 1470 K and exhibit no transformation to the L10
ordered phase. Using HRTEM, we characterized the particles
by morphology. We find that about 87% of the particles are
multiply twinned and 13% are FCC polycrystalline.
We now consider the investigations of the influence of
nitrogen addition during the nucleation of the particles. In
order to keep the same nucleation and flight-time conditions,
e.g. the same gas pressure and the same total carrier gas
flow, we substitute argon by nitrogen. We have investigated
several nitrogen-to-argon ratios between 1 : 9 and 1 : 2. The
nucleation pressure and sintering temperatures were set as
0.5 mbar and 1270 K, respectively. In table 1we present
the morphological and structural changes of the particles
generated at different nitrogen contents. The mean diameter
of the particles remains nearly the same as in the case for
particles prepared without nitrogen. However, the geometrical
standard deviation increases from 1.07 to 1.20 with increasing
nitrogen content indicating that the monodispersity is slightly
reduced. This can also be seen in figure 2(a) which shows a
low magnification TEM image of the particles prepared with
nitrogen added to the carrier gas with 1 : 2 ratio. The sputtering
time increases with increasing nitrogen content ranging from
10 min for 0 : 1 to 60min for 1: 2 due to the lower partial
content of argon.
The presence of nitrogen during sputtering indeed leads
to considerable changes in the structure of the particles. As
is shown in table 1, in the presence of nitrogen the amount of
multiply twinned particles reduces from 87% to 6%, whereas
the part of single-crystalline increase from 0% to 70%. The
HRTEM image in the upper part of figure 2(b) shows a single-
crystalline FCC FePt nanoparticle. The plane of the figure is
perpendicular to the [110] crystallographic orientation. The
particle is bounded by {100}and {111}facets and thus has
cuboctahedral morphology [31,32]. The power spectrum of
the image in the lower part of figure 2(b) shows the typical
pattern of the FCC structure seen along the [110] direction.
Furthermore, the single-crystalline and polycrystalline
particles prepared under nitrogen addition are partly
transformed to the L10ordered state. The single-crystalline
FePt nanoparticle shown in the upper part of figure 2(c) exhibits
alternating bright and dark contrasts of the lattice planes
which is typical for L10ordering. The Fourier transformed
pattern shown in the lower part shows supplementary {001}
superstructure reflections in addition to the regular FCC type
diffraction pattern. In table 1we present also the amount of the
particles which shows these characteristic contrasts estimated
by counting HRTEM-images.
Layered contrast such as in the upper part of figure 2(c)
is only observed when the particles are favourably aligned
on the substrate. This means that other single crystalline or
polycrystalline particles, which are not favourably aligned,
can also be in the L10state. In order to estimate the correct
amount of L10ordered particles among the single crystalline
and polycrystalline that show no layered contrast, we use a
statistical method combined with contrast simulations. We
have devised a criteria for the observance of both the L10
super structure and the disordered FCC structure depending
on the orientation of the particles on the substrate [33]. These
criteria are derived from HRTEM contrast simulations taking
into consideration the angle between the electron beam and
an appropriate zone axis. The resulting critical tilt angles for
the visibility of ordered super-lattice and the disordered FCC
lattice are αord =3and αord =5, respectively. Furthermore,
we have taken into account the different possibilities to observe
both lattices with respect to the crystal direction of the
nanoparticles. These criteria correct the statistically counted
amount of the L10ordered and FCC disordered particles.
Using this method, we estimate that at the nitrogen ratio of
1 : 2 about 70% of all single crystalline and polycrystalline
FePt nanoparticles are in the L10state.
3.2. EELS and XAS
In order to investigate how nitrogen interacts with the FePt
particles during preparation, we have made nitrogen-sensitive
spectroscopic measurements using EELS and XAS on primary
(non-sintered) particles and particles sintered at 1270 K. The
primary particles are expected to contain nitrogen, whereas
nitrogen should be absent in the sintered particles due to high
temperature exposure.
Figure 3shows background-treated EELS data in the
energy range 380–500 eV on primary and sintered particles.
The energy resolution is 0.9 eV. For clarity, the data are
smoothed by 20-point adjacent averaging and are represented
O Dmitrieva et al
a) b) c)
Figure 2. Particles prepared in the presence of nitrogen. (a) Low magnification image. (b) The upper high resolution image shows a FCC
cuboctahedral particle with (111) and (100) facets. The lower panel shows the power spectrum related to the FCC structure. (c) L10particle
(upper panel) with alternating bright and dark layers along the c-axis. The lower panel shows the power spectrum related to the L10structure.
Figure 3. Background-treated EELS data on primary and sintered
particles. For better visualization, the data are smoothed by 20-point
adjacent averaging and shown with a heavy line.
by the heavy lines. The data for the primary particles show
an increase in the intensity at the nitrogen ionization edge
at about 400 eV indicating that nitrogen is present in these
particles [34]. On the other hand, the data for the sintered
particles show no significant feature at this energy indicating
that nitrogen is removed from the particles on sintering.
To understand how nitrogen is incorporated in the
particles, we perform XAS studies. Figure 4shows the
normalized XAS in the energy range 395–410 eV in steps
of 0.2 eV. The spectrum of the primary nanoparticles (filled
circles) shows two pronounced absorption features with
maxima at about 398 and 401 eV. The threshold of the first XAS
edge is distributed between about 396 and 397 eV. This energy
range corresponds approximately to the the binding energy
range of 396.5–398 eV for various iron nitrides measured using
x-ray photoelectron spectroscopy [3537]. Calculations show
that the value of the threshold of the x-ray absorption edge
matches well with the binding energy [38]. Therefore, the first
maximum at 398 eV lies in an energy range corresponding to
the absorption edge of iron nitrides.
Figure 4. XAS spectrum of the primary and sintered nanoparticles.
The second peak in the XAS spectrum centred around
401 eV is due to the presence of molecular nitrogen [39,40].
Since no adsorption of nitrogen molecules on the metal surface
is expected at room temperature, some molecular nitrogen must
have been trapped within the particles during their nucleation
and growth.
The XAS spectrum of the sintered nanoparticles is shown
in figure 4by open circles. The spectrum exhibits only
feeble features at energies corresponding to the features of
the spectrum for primary particles. Thus, nitrogen has been
essentially driven out of the particles during sintering at
elevated temperatures.
4. Discussion
HRTEM studies show that the structure and morphology
of the particles are modified by introducing nitrogen into
the sputter process, and the formation of the L10phase is
facilitated. The morphology of FePt nanoparticles, which are
multiply twinned when prepared without nitrogen, becomes
predominantly single crystalline cuboctahedral in the L10
phase when prepared with nitrogen. These results support
Enhancement of L10phase formation in FePt nanoparticles
the fact that the release of the nitrogen out of the particles
during sintering aids the diffusion so that Fe and Pt atoms can
reposition to build the layered L10structure [22].
EELS studies indicate that nitrogen is introduced into the
structure of the primary particles. Using XAS, it is possible to
gain further information on how nitrogen is incorporated into
the particle during their nucleation and growth. We find that
nitrogen is not only found in the form of a compound, but it can
be also trapped in the particle in molecular form. However,
at present, it is not possible to extract detailed information
on which compounds are formed or how molecular nitrogen
becomes trapped within the particle. Nevertheless, it is thought
that molecular nitrogen is trapped into the volume of the
particle due to their rapid cooling in the condensation chamber.
After in-flight sintering at 1270 K, nitrogen is essentially driven
out of the particle.
This can be additionally supported by calculating the
diffusion length of nitrogen atoms for an iron nitride
environment. The diffusion length λcan be written as a
function of annealing time tand the diffusion coefficient D
in the form
λ=Dt . (1)
The diffusion coefficient depends on the diffusion constant D0,
temperature Tand activation energy Eathrough
D=D0exp µEa
Using the known parameter for D0and Eaof nitrogen in
iron nitride 250 cm2s1and 2.75 eV, respectively [41], we
calculate the diffusion coefficient at the sintering temperature
1270 K and annealing time t=0.1 s. λduring the flight-
sintering is then 180 nm, which is much larger than the largest
particle diameter. Thus, the nitrogen atoms are expected to
effuse completely out of the particles during flight through the
annealing area.
5. Conclusion
Using EELS and XAS, we show that by adding nitrogen to
the sputter gas of a gas-phase particle generator with in-flight
sintering, nitrogen is incorporated into the structure of FePt
nanoparticles during their nucleation and growth. Nitrogen
is driven out during the sintering phase causing enhanced
diffusion and leading to the formation of the L10-phase, as
observed with HRTEM.
This work was supported by the Deutsche Forschungsgemein-
schaft (SFB 445).
[1] Weller D, Moser A, Folks L, Best M, Lee W, Toney M F,
Schwickert M, Thiele J U and Doerner M 2000 IEEE Trans.
Magn. 36 10
[2] Weller D and Moser A 1999 IEEE Trans. Magn. 35 4423
[3] Sellmyer D, Yan M, Xu Y and Skomski R 2005 IEEE Trans.
Magn. 41 560
[4] Sun S, Murray C B, Weller D, Folks L and Moser A 2000
Science 287 1989
[5] Dai Z R, Sun S and Wang Z L 2002 Surf. Sci. 505 325
[6] Takahashi Y K, Ohnuma M and Hono K 2001 Japan. J. Appl.
Phys. 40 1367
[7] Harrel J W, Wang S, Nikles D E and Chen M 2001 Appl. Phys.
Lett. 79 4393
[8] Ping D H, Ohnuma M, Hono K, Watanabe M, Iwasa T and
Msumoto T 2001 J. Appl. Phys. 90 4708
[9] Huang Y H, Zhang Y, Hadjipanayis G C and Weller D 2003 J.
Appl. Phys. 93 7172
[10] Qiu J, Judy J H, Weller D and Wang J 2005 J. Appl. Phys.
97 10J319
[11] Stappert S, Rellinghaus B, Acet M and Wassermann E F 2002
Mater. Res. Soc. Symp. Proc. 704 571
[12] Stappert S, Rellinghaus B, Acet M and Wassermann E F 2003
J. Cryst. Growth 252 440
[13] Rellinghaus B, Stappert S, Acet M and Wassermann E F 2003
J. Magn. Magn. Mater. 266 142
[14] Stappert S, Rellinghaus B, Acet M and Wassermann E F 2003
Eur. J. Phys. D24 351
[15] Ino S 1969 J. Phys. Soc. Japan 27 941
[16] Cleveland C and Landman U 1991 J. Chem. Phys. 94 7376
[17] Li L, Kida A, Ohnishi M and Matsui M 2001 Surf. Sci.
493 120
[18] Ravelosona D, Chappert C, Mathet V and Bernas H 2000
Appl. Phys. Lett. 76 236
[19] Lai C H, Yang C H and Chiang C C 2003 Appl. Phys. Lett.
83 4550
[20] Bernas H, Attane J P, Heinig K H, Halley D, Ravelosona D,
Marty A, Auric P, Chappert C and Samson Y 2003 Phys.
Rev. Lett. 91 77203
[21] Acet M, Gehrmann B, Wassermann E F, Bach H and
Pepperhoff W 2001 J. Magn. Magn. Mater. 232 221
[22] Wang H Y, Mao W H, Ma X K, Zhang H Y, Chen Y B, He Y J
and Jiang E Y 2004 J. Appl. Phys. 95 2564
[23] Hsiao H H, Panda R N, Shih J C and Chin T S 2002 J. Appl.
Phys. 91 3145
[24] Dmitrieva O, Acet M, K¨
astner J, Dumpich G and
Wunderlich W 2006 J. Nanopart. Res.
doi 10.1007/s11051-005-9045-6
[25] Yu A C C, Mizuno M, Sasaki Y and Kondo H 2004 Appl.
Phys. Lett. 85 6242
[26] Dai Z R, Sun S and Wang Z L 2001 Nano. Lett. 1443
[27] Stahl B et al 2003 Phys. Rev. B67 014422
[28] Loc Nguyen H, Howard L E M, Giblin S R, Tanner B K,
Terry I, Hughes A K, Ross I M, Serres A, B¨
ummer H
and Evans J S O 2005 J. Mater. Chem. 15 5136
[29] Landolt H and Madelung O 1995 Numerical Data and
Functional Relationships in Science and Technology (New
series Group IV/5e) ed O Madelung (Berlin: Springer) p
[30] Urban J, Sack-Kongehl H and Weiss K 1993 Z. Phys. D28 247
[31] Yacaman M J, Ascencio J A, Liu H B and Gardea-Torresdey J
2001 J. Vac. Sci. Technol. B19 1091
[32] Dai Z R, Sun S and Wang Z L 2002 Surf. Sci. 505 325
[33] Dmitrieva O, Rellinghaus B, K¨
astner J and Dumpich G 2006 J.
Cryst. Growth submitted
[34] Krivanek O L and Ahn C C 1983 EELS Atlas (Warrendale, PA:
Gatan Inc.) p 7
[35] Wang X, Zheng W T, Tian H W, Yu S S, Xu W, Meng S H,
He X D, Han J C, Sun C Q and Tay B K 2003 Appl. Surf.
Sci. 220 30
[36] Zhong W H, Tay B K, Lau S P, Sun X W, Li S and Sun C Q
2005 Thin Solid Films 478 61
[37] Graat P C J, Somers M A J and Mittemeijer E J 1998 Appl.
Surf. Sci. 136 238
[38] Almbladh C O and Hedin L 1983 Handbook of Synchtrotron
Radiation vol 1b ed E Koch (Amsterdam: North Holland)
pp. 607–904
[39] Chen J G 1997 Surf. Sci. Rep. 30 1
[40] Wenzel L, Arvanitis D, Schl¨
ogl R, Muhler M, Norman D,
Baberschke K and Ertl G 1989 Phys. Rev. B40 6409
[41] Chatbi H, Vergnat M, Bobo J F and Hennet L 1997 Solid State
Commun. 102 677
... In PVD technique, mainly three steps are involved: (1) the first step involves vaporization of the materials that are mainly from solid source; (2) in the second step, vaporized material is transported; and (3) the third step is the nucleation and growth for the generation of thin films and nanoparticles (Okuyama and Lenggoro 2003). Some of the frequently used PVD methods for synthesizing the nanomaterials (Hatakeyama et al. 2011;Veith et al. 2007;Bouchat et al. 2013;Asanithi et al. 2012;Ichida et al. 2014;Ghosh et al. 2007;Veith et al. 2005;Ramalingam et al. 2013;Zhang et al. 2004;Takahashi et al. 2004;Dmitrieva et al. 2006;Stellacci et al. 2002;Hsieh et al. 2010; Naessens Ong et al. 2008;Andrea et al. 2009;Jing et al. 2014) are (i) sputtering, (ii) vacuum arc, (iii) pulsed laser deposition and (iv) electron beam evaporation. ...
Free enzymes do not possess properties of recovery and reusability, and also they are not stable at wide pH and temperature range. Therefore, new ways which can enhance enzyme stability and reusability should be developed, and hence, the immobilization technique is one such approach. These immobilization techniques offer such materials which have the ability to be active in the much wide range of pH and temperature, and also they are more stable than the free enzymes. Immobilization is carried out on the nanosized material either by adsorption, covalent coupling, entrapment, encapsulation or cross-linking. These nanomaterial-immobilized enzymes show several advances over the free enzymes because of large surface area-to-volume ratio, lower mass transfer resistance and high mobility. Several nanomaterials are used for immobilizing the enzymes; however, their recovery from the reaction mixture is very poor. Therefore, the magnetic nanomaterials are more attractively used in immobilization because the enzyme immobilized through magnetic nanomaterial has the tendency to be easily separated out from the reaction mixture. These nanomaterial-immobilized enzymes show wide range of applications in biotechnology, bioanalysis, biomedicine, pathology and biosensors.
... The laser irradiation and nitriding treatment produce thus both the occurrence of a two-phase magnetic behavior, similar to an exchange spring mechanisms, and an increase in the crystallinity and of the coercive fields. Enhancements in magnetic behavior induced by nitrogen presence have been previously reported in [46][47][48][49]. In the present case, values of 934 and 688 kA/m are obtained for the samples with 6 and 9 at.% ...
Full-text available
In melt-spun FePtB-based ribbons, the addition of Ag has been proven to decrease the temperature of phase transformation from the A1 fcc FePt phase to the hard magnetic tetragonal L10 phase. Alloys with 6 and 9 at.% Ag added to the initial FePtB have been synthesized by rapid solidification from the melt. The samples have been laser irradiated and submitted to nitriding procedure. This procedure has been proven beneficial for inducing complete transformation of A1 to L10 phase and a strong (001) texturing. Ag segregation combined to mechanisms of creation of vacancies and diffusion of N give rise to the formation of an intergranular disordered region and due to an improved interfacial coupling between FePt grains, enhanced coercivity and two-phase magnetic behavior is obtained.
... Il semblerait que le titane s'insère dans le réseau de CoPt et induise des déformations importantes (à cause de la grande taille des atomes de Ti). L'azote peut s'insérer dans le réseau cristallin de FePt comme l'ont montré des résultats de spectroscopie d'absorption X. L'azote en raison de son fort coefficient de diffusion augmente la diffusion des atomes métalliques et permet d'obtenir la phase L1 0 à plus basse température [Dmitrieva 2006]. Le cas du cuivre est particulier puisque dans certains travaux il a pour effet de diminuer la température de mise en ordre chimique [Gibot 2005] et dans d'autres cas aucun effet n'est observé ], mais il faut tenir compte du fait que le cuivre forme un alliage avec le FePt : la Tulameenite (FeCuPt 2 ) qui présente également une phase L1 0 , un alliage semblable n'a pas été observé avec le cobalt platine. ...
Full-text available
The aim of this thesis is to study the links between morphological, structural and magnetic properties of CoPt nanoparticles. The importance of this study is that CoPt nanoparticles constitute a promising material in the field of magnetic recording. It is therefore essential to determine the relationship between the magnetic properties and organization of nanoparticles at atomic scale. First we studied the structure and morphology of these objects by using complementary techniques: electron microscopy, X-ray scattering and X-ray absorption. The results show structural transitions depending on parameters such as temperature, size or nanoparticles growth mode. Thus, when growth atom by atom, transitions from icosahedral to face-centered cubic structures are observed. However, coalescence allows the formation of decahedral structure. The chemically ordered structure that is most interesting for the magnetic data storage has been obtained by annealing at 630 ° C. In a second step we studied the magnetic properties of nanoparticles by SQUID magnetometry and X-ray magnetic circular dichroism. Results show clear links between structures and magnetic properties. These studies were conducted on alloyed and core / shell particles. Interface effects were identified, and annealed samples shows an increase of the magnetic moment.
... Bimetallic FePt and CoPt nanoclusters have been subject of intense experimental [40][41][42][43][44][45][46][47][48] as well as theoretical [14,[49][50][51][52][53][54][55][56] investigations, in particular because of their potential use in miniaturized storage media requiring monodisperse FePt nanoparticles and ferromagnetic FePt nanocrystal superlattices [57][58][59] and in nanocatalysis where the catalytic properties of highly dispersed supported bimetallics are used [60,61]. ...
Full-text available
This brief overview summarizes the state-of-the-art of simulations of transition metal nanoclusters based on density functional theory calculations. Besides the monometallic clusters like iron, we focus on alloy nanoclusters like Fe-Pt, Co-Pt and (Ni, Co)-Mn-Ga which are of current interest for recording media and actuators involving the magnetic shape memory effect, respectively. Although catalysis is not the subject of the present paper, trimetallic nanoclusters are of special interest because the third element can be used to achieve higher catalytic and selective properties compared to the corresponding monometallic and bimetallic clusters. For clusters of Fe-Pt and Co-Pt below a critical size, the L12 structure with its technologically relevant high magnetocrystalline anisotropy, is difficult to stabilize. For trimetallic systems like Ni-Mn-Ga, the rather versatile properties of the bulk material can be used to achieve shape changes or magnetocaloric effects (depending on the composition) also in nanoclusters. More importantly, it might be cheaper to manufacture the nanocrystalline materials from the trimetallic nanoclusters than to fabricate corresponding single-crystal bulk systems.
FePt system attracts currently a great deal of interest for applications as future RE free permanent magnets. Among the key issues to be solved one may count the decreasing of the ordering temperature and improvement of magnetic behavior. For that purpose we have studied the effect of a nitriding post-synthesis procedure on the FePtAgB melt spun ribbons, aimed at refining the microstructure and enhancing the magnetic performances. Deep structural characterization by transmission electron microscopy, electron diffraction, energy dispersive X-ray spectroscopy and X-ray diffraction allowed us to observe the morphology and to correctly assign and identify the nature of the main granular phases observed. Nitriding procedure is shown to strongly enhance the (001) texturing and the degree of ordering of the L10 FePt phase, as well as largely increase of coercivity, compared to the as-cast state. These changes are interpreted in terms of Ag segregation towards intergranular region associated to N diffusion and creation of vacancies that favor consistently the process of ordering the FePt grains into the L10 tetragonal phase.
Full-text available
We have investigated the spin and orbital magnetic moments of Fe in FePt nanoparticles in the $L$1$_{0}$-ordered phase coated with SiO$_{2}$ by x-ray absorption spectroscopy (XAS) and x-ray magnetic circular dichroism (XMCD) measurements at the Fe $L_{\rm 2,3}$ absorption edges. Using XMCD sum rules, we evaluated the ratio of the orbital magnetic moment ($M_{\rm orb}$) to the spin magnetic moment ($M_{\rm spin}$) of Fe to be $M_{\rm orb}/M_{\rm spin}$ = 0.08. This $M_{\rm orb}/M_{\rm spin}$ value is comparable to the value (0.09) obtained for FePt nanoparticles prepared by gas phase condensation, and is larger than the values ($\sim$0.05) obtained for FePt thin films, indicating a high degree of $L$1$_{0}$ order. The hysteretic behavior of the FePt component of the magnetization was measured by XMCD. The magnetic coercivity ($H_{\rm c}$) was found to be as large as 1.8 T at room temperature, $\sim$3 times larger than the thin film value and $\sim$50 times larger than that of the gas phase condensed nanoparticles. The hysteresis curve is well explained by the Stoner-Wohlfarth model for non-interacting single-domain nanoparticles with the $H_{\rm c}$ distributed from 1 T to 5 T.
Fe–Pt alloys are known to exhibit high coercivity due to high magnetocrystalline anisotropy (MCA) of the L10 FePt phase. The main intrinsic magnetic properties of this itinerant-electron ferromagnet are reported to be Tc =750 K (the Curie temperature), Js = 1.43 T (the spontaneous magnetisation at room temperature), and K1 = 6.6MJ/m3 (the firstmagnetic anisotropy constant at room temperature). L10-based Fe–Pt alloys have a huge potential for a variety of applications, most importantly, in the field of magnetic recording and specialized permanent magnet applications. Because of the unique combination of excellent intrinsic magnetic properties and good corrosion resistance, L10-based Fe–Pt thin films and nanoparticles are promising candidates for ultrahigh-density magnetic storage media. FePt nanoparticles are also considered for such applications as contrast agents in magnetic resonance imaging for bio-medical applications or for catalysis. The structure and, consequently, the magnetic properties of the Fe–Pt alloys differ depending on the approach used for their preparation, especially when such aspects as the size dependence of the chemical ordering in very small nanoparticles or the influence of a lattice misfit between a film and a substrate start to play a major role.
Full-text available
Size-selected SnO1.8:Ag mixed nanoparticle films have been prepared using a gas phase condensation method. Transmission electron microscopy showed that the applied size-selection technique yields well-defined, monodisperse and spherical SnO1.8 and Ag nanoparticles, both with a fixed diameter of 20 nm. The technique allows an independent variation of the particle size of both materials as well as the concentration of Ag. It allows to assess the influence of these parameters on the gas-sensing properties of the films, here for ethanol vapor in synthetic air. SnO1.8:Ag nanoparticle films show optimal values of the sensor signal and response time at a Ag nanoparticle concentration of 5%. Due to the fact that the Ag nanoparticles are clearly distinct from the SnO1.8 nanoparticles in the film, the most probable mechanism leading to improved sensor properties is chemical sensitization via a spill-over effect.
Full-text available
The energetics of nickel clusters over a broad size range are explored within the context of the many-body potentials obtained via the embedded atom method. Unconstrained local minimum energy configurations are found for single crystal clusters consisting of various truncations of the cube or octahedron, with and without (110) faces, as well as some monotwinnings of these. We also examine multitwinned structures such as icosahedra and various truncations of the decahedron, such as those of Ino and Marks. These clusters range in size from 142 to over 5000 atoms. As in most such previous studies, such as those on Lennard-Jones systems, we find that icosahedral clusters are favored for the smallest cluster sizes and that Marks’ decahedra are favored for intermediate sizes (all our atomic systems larger than about 2300 atoms). Of course very large clusters will be single crystal face-centered-cubic (fcc) polyhedra: the onset of optimally stable single-crystal nickel clusters is estimated to occur at 17 000 atoms. We find, via comparisons to results obtained via atomistic calculations, that simple macroscopic expressions using accurate surface, strain, and twinning energies can usefully predict energy differences between different structures even for clusters of much smaller size than expected. These expressions can be used to assess the relative energetic merits of various structural motifs and their dependence on cluster size.
The initial oxidation of α-Fe and ε-Fe2N1−x, subjected either to a sputter cleaning pretreatment or a sputter cleaning plus additional annealing pretreatment, was investigated with XPS. The samples were oxidised at pO2=8·10−5 Pa and temperatures ranging from 300 to 600 K. From the Fe 2p and O 1s spectra the thickness and composition of the oxide film was determined. The composition of the oxide films formed on α-Fe and on ε-Fe2N1−x was only a function of oxidation temperature and film thickness and was independent of the composition or pretreatment of the substrate. Analysis of the N 1s spectra provided information on the electric charge on nitrogen atoms and the depth distribution of nitrogen. A linear relation was found between the N 1s electron binding energy and the nitrogen concentration in the substrate. Upon oxidation of the iron nitride, nitrogen atoms accumulated underneath the oxide film. If the nitrogen concentration at that location exceeded the maximum solubility of nitrogen in ε-Fe2N1−x an additional N peak appeared in the N 1s spectrum, which indicated the formation of a nitrogen containing phase other than ε-Fe2N1−x at the nitride–oxide interface. For the oxygen exposures applied, the oxide-film thickness decreased with increasing nitrogen concentration in the substrate. The effect of nitrogen in the substrate on the initial oxidation was evaluated from the results.
Growth mode of epitaxial ultrathin γ-Fe film on the Cu(0 0 1) substrate with OR45° reconstruction surface has been investigated via reflection high energy electron diffraction (RHEED) observation. The γ-Fe structure and surface OR45° reconstruction were kept till 45 monoatomic layers (ML) with the streak pattern as proof of non-island growth mode. All oxygen atoms adsorbed on Cu(0 0 1) are floated onto 45 ML γ-Fe surface considering from the RHEED patterns. The oxygen in the state of Cu(0 0 1)–OR45° reconstruction surface exhibited a strong surfactant function for the γ-Fe growth on the Cu(0 0 1) with stabilizing effect on the film. We succeeded to prepare relatively thick and pure γ-Fe film on Cu(0 0 1) by this method. Magnetization of a series of γ-Fe films was measured as a function of Fe thickness. A part of γ-Fe layer shows non-magnetism in the low spin state.
Different phases of iron nitrides were prepared by reactive sputtering. The release of nitrogen and the crystallographic transformations during annealing were monitored by thermal desorption spectrometry and X-ray diffraction experiments. Finally, the diffusivity of nitrogen in the γ′-Fe4N phase was evaluated.
Multislice calculations have been performed for Ag, Pd and Au clusters in the size range of ≃ 5.0 nm diameter of cuboctahedral, icosahedral and decahedral structures. It could be shown that tilt series are necessary for the classification of the structures. Particularly for arbitrary orientations, i.e. deviations from main directions such as 2-, 3- and 5-fold axes, the performance of computer simulations is mandatory. The influence of absorption is also studied for the case of a 100 kV microscope by introducing a complex potential.
We demonstrate that the long-range order parameter S of sputtered FePt (001) films may be improved by using postgrowth He ion irradiation. This was demonstrated both on disordered (S~0) and partially ordered (S~0.4) films in which S was increased up to 0.3 and 0.6, respectively. X-ray diffraction analysis showed that these changes are due to irradiation-induced chemical ordering. The changes in the magnetic hysteresis loops correlate with the expected perpendicular magnetic anisotropy increase. This method may find applications in ultrahigh-density magnetic recording.
Nanoparticles are a state of matter that has properties different from either molecules or bulk solids. In the present work, we review the shape and structure of nanometer-sized particles; several shapes are discussed, such as the octahedron and truncated octahedron, the icosahedron, the Marks decahedron, the truncated “star-like” decahedron, the rounded decahedron and the regular decahedron. Experimental high-resolution transmission electron microscopy (TEM) images of each type of particle are presented together with the Fast Fourier Transform and a model of the particle. We consider only gold particles grown by vapor deposition or by colloidal methods. High-resolution TEM images of the particles in different orientations are shown. We discuss two basic types of particles uncapped and capped. Data for other metals and semiconductors are reviewed. We have also performed very extensive simulations obtaining the total energy and pair correlation functions for each cluster under study. Furthermore, distributions of single atom energy for every cluster are displayed in order to reveal the effect of surface on the stability of different types and sizes of clusters. We discuss the structure of the particles from ∼1 to ∼100 nm. The mechanisms for stress release as the particles grow larger are reviewed and a mechanism is suggested. Finally, we discuss the parameters that define the shape of a nanoparticle and the possible implications in technological applications. © 2001 American Vacuum Society.
We report that fully ordered L1(0) FePt magnetic thin films with high coercivity can be fabricated by a simple sputtering method at a substrate temperature of 300degreesC. Although fcc --> L1(0) ordering proceeds only above 600degreesC in ex situ annealing of disordered FePt films, the ordering occurs spontaneously by surface diffusion during the sputtering process of FePt alloy upon heating the substrate at 300degreesC. High coercivity larger than 7 kOe was achieved from the as-sputtered state.