Characteristics of FeCo nano-particles synthesized using plasma
Transcript of Characteristics of FeCo nano-particles synthesized using plasma
INSTITUTE OF PHYSICS PUBLISHING JOURNAL OF PHYSICS D: APPLIED PHYSICS
J. Phys. D: Appl. Phys. 39 (2006) 2212–2219 doi:10.1088/0022-3727/39/10/033
Characteristics of FeCo nano-particlessynthesized using plasma focusT Zhang1, K S Thomas Gan2, P Lee1, R V Ramanujan2
and R S Rawat1
1 National Institute of Education, Nanyang Technological University, Singapore 637616,Singapore2 School of Material Science and Engineering, Nanyang Technological University, Singapore639798, Singapore
E-mail: [email protected]
Received 13 January 2006, in final form 26 February 2006Published 5 May 2006Online at stacks.iop.org/JPhysD/39/2212
AbstractIn our recent investigation, a 3.3 kJ Mather type plasma focus was used tosynthesize FeCo nano-particles. The tops of the anodes normally used werereplaced by the materials to be synthesized using different numbers of focusshots. It is observed that the characteristics of the nano-particles depend notonly on the numbers of focus shots but also on the angular position of themounted sample. SEM results show that the size of the nano-particles issmaller when the sample is mounted at a bigger angular position. The sizeof the particles increases linearly with the increase in the number of focusshots. EDX shows that the FeCo nano-particles are stoichiometric. Themagnetic properties measured using a vibrating sample magnetometeridentify the soft magnetic nature of the FeCo nano-particles. The saturationmagnetization has been found to increase in samples prepared with a highernumber of focus shots using repetitive plasma focus NX2.
1. Introduction
The current curiosity about nano-particles has been phenome-nally increasing because of their interesting electronic, optical,magnetic, mechanical and chemical properties [1]. Interest inthese materials is motivated by the fact that the small grain sizegives the nano-particles unique physical and chemical proper-ties which are totally different from those of their bulk counter-parts. The metallic nano-particles of magnetic materials drawspecial attention owing to their notable uses and applicationsin ultrahigh density data storage, gas sensor, toner materialfor high quality colour copiers and printers, new generationelectric motors and generators, environment friendly refriger-ants and biomedicine [1–3]. Each of the applications requiresthat the magnetic nano-particles have different properties [2].Hence, the synthesis of magnetic nano-particles in a controlledmanner is still a real challenge for their practical usages. Sev-eral chemical and physical methods such as sol–gel, ion im-plantation and sputtering, spray and laser pyrolysis, chemicaland physical vapour deposition and pulsed laser ablation de-position (PLAD) have been employed for synthesis of nano-particles [4–8]. In particular, molecular beam epitaxy, triode
sputtering, ultrasound-assisted electrochemistry, dc magnetronsputtering, laser ablation and e beam were also used to depositmagnetic particles [8–11] and thin films [12–20].
Compared with these methods, plasma focus, as acopious source of energetic ions, has advantages such ashigh deposition rate, energetic deposition process and possibledeposition under reactive background gas pressures. Thedeposition process in dense plasma focus (DPF) is donethrough heating, compressing and ionizing the filling gas toform plasma. The plasma then disintegrates due to plasmainstabilities which generate energetic ions and relativisticelectrons. The energetic electrons along with plasma jet areresponsible for the ablation of the anode material and theablated material is deposited on the substrate. The thin filmdeposition mechanism in a plasma focus device is describedin detail by Soh et al [21]. In our present investigation, asingle shot plasma focus machine was used to synthesize FeComagnetic nano-particles and they were characterized usingdifferent methods including scanning electron microscopy(SEM), energy dispersive x-ray (EDX) spectroscopy, x-raydiffraction (XRD) and vibrating sample magnetometry (VSM).
0022-3727/06/102212+08$30.00 © 2006 IOP Publishing Ltd Printed in the UK 2212
Characteristics of FeCo nano-particles synthesized
Figure 1. (a) Plasma focus setup for material synthesis and (b)specially designed holder with five fingers for material synthesis atdifferent angular positions.
2. Experimental setup
In our investigation, a 3.3 kJ single shot Mather-type DPFdevice was employed, as shown in figure 1, to synthesize FeConano-particles using 5 to 30 focus shots. The top of the anodewas replaced by a FeCo solid tip. A silicon wafer, working asthe substrate, is cut into smaller dimensions of 10 mm × 10 mm× 0.68 mm and then washed by soaking it in acetone, alcoholand de-ionized water, respectively, for durations of 5 min in anultrasonic bath. A specially designed substrate holder was usedto mount silicon substrates at different angular positions withrespect to the anode axis which are named off-centre (18◦) andoutermost (36◦) positions. There are numerous investigationsabout the ion emission characteristics of the plasma focusdevice [22, 23]. The ion flux and energies are reported tochange/decrease with the increase in angle [24] with respectto the anode axis and this results in the different nature of thethin films deposited at different angular positions. We avoidedmaking the deposition along the anode axis (i.e. 0◦) as thethin films deposited along this direction are not very uniformbecause of high flux of high energy ions and also the very highenergy of the ions creates surface defects such as cracks andcraters.
It is well known that a plasma focus device needs someconditioning shots (3–5 in our case) if it is subjected to
atmosphere. Since different samples need a new set of Sisubstrates, the system requires conditioning shots after eachfresh loading of samples. A manual shutter was used, betweenthe anode and the silicon substrates, to avoid deposition duringconditioning shots so that the deposition is done under uniformfocus operation. Once a strong focus peak was observed in thevoltage probe signal (typically within 3–5 focus shots in ourcase) the shutter was removed and a thin film was depositedusing a designated number of deposition shots. The distancebetween the substrate holder and the anode top was fixed at120 mm for the entire set of depositions.
In our investigation, neon is used as the filling gas for FeConanomaterial depositions as, owing to its inert nature, it doesnot react with FeCo and the silicon substrate to give differentcompounds. The plasma focus machine was operated at 14 kVand at neon filling gas pressure of 3 mbar. The interval of twofocus shots is 1 min. A JOEL JSM-6700F SEM with an OxfordEDX attachment, Siemens D5005 XRD, and Lakeshore 7400VSM were used to observe the surface morphology, crystalstructure and magnetic properties of the synthesized nano-particles.
3. Results and discussions
3.1. Topography of FeCo thin film using SEM
Figure 2 shows the topography of FeCo samples depositedat different angular positions using different numbers of focusshots. The FeCo nano-particles synthesis is very much evidentin all the SEM images of figure 2. The variation in FeConano-particle size with the variation in the number of focusshots is plotted in figure 3. It may be noticed that the nano-particle size increases linearly with the increase in the numberof focus shots. For a smaller number of focus shots the textureof the sample surface looks smoother as the nano-particle sizedecreases and they appear to pack more densely. A comparisonof the samples prepared at different angular positions but usingthe same number of focus shots shows that nano-particle sizeat the outermost position is smaller than the nano-particle atthe off-centre position. This is due to the fact that with theincrease in the angular position of the deposited sample, theflux and the energy of the ablated ions is reduced resulting inthe decrease in the size of the nano-particles. Sizes of nano-particles as small as 30 nm are achieved with 5 shots at theoutermost position. The decrease in nano-particle size withincreasing angular position results in smoother surface texturefor depositions at the outermost positions as compared to theone at off-centre positions.
3.2. Quantitative analysis of FeCo thin film using EDX
Figure 4 shows a typical EDX spectrum of a samplesynthesized at the off-centre position for 30 shots. Thequantitative analysis of Fe and Co using a background scanin EDX for the combination of different gas pressures anddeposition shots is plotted in figure 5(a). It must be noted thatfor a similar number of focus shots the sample prepared at theoff-centre position has a higher atomic percentage of Fe andCo as compared to the one prepared at the outermost position.The Fe and Co atomic percentage is also seen to increase withthe increase in the number of shots. These trends are consistent
2213
T Zhang et al
Off-centre
Figure 2. Typical micrographs of FeCo samples synthesized using different number of focus shots at different angular positions (the scalingbar shown is 100 nm).
20
30
40
50
60
70
80
90
0 5 10 15 20 25 30 35Focus shots
Siz
e of
nan
o-pa
rtic
les
(nm
)
Offcentre Outermost
Figure 3. Size of nano-particles of FeCo samples synthesized usingdifferent numbers of focus shots at different angular positions.
with the SEM results and the ion emission characteristics ofthe plasma focus device are responsible for them. The ratio ofatomic percentage of Fe to Co is also plotted in figures 5(b) and(c). It can be seen from the trend that the products obtainedare quite stoichiometric for the different angular positions forthe different number of deposition shots.
3.3. Structural characteristics of FeCo thin film
The XRD spectra for samples prepared at the outermostposition are shown in figure 6. It can be noticed that theproducts deposited with 5, 10 and 20 shots are amorphousin nature as their XRD do not show any diffraction peaks.Crystalline peaks of (220) CoFe2O4 at 29.33◦, (311) Fe3O4 at35.87◦ and (110) FeCo at 44.83◦ are observed for the sampleprepared using 30 focus shots.
2214
Characteristics of FeCo nano-particles synthesized
Figure 4. EDX results for sample synthesized at off-centre position using 30 focus shots.
0
2
4
6
8
10
12
14
0 5 10 15
Co At %
Fe
At
%
0
5
10
15
20
25
30
0 5 10 15 20 25
Fe At %
Co
At
%
Outermost position (b) Off-centre position (c)
(a)
0
5
10
15
20
25
0 5 10 15 20 25 30 35Focus shots
Ato
mic
%
Outermost Fe Outermost Co
Offcentre Fe Offcentre Co
Figure 5. (a) Atomic percentage of Fe and Co for various samples, (b) relative atomic percentage of Fe and Co for samples synthesized atoff-centre positions and (c) relative atomic percentage of Fe and Co for samples synthesized at outermost positions.
The sample prepared with 30 shots shows the presence ofa crystalline structure which may be due to the fact that withthe higher number of shots there was enough energy, more thanthe free energy required for crystallization, transferred to thesample surface by the high energy ions of the filling gas speciesof the plasma focus device. Another reason could be that withthe increase in the number of shots there is more material onthe surface and hence correspondingly there will be more x-rayphotons that could be diffracted. The product obtained with30 shots is partially oxidized. This is due to the fact that theSi substrate has an oxide layer. When neon plasma, initially,hits the substrate, an etching process takes place causing thecleaning of the substrate and ablation of oxygen atoms from
the Si substrate surface. Subsequently, when the ablated Feand Co ions condense on the substrate they would react withthe oxygen ions contributed by the oxide layer ablation fromthe Si substrate and also from the impure oxygen atoms presentin the chamber as the base pressure is not very low, typically10−3 mbar.
XRD spectra at the off-centre position are shown infigure 7. It can be seen that the products synthesized using5 and 10 focus shots are amorphous in nature. However,the samples prepared using 20 and 30 shots are crystalline.The early crystallization in this case, i.e. crystallization at 20deposition shots as compared with 30 shots for the outermostposition sample, is due to the fact that ion energy and flux are
2215
T Zhang et al
0
10
20
30
40
50
60
70
20 25 30 35 40 45 502 θ
Inte
nsi
ty
(220
) C
oFe 2
O4
(311
) Fe
3O4
(110
) Fe
Co
5 focus shots
10 focus shots
20 focus shots
30 focus shots
Figure 6. Typical XRD results for samples synthesized at outermost positions.
0
10
20
30
40
50
60
70
80
90
20 25 30 35 40 45 502 θ
Inte
nsity
(220
) C
oFe 2
O4
(311
) Fe
3O4
(400
) C
oFe 2
O4
(110
) Fe
Co
(316
)Fe
2O3
(024
)Fe
2O3
(109
)Fe
2O3
5 focus shots
10 focus shots
20 focus shots
30 focus shots
Figure 7. Typical XRD results for samples synthesized at off-centre positions.
higher at the lower angular (off-centre) position. Crystallinepeaks of (220) CoFe2O4 at 29.27◦, (311) Fe3O4 at 35.87◦,(109) Fe2O3 at 39.51◦ (400) CoFe2O4 at 43.16◦, (110) FeCoat 44.83◦, (316) Fe2O3 at 47.48◦ and (024) Fe2O3 at 48.59◦
are seen. The high-energy high-fluence ions therefore play animportant role in the growth of as-deposited crystalline FeConano-particles at the off-centre position.
The intensity of the (220) CoFe2O4 peak has been foundto decrease from 20 to 30 shots. This is expected because therest of the peaks present at 30 shots have increased leading tothe compensation for the decrease in intensity of the (220)CoFe2O4 peak. It can also be observed that all the peakspresent at 20 shots have shifted when 30 shots are fired. Thisis probably due to changes in the stress level that the productsexperience when a higher number of shots are used for samplesynthesis. According to the stress formula given by [25]
�d
d= d(observed) − d(JCPDS)
d(JCPDS)
a shift in the peak can be used to calculate the residual stressin the sample. Taking the elastic constants as k, �d/d × k =residual stress.
At 20 shots for the (220) CoFe2O4 peak at 29.27◦, theresidual stress is −0.0259k.
At 30 shots for the (220) CoFe2O4 peak at 29.60◦, theresidual stress is −0.0160k.
Therefore, there is a greater amount of compressiveresidual stress with 20 shots compared to 30 shots. Thecompressive residual stress experienced with 20 shots is about1.62 times that of 30 shots.
3.4. Magnetic field measurement using vibrating samplemagnetometer
A typical VSM result for the off-centre position for 30 shotsdeposition is shown in figure 8. The magnetism for differentshots at the similar angular position is plotted in figure 9.It can be seen that magnetization increases from 5 to 30shots. There is a relationship between the number of shotsand magnetization and a higher number of shots gives greatermagnetization. The outermost position is not taken intoconsideration because the data are not very reliable. The off-centre position does not give good saturation magnetizationvalues of FeCo. However, there is a trend of increasingmagnetization with the increase in the number of shots fired.
2216
Characteristics of FeCo nano-particles synthesized
-0.008
-0.006
-0.004
-0.002
0
0.002
0.004
0.006
0.008
-2000 -1000 0 1000 2000Field strength (Oe)
Mag
neti
zati
on (
emu)
Figure 8. Typical VSM result for 30 shots sample deposited atoff-centre position.
1
2
3
4
5
6
7
8
9
0 10 2 0Focus shots
Mag
neti
sm (
mem
u)
Figure 9. Typical magnetism on samples synthesized at off-centrepositions.
3.5. Magnetism improvement using larger number ofdeposition shots
As mentioned in the previous sub-section, a relatively lowsaturation magnetization is observed in deposited FeCosamples and this saturation magnetization is found to increasewith the increase in deposition shots. Therefore, a repetitiveplasma focus machine designated as NX2 [26] was employedto deposit FeCo samples for a large number of depositionshots as using a single shot machine which operates at 1 minfrequency would take a long time to synthesize a 100 shotsample. The NX2 is a 27.6 µF (1.98 kJ at 12 kV charging)Mather-type DPF device which can be operated at a maximumof 16 Hz repetition rate. In the present investigation, it wasoperated at 3 mbar neon, 12 kV charging and 1 Hz repetitionrate.
At first, the NX2 machine was fired for ten deposition shotsto compare the results with the results obtained in a single shotplasma focus machine. Figure 10 is the topography of FeCosample deposited using ten NX2 shots but in repetitive mode(1 Hz). A uniform FeCo thin film composed of agglomerateswith a size of 100 nm was observed. EDX scanning showsthe Fe and Co atomic percentage to be 7.92% and 7.62%,respectively. Compared with the results for the similar numberof shots using the single shot DPF machine, we can see that
Figure 10. Typical SEM results for FeCo sample prepared using 10focus shots from repetitive NX2 focus machine (the scaling barshown is 100 nm).
Figure 11. Typical SEM results for FeCo sample prepared using100 focus shots from repetitive NX2 focus machine (the scaling barshown is 100 nm).
the atomic percentages of Fe and Co are higher on the sampledeposited using the repetitive NX2 focus device. The XRDshowed that this sample is not crystalline. Next, a much largernumber of deposition shots (100 shots) was used to deposit theFeCo samples using the NX2 focus device operated at 1 Hz.The topography of samples deposited using 100 focus shots isshown in figure 11. It shows that the FeCo thin film is madeup of agglomerates with sizes several hundred nanometres andthese agglomerates are composed of small grains having thesame size as the low number of shots. Figure 12 is the EDXresult for 100 deposition shots using the repetitive plasma focusdevice. From it we can see that deposited FeCo thin film isso thick that Si Kα emission from the surface of samples islow and the atomic percentages of Fe and Co are quite high.The FeCo sample is shown to be the crystalline in nature usingXRD, as shown in figure 13. The significant peaks shown infigure 13 correspond to (110) FeCo, (200) FeCo, and (210)FeCo. A saturation magnetization of 76 memu is measuredon 100 shot depositions, as shown in figure 14, indicatingthat the samples deposited using 100 deposition shots are of
2217
T Zhang et al
Figure 12. Typical EDX results for FeCo sample prepared using 100 focus shots from repetitive NX2 focus machine.
0
500
1000
1500
20 40 60 80 1002 θ
Inte
nsity
(a.
u.)
(110) FeCo
(200) FeCo (210) FeCo
Figure 13. Typical XRD results for FeCo sample prepared using100 focus shots from repetitive NX2 focus machine.
-0.03
-0.02
-0.01
0
0.01
0.02
0.03
-20000 -10000 0 10000 20000
Field strength (Oe)
Mag
netiz
atio
n (e
mu)
Figure 14. Typical VSM results for FeCo sample prepared using100 focus shots from repetitive NX2 focus machine.
intermediate magnetic strength and can be used for data storageapplications. A more systematic study of FeCo nano-phasematerial synthesis using a larger number of deposition shotsusing the NX2 repetitive plasma focus is planned and will bereported later.
4. Conclusions
FeCo nano-particles have been successfully synthesized usinga 3.3 kJ Mather type plasma focus machine. FeCo nano-particles with sizes ranging from 30 to 80 nm have been
synthesized using different numbers of focus shots at differentangular positions. Nano-particles of smaller size with bettersurface uniformity have been synthesized at the outermostposition, i.e. at a bigger angular position. This size dependenceis a direct consequence of the ion emission characteristics of theplasma focus device. The increase in the number of depositionshots increases the size of nano-particles. EDX spectroscopyshowed that the FeCo samples were stoichiometric. XRDscanning shows the crystal characteristics changing with thevariations in the positions and deposition shots. The saturationmagnetization was found to increase significantly with theincrease in the number of focus deposition shots to 100. Thisopens up the scope for an interesting systematic investigationof nano-phase magnetic material synthesis using a much largernumber of focus shots in a repetitive plasma focus device.
Acknowledgments
The authors are thankful to the National Institute of Education,Singapore, for providing the AcRF Grant RI 17/03/RSR. Oneof the authors, TZ, would like to thank NIE/NTU for providinga research scholarship. The authors are also thankful to thereferees for their valuable suggestions.
References
[1] Nalwa H S (ed) 1999 Hand Book of Nanostructured Materialsand Nanotechnology (San Diego: Academic Press)
[2] Tartaj P, Morales M P, Veintenillas-Verdagure S,Gonzalez-Carreno T and Serna C J 2003 J. Phys. D 36 R182
[3] Ouellette J 1997 Industrial Phys. 3 12–15[4] Brinker C J and Scherer G W 1990 The Physics and Chemistry
of Sol Gel Processing (Boston: Academic Press)[5] Swihart M T 2003 Curr. Opin. Colloid. Interface Sci. 8 127–33[6] Chen J P, Lee K M, Sorensen C M, Klabunde K J and
Hadjipanayis G C 1994 J. Appl. Phys. 75 5876–8[7] Daroczi L, Beke D L, Posgay G and Kis-Vargal M 1995
Nanostruct. Mater. 6 981–4[8] Zbroniec L, Sasaki T and Koshizaki N 2001 Scr. Mater. 44
1869–72[9] Li Q, Sasaki T and Koshizaki N 1999 Appl. Phys. A 69 115–8
[10] Gonzalez J M, Montero M I, Vazquez L, Martin Gago J A,Givord D, Julian C and O’Grady K 1999 J. Magn. Magn.Mater. 196–197 96–8
[11] Sasaki T, Terauchi S, Koshizaki N and Umerhara H 1998 Appl.Surf. Sci. 127–129 398–402
2218
Characteristics of FeCo nano-particles synthesized
[12] Swerts J, Temst K, Vandamme N, Opperdoes B, VanHaesendonck C and Bruynseraede Y 2002 Thin Solid Films413 212–15
[13] Cho S J, Kang H Y, Lee C H, Kim Y J and Krist T 2004 Nucl.Instrum. Methods Phys. Res. A 529 94–7
[14] Farnan G A, Fu C L, Gai Z, Krcmar M, Baddrof A P, Zhang Zand Shen J 2003 Phys. Rev. Lett. 91 226106-1-4
[15] Dumm M, Uhl B, Zolfl M, Kipferl W and Bayreuther G 2002J. Appl. Phys. 91 8763–5
[16] Mancier V, Delplancke J, Delwiche J, Franskin M J H, PiquerC, Rebbouh L and Grandjean F 2004 J. Magn. Magn.Mater. 281 27–35
[17] Yu W, Bain J A, Peng Y and Laughlin D E 2002 IEEE Trans.Magn. 38 3030–2
[18] Jung H S, Doyle W D, Wittig J E, Al-Sharab J F and Bentley J2002 Appl. Phys. Lett. 81 2415–7
[19] Bowen M et al 2001 Appl. Phys. Lett. 79 1655–7
[20] Gupta R, Muller G A, Schaaf P, Zhang K and Lieb K P 2004Nucl. Instrum. Methods Phys. Res. B 216 350–4
[21] Soh L Y, Lee P, Shuyan X, Lee S and Rawat R S 2004 IEEETrans. Plasma Sci. 32 448–55
[22] Mohanty S R, Bhuyan H, Neog N K, Rout R K and Hotta E2005 J. Appl. Phys. 44 5199–205
[23] Kelly H, Lepone A, Marquez A, Sadowski M J, Baranowski Jand Skladnik-Sadowska E 1998 IEEE Trans. Plasma Sci. 26113–17
[24] Bertalot L, Herold H, Jager U, Mozer A, Oppenlander T,Sadowski M and Schmidt H 1980 Phys. Lett. A 79 389–92
[25] Rawat R S, Arun P, Videshwar A G, Lee P and Lee S 2004 J.Appl. Phys. 95 7725–30
[26] Zhang T, Mohanty S R, Hassan S M, Patran A, Springham S V,Tan T L, Lee P, Lee S and Rawat R S 2005 DenseZ-Pinches: 6th Int. Conf. on Dense Z-Pinches (Oxford, UK)ed J Chittenden, pp 231–4
2219