FePd Microwave

8/3/2019 FePd Microwave

http://slidepdf.com/reader/full/fepd-microwave 1/8

Synthesis of monodispersed fcc and fct FePt/FePd nanoparticles bymicrowave irradiation{

H. Loc Nguyen,a Luciano E. M. Howard,a Sean R. Giblin,bd Brian K. Tanner,b Ian Terry,b

Andrew K. Hughes,a Ian M. Ross,c Arnaud Serres,b Hannah Burckstummera and John S. O. Evans*a

Received 19th August 2005, Accepted 5th October 2005First published as an Advance Article on the web 24th October 2005

DOI: 10.1039/b511850f

A simple microwave heating method has been used for the stoichiometrically controlled synthesis

of FePt and FePd nanoparticles using Na2Fe(CO)4 and Pt(acac)2/Pd(acac)2 as the main reactants.

By varying the solvents and surfactants, the microwave assisted reactions have shown a significant

advantage for the rapid production of monodisperse fcc FePt nanoparticle metal alloys which can

be converted to the fct phase at low temperatures (364 uC). Microwave reactions at high pressure

(closed system) have led to the direct formation of a mixture of fcc and fct phase FePt

nanoparticles. Room temperature structural and magnetic properties of materials have been

characterized by X-ray diffraction, HRTEM and magnetic measurements. The onset of ordering

has been investigated by in situ high temperature X-ray diffraction studies.

Introduction

The preparation of nanoscale magnetic materials is an

extremely active research area due to their potential uses in

magnetic recording devices, biomedical applications, magne-

tooptical systems and in numerous other areas.1 Of the many

nanoparticle alloys that have been studied for future genera-

tion magnetic storage applications, self-assembled Ll0 FePt

nanoparticle arrays are promising candidates owing to their

large uniaxial magnetocrystalline anisotropy [K u $ 7 6

107 erg cm23] and good chemical stability.2 Calculations

indicate that particles as small as 2.8 nm have a sufficient

anisotropy energy K uV (V is the magnetic grain volume) to be

exploited for permanent data storage, leading to significant

advances in hard disk drive areal densities over materials

currently used.3

Many approaches to the preparation of metal nanoparticles

have been reported4 including chemical reduction,5 UV

photolysis,6 thermal decomposition,7 metal vapour decom-

position,8 electrochemical synthesis9 and sonochemical decom-

position.10 Chemical routes11 appear to offer the best route to

monodisperse FePt nanoparticles.2,7a In a typical preparation

simultaneous decomposition of iron pentacarbonyl and reduc-

tion of platinum acetylacetonate by polyol reducing agents

or co-reduction of iron and platinum salts in the presence

of surfactants leads to formation of face centered cubic (fcc)

FePt alloys.

To obtain self-assembled Ll0 FePt nanoparticle super-

lattices, which are required for storage applications, the

as-synthesized nanoparticles typically have to be annealed at

high temperature to transform the material from the fcc Fe/Pt

disordered phase to the face centered tetragonal (fct) Fe/Pt

ordered phase, the so called Ll0 structure (Fig. 1). During the

annealing process, however, agglomeration of the particles

can lead to a dramatic increase in both particle size and

size dispersion.8c,12 This hinders applications as high-density

recording materials. Different methods have been attempted to

lower the FePt phase transition temperature (T t) and particle

sintering or to establish a direct route to fct nanoparticleformation. Introduction of a third metal into FePt alloys,13

although reported at lower T t, has resulted in particles which

retain the problems of agglomeration or decomposition on

further annealing at higher temperature. Partially ordered fct

FePt nanoparticles have recently been obtained by chemical

routes including the simultaneous reduction of Fe(II)/Pt(II)

salts and from Fe(CO)5/Pt(acac)2 using conventional heating

methods.14 These preliminary results generally show a low

ordering ratio, small room temperature (RT) coercivity of fct

particles and frequently relatively broad particle size disper-

sion. Recently a multistep process involving coating particles

with an inert silica coating during annealing followed by its

subsequent removal in base has been described. This process

aDepartment of Chemistry, University Science Laboratories, Universityof Durham, South Road, Durham, UK DH1 3LE.E-mail: [email protected] bDepartment of Physics, University Science Laboratories, University of Durham, South Road, Durham, UK DH1 3LE cDepartment of Electronic and Electrical Engineering, University of Sheffield, Mappin Building, Mappin Street, Sheffield, UK S1 3JDd Present address: ISIS facility, Rutherford Appleton Laboratory,Chilton, Didcot, OXON.{ Electronic supplementary information (ESI) available: Magneticresults of sample 1, 2, 4, HRTEM and SAED results of sample 1 of Table 1. See DOI: 10.1039/b511850f

Fig. 1 Schematic representation of the FePt phase transformation

from the fcc to fct structure.

PAPER www.rsc.org/materials | Journal of Materials Chemistry

5136 | J. Mater. Chem., 2005, 15, 5136–5143 This journal is ß The Royal Society of Chemistry 2005

View Online

http://dx.doi.org/10.1039/B511850F



can apparently lead to particle ordering without sintering.15

Disordered fcc particles have also been annealed to the fct

phase with minimal sintering in a NaCl matrix.16

In a recent communication, we have presented a straight-

forward stoichiometrically controlled synthesis of FePt nano-

particles using Collman’s reagent, Na2Fe(CO)4, as a reducing

agent for platinum acetylacetonate, Pt(acac)2.17 An advantage

of the method is that the electrons required to reduce Pt(II) arelocated on the Fe source rather than on an additional species

(the reaction can be schematically written as Fe22 + Pt2+ A

FePt). This process assures the ideal 1 : 1 stoichiometry is

achieved which is important since the magnetically important

fct phase only forms over an Fe12xPt range of x y 0.4–0.6;

other workers have shown that conventional procedures lead

to individual particles with a range of stoichiometries.18

Further, the reduction step that is key to nanoparticle alloy

formation requires the simultaneous presence of Fe and Pt

ions to occur, leading to the product alloy being intimately

mixed on an atomic scale. Using this route we have shown

that it is possible to produce fcc FePt nanoparticles which can

be converted to the fct structure at low temperatures withminimal agglomeration without the presence of a third metal.

By varying the surfactants and temperature regimes it was also

possible to synthesise FePt nanoparticles with the important

fct structure directly in solution, without any post-synthesis

heat treatment.17

In order to improve the FePt nanoparticle preparation

reported in our preliminary work, an alternative method of

energy supply has been investigated for heating reactions more

efficiently. Microwave dielectric heating has recently attracted

the attention of chemists for, inter alia, organic reactions,19

molecular sieve preparation,20 and syntheses of inorganic

complexes21 as it can lead to much higher heating rates

than those achieved by conventional heating. The rapid anduniform heating provided by microwaves has potential benefits

for nanoparticle synthesis. Microwave irradiation was recently

reported to be a successful synthetic method for single metal

nanoparticles such as Pt, Ir, Rh, Pd, Au, Ru22 and Ag.23

Application of microwave dielectric heating for binary Pt–Ru

nanoparticles was also reported, in which a uniform size of

2–3 nm was obtained in the presence of a polymer as a

protective layer for the particles.24 We are only aware of one

previous publication on the synthesis of FePt by microwave

methods in which platinum(II) chloride and iron(II) acetate

were reduced in ethylene glycol; to achieve a 1 : 1 stoichiometry

excess Fe was used.25 In this work the as-prepared super-

paramagnetic material was described as being amorphous (nopeaks were present in its X-ray diffraction pattern) and

crystalline FePt was only formed on heating to 600 uC. Little

characterisation of the as-prepared material was given and

the annealed material was reported as having a bimodal size

distribution. Selected area electron diffraction (SAED) pat-

terns reported did not show the ordering peaks one would

expect for fct FePt and certain ordering peaks (e.g., the 001,

112 and 113 peaks based on a pseudo-cubic cell setting)

appeared to be missing from the X-ray data presented.

In this work we have investigated the use of microwave

irradiation as an energy source for the preparation of FePt

using the Fe22/Pt2+ methodology. We show that monodisperse

crystalline fcc particles of controlled size (and thus suitable for

self-assembly) can be readily produced. The reaction has been

performed using a variety of solvents and surfactants leading

to control over particle size. Under certain conditions it is also

possible to prepare fct particles directly. We have also extended

this chemistry to the microwave synthesis of FePd nano-

particles. The structure and properties of key materials have

been characterized by X-ray diffraction (XRD), transmissionelectron microscopy (TEM) and magnetic measurements.

Experimental

Materials and instruments

Platinum acetylacetonate [Pt(acac)2] was purchased from

STREM, Pd(acac)2, disodium tetracarbonylferrate-dioxane

complex [Na2Fe(CO)4?1.5C4H8O2], dioctyl ether, oleylamine

and oleic acid from Aldrich, n-nonadecane from Lancaster.

Octyl ether was degassed for 15 min before each use. All

chemicals were weighed, placed into reaction flasks and

sealed in a N2 filled glove box before transferring to microwave

apparatus. Other AR (analytical reagents) grade organicsolvents used for purification (e.g., hexane and absolute

ethanol) were used as purchased. The microwave-assisted

reactions were carried out in a CEM 300W Discover Focus

Synthesis Microwave with a 2.45 GHz working frequency.

Reactions under ambient pressure were performed in 100 ml

thick glass walled vessels connected to a condenser under Ar.

Reactions at elevated pressure and temperatures were

performed in 10 ml sealed vials.

Nanoparticle preparation in an open vessel

In a typical reaction a mixture of Pt/Pd acetylacetonate

(0.3 mmol), disodium tetracarbonylferrate (0.3 mmol) with

appropriate amounts of surfactants and solvents (Table 1) was

placed in a 100 ml thick walled glass vessel connected to a

condenser and Ar input. The mixture was sonicated at 60 uC

for 1 h before transferring into the microwave apparatus.

Reaction was typically carried out using control parameters of

300 W max power, 250 uC max temperature and 250 psi max

pressure. A high reaction temperature (215 uC) was obtained

using octyl ether as solvent whilst lower temperature (130–

150 uC) was achieved for reactions carried out in nonadecane.

After reaction the dark product mixture was allowed to cool to

room temperature before adding 100 ml of absolute ethanol to

precipitate dark particles. The product was separated by

centrifugation then dispersed in hexane (20 ml) in the presence

of appropriate surfactants and precipitated by adding ethanol

(40 ml). After centrifugation, the material was washed one

more time in a similar solvent mixture, dried in air at room

temperature and stored under N2.

Nanoparticle preparation in a closed vessel

FePt nanoparticles were synthesized from a mixture of plati-

num acetylacetonate (0.17 mmol), disodium tetracarbonyl-

ferrate (0.17 mmol) with appropriate amounts of surfactants

and solvents for specific reactions (Table 1). Rapid heating

(1 to 5 min) to the desired reaction temperature could be

readily achieved. Safety warning: when using sealed systems

This journal is ß The Royal Society of Chemistry 2005 J. Mater. Chem., 2005, 15, 5136–5143 | 5137

View Online




there is a potential hazard due to rapid build up of high

pressure in the reaction vessel and the release of CO during the

reaction. The control system of the microwave reactor used is

designed to remove power if a rapid pressure build up is

encountered. To minimise such risks, sealed-system reactions

were typically heated to the desired temperature in 2 stages,

each taking y1 min. Despite these precautions rapid heating

that could lead to explosions was experienced on occasions;

we believe this is related to the formation of large particles

which provide a self-accelerating heating mechanism.We therefore advise caution and the use of appropriate

containment/shielding methods when such reactions are

attempted. At the end of reactions the pressure was released

and the dark mixture was allowed to cool to room temperature

before adding absolute ethanol to precipitate dark particles.

After centrifugation, the black product was dispersed in

hexane (5 ml) in the presence of the surfactants used during

the synthesis, precipitated by adding ethanol (10 ml) and

centrifuged. The materials were washed one more time, dried

in air at room temperature and stored under N2.

Characterisation methods

XRD data used to confirm sample purity and particle size were

collected on a Bruker D8 Advance diffractometer equipped

with a Cu tube and a Sol-X energy dispersive detector. The

sample was mounted on a zero background (511) silicon wafer.

Data were typically collected from 10–90u 2h (step size 5 0.02u

and time per step 5 10 s) at room temperature. A variable

divergence slit giving a constant area of sample illumination

was used. In situ variable temperature X-ray diffraction data

were collected using a Bruker AXS D8 Advance diffractometer

equipped with a Cu tube, a Ge(111) incident beam mono-

chromator (l 5 1.5406 A) and a Vantec-1 PSD. High tem-

perature measurements were performed using an Anton Parr

HTK1200 high temperature furnace. Temperature calibrationwas determined using an external Al2O3 –Si mixture of

standards.26 The powdered sample was mounted on an

amorphous silica disc. Variable temperature XRD data

were collected over a temperature range of 297–924–296 K.

Measurements (48 in total) were recorded over 48 h (every

25 K, 60 min each, a 0.2 K s21 heating/cooling rate between

temperatures, a 2h range of 5–130u and a step time of 0.33 s).

Data were rebinned onto a step size of 0.05u for Rietveld

analysis. A slow flow of 5% H2 –95% Ar gas was passed over

the sample for the experiment’s duration. XRD derived

particle sizes quoted throughout the paper were obtained

from Rietveld refinements of data sets. Peak shapes were fitted

by convolution of a Scherrer-type broadening term of form

(l/size)cosh and a strain term of the form strain 6 tanh with an

instrumental resolution function derived from a highly crystal-

line CeO2 standard recorded under equivalent conditions.

Samples for transmission electron microscopy (TEM)

analysis were prepared as dilute dispersions in hexane with a

small amount of surfactants. A drop of particle dispersion was

allowed to evaporate slowly on an amorphous carbon film

supported on a standard 3 mm copper grid (200 mesh, Agar

Scientific). High resolution TEM (HRTEM) was performed ina JEOL 2010F field-emission gun (FEG) TEM operating at

200 kV. This instrument is capable of forming sub-nanometre

analytical electron probes facilitating high spatial resolution

compositional analysis via an Oxford Instruments LINK/ISIS

X-ray energy-dispersive spectrometer (EDS) (Si/Li detector,

1024 channels, 20 keV range). EDS spectra were acquired

from single nanoparticles and also regions of the specimen

containing clusters of y300 particles using a 30 s preset live

time acquisition. Quantification of the data was performed

using the Cliff–Lorimer thin section technique assuming an

average material density of 14.6 g cm23 and a specimen

thickness equal to the average projected diameter of the

particle(s) being studied.Magnetic studies were carried out using a Quantum Design

SQUID magnetometer. Magnetization curves as a function

of applied field were measured with fields up to 50 kOe at

temperatures of 10 K and 290 K. Zero-field cooling/field

cooling (ZFC/FC) experiments were made at 100 Oe with

temperatures ranging from 2 to 300 K on samples mounted in

low background gelatin capsules. Data are presented per g

of sample owing to the difficulty in accurately assessing the

percentage of surfactant molecules in an individual sample.

Result and discussion

Initial reactions (Table 1) were performed using a 1 : 1 molarratio of Fe and Pt sources and a molar equivalent of oleyl

amine and oleic acid surfactants. Nonadecane was chosen as

solvent since experiments using conventional heating have

shown that it can lead to low particle agglomeration during

subsequent heat treatment.17 Heating these reagents to 150–

170 uC in an open system for 40 min led to the formation of a

black suspension. The XRD of this material (Fig. 2a) is typical

of a chemically disordered fcc structure possessing broad peaks

at 41, 47, 68 and 82u 2h which are indexed as the (111), (200),

(220) and (311) reflections respectively. The fcc structure of

the particles was also verified by electron diffraction where

d -spacings calculated from radii of pattern rings were

Table 1 Reaction conditions and XRD particle sizes of fcc FePt nanoparticles synthesized by microwave heating

Reaction System Temperature/uCHeatingtime/min

Holdtime/min Solvent Surfactants (ratio used)

Particlesize/nm

1 FePt/open 150 30 10 Nonadecane Oleyl amine–oleic acid (1 : 1)a 2.66(7)b

2 FePt/open 215 10 30 Octyl ether Oleyl amine–oleic acid (1 : 1) 2.19(5)3 FePt/closed 150 + 250 1.5 + 1 2 + 2 Octyl ether Oleyl amine–oleic acid (1 : 1) 3.26( 7)4 FePt/closed 280 5 55 Octyl ether Oleyl amine–oleic acid (1 : 1) 3.16(6)5 FePd/open 282 12 80 Octyl ether Oleic acid (3) 7.20(2)a Molecular mole ratio of surfactants in comparison with molecular mole ratio of main reactants. b Estimated standard uncertainties inparentheses, see text for definition of particle size.


View Online




consistent with XRD results (see electronic supplementary

information (ESI){). Rietveld refinement of the data performed

using Topas Academic27 suggested an average diameter of

particles of 2.66(7) nm and cell parameter of 3.8744(9) A,

similar to fcc FePt nanoparticles synthesised under similar

conditions by normal heating.17 We note that the temperature

at which this reaction was performed is significantly lower

than required using conventional methods (150 uC vs 330 uC).

Further attempts using different conditions of temperature,

time, solvent and vessel (open or closed) system have also led

to production of fcc FePt nanoparticles as shown in Table 1.

FePd nanoparticles could also be prepared by a similar route.

Fig. 2b shows the XRD pattern of y7.2 nm FePd particles

prepared in dioctyl ether with a 1 : 3 molar ratio of metal–oleic

acid surfactant at 282 uC. The refined cell parameter was

3.8972(4) A.

Fig. 3 shows a TEM image of FePt particles produced

by this route. These can be seen to be essentially spherical

and of uniform size. Using image analysis software,28 size

measurement of 188 randomly selected particles shows that the

FePt particles have a narrow size distribution. Fitting with a

log-normal distribution leads to a measured mean diameter of

2.58 nm and a dispersion s of 5% (Fig. 4). The size is consistent

with that indicated by XRD. The HRTEM images (Fig. 3

insert) of individual particles demonstrated that they were

single crystals with lattice fringes consistent with the (200) and

(220) d -spacing of y

1.9 A˚

andy

1.3 A˚

respectively. EDSanalysis of clusters containing y300 particles gave an overall

average Fe : Pt stoichiometry of 52(3) : 48(3). Analysis of

individual particles revealed that a range of compositions are

present with some particles either Fe or Pt rich. Averaging a

large number of individual particles gave a stoichiometry of

48(7) : 52(7). Yu and co-workers have reported that individual

particles prepared by the conventional polyol synthetic route

can have a wide stoichiometry range with a significant

proportion of particles being either Fe- or Pt-rich. In fact,

they report that only 29% of individual particles lie in the

0.4 , x , 0.6 FexPt12x range that would allow the transition

to the L10 phase to occur.18 We note that unlike Yu et al.

we find that 75% of the individual particles lie within thisrange. This suggests that significantly better control over

individual particle stoichiometry is achieved with our

Fe22/Pt2+ synthetic route.

The fcc particles prepared by this route can be converted to

the magnetically important fct L10 phase by annealing under a

Fig. 2 Rietveld fits of X-ray patterns of (a) as-prepared fcc FePt

samples (observed and calculated patterns for FePt with differencebelow) and (b) fcc FePd.

Fig. 3 HRTEM image of monodispersed fcc FePt particles, insert

shows nanoparticles with a fringe spacing consistent with the (200)

plane of fcc structure.

Fig. 4 Particle diameter histogram of fcc FePt nanoparticles; the line

plotted corresponds to the fit using a log-normal distribution.


View Online




flow of 5% H2 in Ar. Fig. 5 shows a narrow 2h range of a series

of diffraction experiments recorded at increasing temperatures;

each pattern was recorded over y1 h with rapid heating

between data collections. Two different phenomena can be

observed from these data. Firstly from a temperature of

around 637 K (364 uC) extra peaks appear at 2h values of y24

and 33u. These peaks can be indexed as the (001) and (110)

reflections of the fct phase (using a pseudo-cubic cell of a y 3.85, c y 3.71 A) and provide direct evidence of the

ordering phase transition. The recorded ordering temperature

is consistent with previous experiments on annealing FePt

nanoparticles synthesized using this synthetic method and

conventional heating, and is significantly lower than the

600 uC required to order previously reported materials.25 It

is also clear from Fig. 5 that peaks sharpen on heating which

is evidence of particle growth.

Quantitative information on both these processes has been

obtained by Rietveld refinement using the Topas Academic

software suite. It is difficult to extract reliable quantitative

information on the early stages of ordering for materials such

as this as much of the information is contained in the relativelybroad superlattice peaks. Due to correlations with the back-

ground (which itself has significant slowly varying contribu-

tions due to the sample mounting and furnace environment for

variable temperature experiments), it is extremely hard to

estimate their intensity correctly. We have therefore adopted a

strategy in which a sixth order background polynomial was

fitted to each of the 48 data sets in an initial round of

Rietveld refinements using a fully disordered model in which

2h ranges corresponding to ordering peaks were excluded.

We believe that this produces the ‘‘least biased’’ estimate of the

background at each temperature that can be achieved. These

background polynomials were then used as fixed functions

in a separate round of Rietveld refinements in which six

parameters (scale factor, a and c cell parameters, overallatomic displacement parameter, particle size and order

parameter) were refined at each temperature. The sample

height was found to vary smoothly with temperature and was

introduced to each refinement as a fixed though temperature-

dependent parameter. To allow refinement through the

fccA fct phase transition a psuedo-cubic cell setting was used

throughout in space group P 4/mmm (Fe at 1a and 1c Wyckoff

sites and Pt at 2e), and the fractional occupancy (frac) of Fe

on Pt sites and Pt on Fe allowed to refine. The order parameter

for the phase transition is thus given by 1–2frac, and ideally

varies from 0 (fcc) to 1.0 (fct).

Fig. 6a shows the temperature dependence of the unit cell

size. Below the fccA

fct ordering temperature individual valuesof a and c are ill-defined by Rietveld refinement (the material is

cubic and peaks are broad so psuedo-cubic tetragonal values

show considerable scatter) so we choose to plot (volume)1/3 on

warming which provides an average measure of cell parameter

over the whole temperature range. A significant reduction in

this parameter is seen from around 450 K. A reduction in

volume is expected for the fccA fct transition and is well

known in, for example, AuCu binary alloys. We note that the

reduction in cell volume occurs before significant ordering

peaks are visible in Fig. 5, and before significant particle

growth. Cell volume is therefore perhaps the most accessible

indication that particle ordering is beginning to occur.

On cooling the material retains the fct structure as expected.The overall cell volume reveals a positive thermal expansion

coefficient throughout, though the c-axis shows a small thermal

contraction (aa 5 +2.0 6 1025, ac 5 25.2 6 1026 K21 from

925–295 K) with the c/a ratio varying from 0.9464(2) at

925 K to 0.9615(3) at 295 K. The c-axis also shows a marked

contraction just above 700 K which is presumably associated

with the Curie temperature which is around 723 K.29 Cell

parameters on cooling of a 5 3.857(7), c 5 3.708(1) A at

296 K compare to literature values of a 5 3.855, c 5 3.71130

or a 5 3.85 and c 5 3.71 A.31 A recent neutron scattering

measurement on large single crystals of bulk FePt shows a

similar temperature dependence of the c cell parameter.32

The order parameter and particle size dependence ontemperature are shown in Figs. 6b and 6c. Below around

450 K the order parameter is approximately constant. Early in

the ordering process precise values of the order parameter are

hard to derive and the low temperature values plotted of y0.2

on warming are probably not significantly different from zero;

the increase in R-factor on forcing the order parameter to be

exactly 0.0 for these refinements is , 0.15% for temperatures

below 558 K. A significant rise in order parameter can be seen

above y500 K, a temperature slightly higher than that at

which the cell volume decrease occurs. On cooling the material

retains its fct structure with the room temperature order

parameter refining to 1.01(3). The indication of perfect Fe/Pt

Fig. 5 25 diffraction patterns recorded on annealing as-prepared fcc

FePt as a function of temperature from 297 to 924 K. Indication of

ordering is clearly visible at 637 K (364 uC) and above. Data collected

on cooling are not shown but derived quantities are included in Fig. 6.


View Online




order provides further support for the stoichiometric nature of

the particles. The mean particle size also increases significantly

from y500 K indicating particle sintering as the protective

surfactants burn off. The particle size remains, as expected,

essentially unchanged on cooling.

Fig. 7a shows the magnetization of sample 1 as a function of

increasing temperature after cooling in a small residual field of

,1 Oe (‘‘zero’’ field cooling) and in a 100 Oe field. Clear

evidence for superparamagnetic behaviour with a blocking

temperature of T y 17 K is seen. The sharp rise of the ZFC

data indicates that the particle size dispersion is low, in support

of the TEM conclusions. The magnetization vs field loops

(Fig. 7b) of the as-synthesized sample showed small coercivity

values at both 10 and 290 K, confirming the superparamag-

netic properties of the particles. The data also indicate the

presence of a very minor component which saturates at low

field. Hysteresis loops of an annealed sample (see Fig. S4 in the

ESI{) gave coercivities of 14.7 and 10.6 kOe at 10 and 290 K

respectively.

We have recently reported the direct preparation of orderedfct FePt nanoparticles using the Fe22/Pt2+ route with conven-

tional heating at 389 uC in tetracosane. This has prompted

us to attempt the preparation of ordered fct particles using

microwave irradiation. A closed microwave system was chosen

to access the high temperatures at which fcc FePt particles

might be transformed directly to the ordered phase in solution.

It proved, however, difficult to control heating rates of such

reactions and pressure build-up in reaction vessels sometimes

led to reaction vessel bursting. Fig. 8 shows a diffraction

pattern of an FePt sample prepared with octyl ether and an

oleyl amine surfactant (2 : 1 surfactant to metal ratio) at 280 uC

under microwave irradiation for 10 min. Superlattice peaks are

Fig. 6 (a) Temperature dependence of the (cell volume)1/3 on

warming and cooling (closed/open triangles respectively) and a (open

circles) and c (open squares) cell parameters on cooling; (b) order

parameter on warming (closed symbols) and cooling (open symbols);

(c) particle size on warming (closed symbols) and cooling (open

symbols). Error bars show ¡1 standard uncertainty as derived by

Rietveld refinement, and are probably an underestimate of the true

uncertainty on parameters. Where error bars are not shown they aresmaller than the size of the plotted symbol.

Fig. 7 (a) ZFC and FC magnetisation curves as function of tem-

perature from 5 to 290 K at a field 100 Oe; (b) Magnetic hysteresisloops measured at 10 and 290 K of as-prepared fcc FePt particles as

prepared in reaction 1 of Table 1.


View Online




clearly present at y24 (001) and 33u (110) 2h indicating FePt

particles in the fct phase have formed. The overlap of sharp

diffraction peaks on broader peaks at y40.5 (111) and 47u

(200) 2h respectively suggests the coexistence of fcc and fct

FePt structural phases. Rietveld refinement confirms the

existence of the ordered FePt structure giving a and c cell

lattice parameters of 3.8463(2) and 3.7214(3) A, an order

parameter of 0.90(1), and an estimated particle size of

24 nm. The cell parameter of the cubic component refines to

3.872(2) A. These results suggest that rapid heating had

simultaneously caused a phase transformation from the fcc to

fct phase and decomposition of surfactants leading to rapidparticle size increase. Hysteresis loops measured at 10 K and

290 K (Fig. S5 in the ESI{) also suggest two phase behaviour

with a kink at low field. The measured coercivities were 7.0 kOe

at 290 K and 9.0 kOe at 10 K.

Conclusions

The general synthetic route presented here provides a

straightforward and stoichiometrically controlled synthesis of

FePt nanoparticles. Using microwaves for reaction heating

shows significant advantages for production of monodispersed

fcc FePt nanoparticle alloys which can be conveniently

converted into the ordered fct phase on annealing at lowtemperature (364 uC). Reactions can be performed very rapidly

(6 minutes or less) and at temperatures lower than using

conventional heating. The Fe22/Pt2+ route allows good

control over both the overall stoichiometry and the stoichio-

metry of individual particles. High temperature reactions

in the microwave led to the direct formation of a mixture

of fcc and fct FePt nanoparticles. The fct nanoparticles

were shown to have a particle size of y24 nm and strong

coercivity indicating ferromagnetic behavior. Further exten-

sions of FePt nanoparticles synthesis by microwave heating

with varying solvent, surfactants and metals have been

investigated.

Acknowledgements

The authors thank Vivian Thompson for TEM images,

Prof Todd Marder and Dr Patrick Steel for access to

microwave facilities EPRSC and ONE-NE, via the Durham

Nanotechnology Innovation Centre and Seagate Technology

for financial support.

References

1 (a) K. J. Klabunde, Nanoscale Materials in Chemistry, John Wiley& Sons, New York, 2001; (b) D. L. Peng, T. Hihara and

K. Sumiyama, J. Magn. Magn. Mater., 2004, 277, 201; (c)V. F. Puntes, K. M. Krisknan and A. P. Alivisatos, Science,2001, 291, 2115.

2 (a) S. Sun, C. B. Murray, D. Weller, L. Folks and A. Moser,Science, 2000, 287, 1989; (b) D. Weller and A. Moser, IEEE Trans.Magn., 1999, 35, 4423.

3 M. Plumer, J. Van Ek and D. Weller, The Physics of Ultra-High-Density Magnetic Recording , Springer, New York, 2001.

4 (a) C. Burda, X. Chen, R. Narayanan and M. A. El-Sayed, Chem.Rev., 2005, 105, 1025; (b) B. L. Cushing, V. L. Kolesnichenko andC. L. O’Connor, Chem. Rev., 2004, 104, 3893.

5 (a) S. Sun, S. Anders, T. Thomson, J. E. E. Baglin, M. F. Toney,H. F. Hamann, C. B. Murray and B. D. Terris, J. Phys. Chem. B ,2003, 107, 5419; (b) T. Iwaki, Y. Kakihara, T. Toda, M. Abdullahand K. Okuyama, J. Appl. Phys., 2003, 94, 6870; (c) B. Jeyadevan,A. Hobo, K. Urakawa, C. N. Chinnasamy, K. Shinoda andK. Tohji, J. Appl. Phys., 2003, 93, 7574; (d ) B. M. Leonard,N. S. P. Bhuvanesh and R. E. Schaak, J. Am. Chem. Soc., 2005,127, 7326.

6 K. Torigoe and K. Esumi, Langmuir, 1993, 9, 1664.7 (a) S. Sun, E. E. Fullerton, D. Weller and C. B. Murray,

IEEE Trans. Magn., 2001, 37, 1239; (b) B. Stahl, N. S. Gajbhiye,G. Wilde, D. Kramer, J. Ellrich, M. Ghafari, H. Hahn, H. Gleiter,J. Weißmuller, R. Wurschum and P. Schlossmacher, Adv. Mater.,2002, 14, 24; (c) Y. Hou, H. Kondoh, T. Kogure and T. Ohta,Chem. Mater., 2004, 16, 5149; (d ) M. Chen and D. E. Nikles,J. Appl. Phys., 2002, 91, 8477.

8 (a) K. Kakizaki, Y. Yamada, Y. Kuboki, H. Suda, K. Shibata andN. Hiratsuka, J. Magn. Magn. Mater., 2004, 272–276, 2200; (b)P. T. L. Minh, N. P. Thuy and N. T. N. Chan, J. Magn. Magn.Mater., 2004, 277, 187; (c) D . L . P en g, T . H ih ara a ndK. Sumiyama, J. Magn. Magn. Mater., 2004, 277, 201.

9 M. T. Reetz and W. Helbig, J. Am. Chem. Soc., 1994, 116, 7401.

Fig. 8 Rietveld fit of X-ray diffraction data of as-prepared FePt samples containing a mixture of fcc and fct particles. (observed and calculated

patterns for FePt, and difference below).


View Online




10 K. S. Suslick, M. Fang and T. Hyeon, J. Am. Chem. Soc., 1996,118, 11960.

11 T. Hyeon, Chem. Commun., 2003, 8, 927.12 (a) T. Thomson, B. D. Terris, M. F. Toney, S. Raoux,

J. E. E. Baglin, S. L. Lee and S. Sun, J. Appl. Phys., 2004, 95,6738; (b) H. Zeng, S. Sun, T. S. Vedantam, J. P. Liu, Z.-R. Dai andZ.-L. Wang, Appl. Phys. Lett., 2002, 80, 2583.

13 (a) S. Wang, S. S. Kang, D. E. Nikles, J. W. Harrell and X. W. Wu,J. Magn. Magn. Mater., 2003, 266, 4 9 ; (b) Y. K. Takahashi,M. Ohnuma and K. Hono, J. Magn. Magn. Mater., 2002, 246,

259; (c) C. L. Platt, K. W. Wierman, E. B. Svedberg, R. VanDe Veerdonk, J. K. Howard, A. G. Roy and D. E. Laughlin,J. Appl. Phys., 2002, 92, 6104.

14 (a) B. Jeyadevan, K. Urakawa, A. Hobo, N. Chinnasamy,K. Shinoda, K. Tohji, D. D. J. Djayaprawira, M. Tsunoda andM. Takahashi, Jpn. J. Appl. Phys., 2003, 42, L350; (b) S. Kang,Z. Jia, S. Shi, D. E. Nikles and J. W. Harrell, J. Appl. Phys.,2005, 97, 10J318; (c) S. Kang, Z. Jia, S. Shi, D. E. Nikles andJ. W. Harrell, Appl. Phys. Lett., 2005, 86, 62503; (d ) M. Takahashi,T. Ogawa, D. Hasegawa and B. Jeyadevan, J. Appl. Phys., 2005,97, 10J307.

15 S. Yamamoto, Y. Morimoto, T. Ono and M. Takano, Appl. Phys.Lett., 2005, 87, 32503.

16 K. Elkins, D. Li, N. Poudyal, V. Nandwana, Z. Jin, K. Chen andJ. P. Liu, J. Phys. D: Appl. Phys., 2005, 38, 2306.

17 L. E. M. Howard, H. L. Nguyen, S. R. Giblin, B. K. Tanner,

I. Terry, A. K. Hughes and J. S. O. Evans, J. Am. Chem. Soc.,2005, 127, 10140.18 A. C. C. Yu, M. Mizuno, Y. Sasaki and H. Kondo, Appl. Phys.

Lett., 2004, 85, 6242.19 (a) R. N. Gedye, W. Rank and K. C. Westaway, Can. J. Chem.,

1991, 69, 706; (b) M. Larhed and A. Hallberg, J. Org. Chem., 1996,

61, 9582; (c) D. R. Baghurst and D. M. P. Mingos, J. Organomet.Chem., 1990, 384, C57.

20 C. -G. Wu and T. Bein, Chem. Commun., 1996, 8, 925.21 (a) A. R. Barron and C. C. Landry, Science, 1993, 260, 1653; (b)

A. G. Whittaker and D. M. P. Mingos, J. Chem. Soc., DaltonTrans., 2002, 3967; (c) A. G. Whittaker and D. M. P. Mingos,J. Chem. Soc., Dalton Trans., 2000, 1521; (d ) A. G. Whittaker,Chem. Mater., 2005, 17, 3426.

22 (a) W. Tu and H. Liu, J. Mater. Chem., 2000, 10, 2207; (b) W. Yu,

W. Tu and H. Liu, Langmuir, 1999, 15, 6.23 K. Patel, S. Kapoor, D. P. Dave and T. Mukherjee, J. Chem. Sci.,2005, 117, 53.

24 F. Bensebaa, N. Patrito, Y. Le. Page, P. L’Ecuyer and D. Wang,J. Mater. Chem., 2004, 14, 3378.

25 R. Harpeness and A. Gedanken, J. Mater. Chem., 2005, 15, 698.26 (a) K. G. Lyon, G. L. Salinger, C. A. Swenson and G. K. White,

J. Appl. Phys., 1977, 48, 865; (b) Y. Okada and Y. Tokumaru,J. Appl. Phys., 1984, 56, 314; (c) D. Taylor, Br. Ceram. Trans. J.,1984, 83, 92.

27 Topas Academic: http://pws.prserv.net/Alan.Coelho/.28 Image Tool: http://ddsdx.uthscsa.edu/dig/itdesc.html/.29 T. S. Vedantam, J. P. Liu, H. Zeng and S. Sun, J. Appl. Phys.,

2003, 93, 7184.30 (a) S. Saita and S. Maenosono, Chem. Mater., 2005, 17, 3705; (b)

T. J. Klemmer, N. Shukla, C. Liu, X. W. Wu, E. B. Svedberg,

O. Mryasov, R. W. Chantrell, D. Weller, M. Tanase andD. E. Laughlin, Appl. Phys. Lett., 2002, 81, 2220.31 Joint Committee for Powder Diffraction Standards, International

Centre for Diffraction Data, Newton Square, PA.32 Y. Tsunoda and H. Kobayashi, J. Magn. Magn. Mater., 2004,

272-276, 776.


View Online


FePd Microwave

Documents