Top Banner
IOP PUBLISHING JOURNAL OF PHYSICS D: APPLIED PHYSICS J. Phys. D: Appl. Phys. 41 (2008) 065001 (7pp) doi:10.1088/0022-3727/41/6/065001 Microstructure, magnetic and M¨ ossbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets N V Rama Rao 1 , P Saravanan 1 , R Gopalan 1 , M Manivel Raja 1 , D V Sreedhara Rao 1 , D Sivaprahasam 2 , R Ranganathan 3 and V Chandrasekaran 1 1 Defence Metallurgical Research Laboratory, Hyderabad-500 058, India 2 International Advanced Research Centre for Powder Metallurgy and New Materials Hyderabad-500 005, India 3 Saha Institute of Nuclear Physics, Kolkata-700 064, India E-mail: rg [email protected] Received 6 November 2007, in final form 24 January 2008 Published 26 February 2008 Online at stacks.iop.org/JPhysD/41/065001 Abstract Nanocomposite powders comprising Sm–Co–Fe intermetallic phases and Fe(Co) were synthesized by high-energy ball milling and were consolidated into bulk magnets by the spark-plasma sintering (SPS) technique. While the microstructure of the SPS samples was characterized by transmission electron microscopy (TEM), the solubility of Fe in different phases was investigated using M¨ ossbauer spectroscopy. TEM studies revealed that the spark-plasma sintered sample has Sm(Co,Fe) 5 as a major phase with Sm 2 (Co,Fe) 17 , Sm(Co,Fe) 2 and Fe(Co) as secondary phases. The size of the nanocrystalline grains of all these phases was found to be in the range 50–100 nm. The M ¨ ossbauer spectra of the as-milled powders exhibited two different subspectra: a sextet corresponding to the Fe phase and a broad sextet associated with the Fe(Co) phase; while that of the SPS sample showed four different subspectra: a sextet corresponding to Fe and other three sextets corresponding to the Fe(Co), Sm(Co,Fe) 5 and Sm 2 (Co,Fe) 17 phases; these results are in accordance with the TEM observation. Recoil magnetization and reversible susceptibility measurements revealed magnetically single phase behaviour of the SPS magnets. 1. Introduction Among the rare earth permanent magnets, SmCo 5 intermetallic compound has been considered as technologically important, as it exhibits a good combination of magnetic properties such as high magnetic crystalline anisotropy (H A ), high Curie temperature (T C ) and high-energy product [(BH) max ][1]. With the advent of nanomagnetism, the SmCo 5 hard magnet is being revived with a scope to enhance some of the magnetic properties through the development of nanocomposite material—comprising SmCo 5 as the hard phase and α-Fe(Co) as the soft phase. These nanocomposite magnets are expected to have both high coercivity and large saturation magnetization, which are achieved through the exchange- coupling interactions between the soft and hard magnetic grains [24]. The effectiveness of the exchange coupling determines the magnetic properties of the nanocomposite magnets, which in turn depend on microstructural parameters such as grain size, phase distribution and volume fraction of the hard and soft phases. Though, several methods have been developed for processing exchange coupled nanocomposite magnets; in practice, difficulty was encountered during sintering of nanocomposite powders, due to the excessive grain growth—which hampers the advantages of getting nanoscale microstructure and useful magnetic properties. In order to realize the potential of nanocomposite magnets in terms of 0022-3727/08/065001+07$30.00 1 © 2008 IOP Publishing Ltd Printed in the UK
7

Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

Mar 12, 2023

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

IOP PUBLISHING JOURNAL OF PHYSICS D: APPLIED PHYSICS

J. Phys. D: Appl. Phys. 41 (2008) 065001 (7pp) doi:10.1088/0022-3727/41/6/065001

Microstructure, magnetic and Mossbauerstudies on spark-plasma sinteredSm–Co–Fe/Fe(Co) nanocompositemagnetsN V Rama Rao1, P Saravanan1, R Gopalan1, M Manivel Raja1,D V Sreedhara Rao1, D Sivaprahasam2, R Ranganathan3 andV Chandrasekaran1

1 Defence Metallurgical Research Laboratory, Hyderabad-500 058, India2 International Advanced Research Centre for Powder Metallurgy and New MaterialsHyderabad-500 005, India3 Saha Institute of Nuclear Physics, Kolkata-700 064, India

E-mail: rg [email protected]

Received 6 November 2007, in final form 24 January 2008Published 26 February 2008Online at stacks.iop.org/JPhysD/41/065001

AbstractNanocomposite powders comprising Sm–Co–Fe intermetallic phases and Fe(Co) weresynthesized by high-energy ball milling and were consolidated into bulk magnets by thespark-plasma sintering (SPS) technique. While the microstructure of the SPS samples wascharacterized by transmission electron microscopy (TEM), the solubility of Fe in differentphases was investigated using Mossbauer spectroscopy. TEM studies revealed that thespark-plasma sintered sample has Sm(Co,Fe)5 as a major phase with Sm2(Co,Fe)17,Sm(Co,Fe)2 and Fe(Co) as secondary phases. The size of the nanocrystalline grains of allthese phases was found to be in the range 50–100 nm. The Mossbauer spectra of the as-milledpowders exhibited two different subspectra: a sextet corresponding to the Fe phase and a broadsextet associated with the Fe(Co) phase; while that of the SPS sample showed four differentsubspectra: a sextet corresponding to Fe and other three sextets corresponding to the Fe(Co),Sm(Co,Fe)5 and Sm2(Co,Fe)17 phases; these results are in accordance with the TEMobservation. Recoil magnetization and reversible susceptibility measurements revealedmagnetically single phase behaviour of the SPS magnets.

1. Introduction

Among the rare earth permanent magnets, SmCo5 intermetalliccompound has been considered as technologically important,as it exhibits a good combination of magnetic properties suchas high magnetic crystalline anisotropy (HA), high Curietemperature (TC) and high-energy product [(BH)max] [1].With the advent of nanomagnetism, the SmCo5 hard magnetis being revived with a scope to enhance some of the magneticproperties through the development of nanocompositematerial—comprising SmCo5 as the hard phase and α-Fe(Co)as the soft phase. These nanocomposite magnets areexpected to have both high coercivity and large saturation

magnetization, which are achieved through the exchange-coupling interactions between the soft and hard magneticgrains [2–4]. The effectiveness of the exchange couplingdetermines the magnetic properties of the nanocompositemagnets, which in turn depend on microstructural parameterssuch as grain size, phase distribution and volume fraction ofthe hard and soft phases. Though, several methods have beendeveloped for processing exchange coupled nanocompositemagnets; in practice, difficulty was encountered duringsintering of nanocomposite powders, due to the excessive graingrowth—which hampers the advantages of getting nanoscalemicrostructure and useful magnetic properties. In order torealize the potential of nanocomposite magnets in terms of

0022-3727/08/065001+07$30.00 1 © 2008 IOP Publishing Ltd Printed in the UK

Page 2: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

higher energy products, it is therefore essential to have an idealhomogeneous nanostructure with enhanced hard–soft phasecoupling with uniform distribution of the phases. Despite thefact that such microstructures with enhanced energy productshave been obtained in thin film multi-layers [5], sintered bulknanocomposite magnets are yet to give remarkable properties.

Spark-plasma sintering (SPS) is known as one of therapid sintering techniques with a capability of achieving fastdensification and minimal grain growth in a short sinteringtime [6]. Although, the technological realization of theSPS technique in the processing of Nd–Fe–B nanocompositemagnets has been successfully demonstrated by severalresearchers [7–9], its potential has not been utilized inthe consolidation of SmCo5/Fe nanocomposite powders.Recently, we have reported the consolidation of mechanicallymilled SmCo5/Fe nano powders by the SPS technique as afeasible method for fabrication of bulk nanocomposite magnets[10]. Since mechanical milling/alloying of SmCo5 and Fepowders will lead to the formation of various Sm–Co–Feintermetallic as well as Fe(Co) phases, it is important toinvestigate the microstructure of these phases in detail andits effect on magnetic properties. Further, the solubility ofFe in these phases is also important to correlate with themagnetic properties. Hence, this study aims to bring outsuch a detailed microstructural investigation complementedwith Mossbauer characterization and the associated magneticproperties in mechanically milled and spark-plasma sinteredSm–Co–Fe/Fe(Co) nanocomposite magnets.

2. Experimental details

Nanocomposite powders of Sm–Co–Fe/Fe(Co) were preparedby milling SmCo5 alloy powder with 5 and 10 wt% ofα-Fe powder using a high-energy ball mill, followed byconsolidation of the nanopowders by SPS. The Ball millingwas carried out using tungsten carbide balls and vial with ballto powder ratio of 10 : 1. To avoid contamination, powderhandling was conducted in toluene medium. Milling wasconducted at 200 rpm for different milling times (4–50 h),with intermediate stopping for every 4 h. Before sintering,the milled powder (10 h) was packed in a graphite die of15 mm inner diameter and then it was placed in an axialmagnetic field press of 2 T static aligning field to orientthe powders. The aligned powders were subjected to SPS(Sumitomo Coal Mining Company) in a vacuum of <10−3 Pawith an applied load of 10.5 kN (∼70 MPa) for 5 min in atemperature range 700–740 ◦C. A thermocouple placed inthe middle part of the graphite die was used to control thetemperature during the SPS treatment. Prior to the structuraland property characterization, the end-products were polishedwith SiC paper to remove surface contamination from thegraphite die and foil. The phase analysis of the sampleswas carried out using a Philips x-ray diffractometer (XRD)with Cu Kα radiation. Microstructural studies were conductedusing a scanning electron microscopy (SEM, Leo 440i) andan analytical transmission electron microscopy (TEM, Tecnai20T G2). The Mossbauer spectra were recorded using aFAST Comtec (Germany) spectrometer at room temperature

Figure 1. SEM micrographs of powders milled for (a) 10 h and(b) 50 h.

in transmission geometry with a 25 mCi 57Co source inrhodium matrix. Prior to the experiments, the spectrometerwas calibrated using a standard α-Fe foil. The Mossbauerspectra of milled powder and spark-plasma sintered sampleswere recorded for a duration of about 5 days with totalbackground counts up to 8–10 × 105. The Mossbauer spectrathus obtained were fitted with the PCMOS-II programme [11].The magnetic properties of the SPS magnets were evaluatedusing a superconducting quantum interference device up to amaximum field of 7 T in the temperature range 5–300 K. Formagnetic measurements above 300 K we have used a vibratingsample magnetometer with a maximum applied magneticfield of 2 T.

3. Results and discussions

3.1. Microstructural investigation

Figure 1 shows the SEM micrographs of the powders milledfor various milling times (10 and 50 h). As the millingtime increases the particle size becomes submicrometres insize (∼600–900 nm) and also the particles are found to beslightly agglomerated. The repeated breaking and weldingof the particles lead to a change in the particle size andmorphology. When milled for 50 h, the nanosized particles arewelded together to form agglomerates as visible from the SEMmicrograph (figure 1(b)). From the XRD analysis (not shown),

2

Page 3: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

Figure 2. Bright field TEM micrograph of the spark-plasmasintered 10 wt% Fe-containing SmCo5 sample.

the average nanocrystallite size was estimated to be around30–60 nm. The XRD and SEM analysis were only taken asa preliminary investigation and a detailed phase analysis wascarried out using TEM in SPS consolidated samples. Since,both 5 and 10 wt% Fe-containing Sm–Co magnets revealedsimilar microstructural features, we report in the followingdiscussion more in detail the TEM results of the 10 wt%Fe-containing nanocomposite magnet.

The bright field TEM image obtained from SPS sampleof the 10 wt% Fe-containing SmCo5 is shown in figure 2.From TEM and in combination with energy dispersivex-ray spectroscopy analysis (EDX), the phases, namely;Sm(Co,Fe)5, Sm2(Co,Fe)17 and Fe(Co) phases have beenidentified. In addition to these phases, the EDX analysisshowed a very few grains corresponding to the Sm(Co,Fe)2

phase. The average grain size of all these phases was foundto be in the range 50–100 nm, while it was ∼30–60 nm inmilled powders suggesting that no significant grain growth hasoccurred during SPS.

Schneider et al [12] have reported the phase equilibriaat 800 ◦C for the ternary Sm–Co–Fe system. In theirinvestigations, complete miscibility of Co and Fe was foundfor the three phases, namely Sm2(Co,Fe)17, Sm(Co,Fe)3 andSm(Co,Fe)2, whereas only limited solubility exists in SmCo5

at high temperature. However, in the present investigation,the TEM results have revealed the presence of the Sm(Co,Fe)5

phase with about 13 at% of Fe going into the SmCo5 lattice.In mechanically milled and spark-plasma sintered sample,the interdiffusion of Fe and Co could take place leading toformation of the solid solutions such as Sm(Co,Fe)5 and Fe(Co)and in this process it is possible to have extended solubilityof Fe in the SmCo5 phase. In contrast to the existence ofthe Sm(Co,Fe)3 phase at high temperatures—as explained bythe Sm–Co–Fe phase equilibria, the phase does not form inmechanically milled and spark-plasma sintered magnet. Wehave also observed a very few grains (<2%) correspondingto Sm(Co,Fe)2. The TEM investigation (figure 3(a)), alsorevealed that some of the grains have high density of planardefects. A similar type of microstructural defect has also beenreported in a few grains of the melt-spun ribbons of PrCo5

and also in hot-pressed Sm11.9Co59.5Fe23.6Cu5 [13, 14]. The

Figure 3. TEM micrographs of SPS 10 wt% Fe-containing SmCo5

sample: (a) bright field image of a grain having faults and (b) darkfield image of Fe(Co) grains. The insets show the SAED patternsobtained from the corresponding phases. The faulted grain wasidentified as Sm2(Co,Fe)17 phase.

selected area electron diffraction (SAED) pattern obtainedfrom the grains containing these faults corresponds to the[1 1 2 1] zone of axis of the Sm2(Co,Fe)17 rhombohedralstructure. However, it should be mentioned that not all thegrains reveal such microstructural features and most of thegrains remain as the Sm(Co,Fe)5 phase. Yermolenko et al[15] have discussed that the anisotropy of the Co-sublattice inSmCo5 which is equal to the anisotropy of YCo5, will decreaseon substitution of transition elements in Sm(Co1−xMx)5 whereM = Fe, Ni and Cu. Hence, in order to have relatively highcoercivity in mechanically milled and spark-plasma sinteredSm-Co/Fe nano magnets, the amount of Fe that can be takeninto solid solution should be controlled through optimizationof milling conditions and starting precursor Fe powder whichneeds a detailed investigation. As discussed earlier, the high-energy ball milling of SmCo5 alloy and Fe powders has alsoresulted in the formation of the Fe(Co) phase (figure 3(b))through interdiffusion of the elements. The EDX analysisrevealed the composition of the Fe(Co) phase as Fe55Co45 andits typical SAED pattern (inset of figure 3(b)) represents the[0 1 1] zone axis of the bcc Fe(Co) phase.

3

Page 4: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

Figure 4. Energy filtered TEM images of SPS 10 wt%Fe-containing SmCo5 sample showing elemental maps for Sm, Coand Fe. (a) Regions showing EELS maps comprising Sm(Co,Fe)5,Fe and Fe(Co) phases and (b) EELS maps corresponding to theSm2(Co,Fe)17 and Sm2O3 phases.

The elemental mapping in the SPS sample was obtainedusing the electron energy loss spectroscopy (EELS) with GIF200. In order to confirm the phases that have been observedthrough the TEM microstructural analysis, we have carried outthe EELS mapping at different regions of the TEM sample.The EELS maps of Sm, Fe and Co obtained at two typicalregions are shown in figures 4 (a) and (b). The elementalmaps show the existence of a clear contrast for Sm, Co and Fe.In figure 4(a), it can be seen that the grains that are enrichedwith Sm and Co have traces of Fe and they correspond to theSm(Co,Fe)5 phase. It can also be observed from figure 4(a)that some of the regions reveal contrast of Fe alone—indicatingthe presence of unreacted α-Fe particle. We also observed insome regions (figure 4(b)), the grains with faulted featureswhich are rich in Sm and Co with less contrast of Fe. Thesegrains in complement with the EDX results are identified as theSm2(Co,Fe)17 phase. In some regions, it can be seen clearlythat Fe and Co are together enriched indicating the presenceof the α-Fe(Co) phase. The existence of Sm2O3 is alsoevident from the contrast of Sm alone regions in figure 4(b).The results of TEM together with the EELS analysis were

Figure 5. The Mossbauer spectra of 10 wt% Fe-containing SmCo5

sample (a) milled for 10 h and (b) after SPS.

further corroborated with the Mossbauer spectroscopy studiesto understand more in detail the formation of the phases in SPSsamples.

3.2. Mossbauer spectroscopy analysis

The 57Fe Mossbauer experiments were carried out on 10 hmilled SmCo5 + 10 wt% Fe powder as well in SPS sample andtheir representative Mossbauer spectra are shown in figure 5.The Mossbauer spectra obtained from the milled powder canbe fitted well with two subspectra: a sextet corresponding tothe α-Fe phase with a hyperfine magnetic field (Hhf ) of 33.2 Tand a broad sextet associated to α-Fe(Co) having Hhf of 35.3 T.It has been reported that the presence of one Co atom in thevicinity of the Fe atoms can increase the hyperfine magneticfield by about 0.8 T [16]. Although, there is a possibility forformation of the Sm(Co,Fe)5 phase owing to diffusion of Fein SmCo5 during milling, the Mossbauer spectra of the milledpowder could not reveal the corresponding Hhf , which will bein the range 26–33 T [17]. However in the case of the SPSsample, the TEM studies clearly showed the presence of theSm(Co,Fe)5 phase which forms due to the interdiffusion ofFe with the SmCo5 phase driven by the SPS temperature andpressure.

4

Page 5: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

Table 1. The Mossbauer parameters obtained for 10 wt% Fe containing SmCo5 sample before and after SPS.

Hhf IS QS Intensity AssignedSample Sub-spectra (T) (mm s−1) (mm s−1) (%) phase(s)

As-milled Powder S1 33.2 −0.109 0.020 83 FeS2 35.3 −0.026 0.038 17 Fe(Co)

SPS S1 33.1 −0.101 0.015 8 FeS2 32.2 −0.014 0.036 55 Fe(Co)S3 27.5 −0.123 −0.480 15 Sm2(Co, Fe)17

S4 25.6 −0.160 0.050 22 Sm(Co, Fe)5

Note: Hhf —hyperfine field; IS—Isomer shift; QS—quadrupole splitting.

The Mossbauer spectra of the SPS sample have exhibitedfour sextets having hyper fine field (Hhf ) values of 33.1,32.2, 27.5 and 25.6 T. The Hhf value corresponding to 33.1 T,significantly implies the presence of some amount of unalloyedFe in the SPS sample, which is in accordance with the resultsof the EELS mapping. The sextet with the Hhf 32.2 T hasbeen assigned to the Fe(Co) phase accounting to the increasingproportion of Co content in the α-Fe phase. It should berecalled that the Hhf increases to a maximum when Fe alloyswith 30 at% of Co and it decreases on further increase of Cocontent, as similar to the variation of saturation magnetizationobserved in phase diagram of Fe and Co alloys [18]. Thus, theMossbauer results of the SPS sample complement the TEM-EDX observations of the phases. The Hhf values of 27.5 and25.6 T in the SPS samples are lower than that of pure α-Fe(33 T) and are comparable to those of the Sm(Co,Fe) phases iniron–rare earth intermetallics [17]. Of these two Hhf values,the lower one can be assigned to the Sm(Co,Fe)5 phase andthe higher to the Sm2 (Co,Fe)17 phase [19]. The relativeintensities of the magnetic components (proportional to theFe content) obtained for the milled powder as well as for theSPS magnet are listed in table 1 and these results are obtainedby assuming that the recoil-less factor (f ) is equal for all thephases and sites at room temperature. The relative intensity ofthe Fe(Co) phase for the milled powder was 17%, while afterSPS it increased to 55%—which suggests that the formation ofFe(Co) is highly favoured during SPS. We could not detect theSm(Co,Fe)2 phase through the Mossbauer spectra, probablydue to presence of low Fe content (<2%).

3.3. Initial magnetization and recoil demagnetization curves

Figure 6 shows hysteresis loops of the spark-plasma sinteredSmCo5 and Fe (5 and 10 wt%) added samples measured upto a field of 7 T at 300 K. The magnetic properties such asremanence (Mr), saturation magnetization (Ms), coercivity(Hc) and energy product (BH)max are evaluated from figure 6and are displayed as inset in figure 6. The energy product ofthe SPS magnets is found to be ∼7 MG Oe. The low energyproduct obtained in the SPS magnets can be attributed to (i)random assembly of the nanocrystallites of the Sm(Co,Fe)5

phase and (ii) the presence of the other Sm–Co–Fe secondaryphases. It can be seen that the initial magnetization curvesfor all these samples are not saturated even at a field of 7 T. Itis evident from figure 6 that with increase in Fe content, themagnetization increases. The increase in magnetization maybe caused by the formation of the Sm(Co,Fe)5 phase which

Figure 6. Hysteresis curves of the spark-plasma sintered (a) SmCo5

(b) 5 wt% and (c) 10 wt% Fe-containing SmCo5 samples. The insetshows a typical spark-plasma sintered nanocomposite magnet.

has enhanced saturation magnetization than that of the pureSmCo5 phase. Also it is possible that, α-Fe(Co)—a new high-magnetization phase that has formed through interdiffusionof addition of the elements can increase the magnetization.In order to investigate the exchange coupling between theintermetallic Sm(Co,Fe)5 (major phase) and α-Fe(Co), wehave calculated the isotropic remanence ratio (Mr/Ms) andit was found to be >0.5 for Fe-containing the SmCo5 samples.Usually, the value of Mr/Ms > 0.5 is considered as anevidence for spring-exchange-coupled grains in an isotropicnanocomposite magnet. A similar observation has alsobeen reported in the mechanically alloyed SmCo5 + α-Fenanocomposite powders [20,21]. However, it should be notedthat the ratio of Mr/Ms > 0.5 alone cannot be consideredas a universal feature for the exchange spring magnet; indeedit has to be discussed in conjunction with a large reversibledemagnetization curve as a criterion for the presence of theexchange spring mechanism [4]. Hence, to gain furtherinsight into the exchange-coupling behaviour in Fe (5 and10 wt%) containing the SPS samples, we measured the recoilcurves (figure 7) for a number of set maximum fields on thedemagnetization curves of a previously magnetized samplesat 300 K. From figure 7, it can be observed that both 5 and10 wt% Fe-containing samples exhibit reversible recoil curves

5

Page 6: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

for demagnetization fields lower than the coercivity. Thechange in reversible magnetization (Mr) is found to be verysmall up to a field of 0.5 T and thereafter it falls rapidly dueto irreversible rotation at high fields. It can be seen fromfigure 7 that the recoil curves in the SPS samples are closedand are similar to that of a single phase magnet [22]. Thus,it indicates a completely reversible behaviour along the innerhysteresis loop. The analysis of the recoil curves in termsof the recoil susceptibility (χrev = dmrev/dH , where mrev isthe reversible magnetization) can further provide informationabout the critical fields for reversible magnetization. Thereversible component of the recoil susceptibility χrev versusreverse field plots for 5 and 10 wt% Fe-containing samples areis shown in figure 8. From these plots, it can be inferred thatwhen the reversal field is less than the critical field (peak ofthe curve), the recoil susceptibility increases with the reversalfield and reaches a maximum and decreases thereafter withfield. The maximum in the recoil susceptibility indicates theend of coherence of magnetization reversal between the phasesin nanocomposite magnet which behaves like a single phasemagnet. Both samples have maximum recoil susceptibilitynear to the coercivity. The recoil susceptibility of the 10 wt%Fe-containing sample is larger than the 5 wt% Fe sample. Therelatively large reversible magnetization in low fields (<0.5 T)in conjunction with isotropic remanence ratio (Mr/Ms) >0.5in these samples indicate the presence of exchange couplingbetween the different magnetic phases that makes them tobehave like a single phase magnet.

We have also measured the recoil curves at 5 K for both5 and 10 wt% Fe-containing SmCo5 samples (results notshown). It is observed that the reversibility of the recoildemagnetization curves exists up to a field of 1 T for boththe samples which is much higher than that observed at300 K. This could be attributed to the enhancement of themagnetic anisotropy of the hard phase at low temperatureswhich results in a significant reduction of the exchange length[22]. However, the competition between the increase in themagnetic anisotropy and the decrease in the exchange lengthon exchange coupling becomes more complex to explain in ournanocomposite magnets due to the presence of the multiphasessuch as SmCo5, Sm(Co,Fe)5, Sm2(Co,Fe)17 and Sm(Co,Fe)2.Since the magnetic anisotropy is high at 5 K, correspondinglythe coercivity of the samples has also increased with Hc

of 1.75 and 1.4 T for the 5 and 10 wt% Fe-containingSmCo5 magnets, respectively as compared with the valuesat 300 K. It is to be mentioned here that the high-energyball milling of the SmCo5 alloy powder with Fe, results inmultiphase formation and hence the associated microstructuralfeatures and their influence on coercivity are still complex tounderstand in the high-energy ball milled and spark-plasmasintered magnets.

4. Conclusions

In summary, the microstructure of the 5 and 10 wt%Fe-containing SPS samples revealed Sm(Co,Fe)5 as the majorphase with the presence of the secondary phases such asSm2(Co,Fe)17, Sm(Co,Fe)2 and Fe(Co). The average grain

Figure 7. Recoil demagnetization curves measured at 300 K forspark-plasma sintered (a) 5 wt% and (b) 10 wt% Fe-containingSmCo5 samples.

Figure 8. Recoil susceptibility curves measured at 300 K forspark-plasma sintered (a) 5 wt% and (b) 10 wt% Fe-containingSmCo5 samples.

size of these phases was found to be in the range 50–100 nm. The Mossbauer analysis also confirmed the formationof the Sm(Co,Fe)5, Sm2(Co,Fe)17 and Fe(Co) phases asevidenced from the TEM results. The recoil curves and recoilsusceptibility plots at 5 and 300 K exhibited magneticallysingle phase behaviour of the nanocomposite magnet.

6

Page 7: Microstructure, magnetic and Mössbauer studies on spark-plasma sintered Sm–Co–Fe/Fe(Co) nanocomposite magnets

J. Phys. D: Appl. Phys. 41 (2008) 065001 N V Rama Rao et al

Acknowledgments

The authors are grateful to Defence Research and DevelopmentOrganization, Government of India for the financial support tocarry out this work. The keen interest shown by the Director,DMRL and the Director, SINP in this work is gratefullyacknowledged. The authors would also like to thank theDirector, ARCI for providing SPS experimental facilities tocarry out this work.

References

[1] Kaplesh K 1988 J. Appl. Phys. 63 R13[2] Skomski R and Coey J M D 1993 Phys. Rev. B 48 15812[3] Fischer R and Kronmuller H 1998 J. Appl. Phys. 83 3271[4] Kneller E F and Hawig R 1991 IEEE Trans. Magn.

27 3588[5] Zhang J, Takahashi Y K, Gopalan R and Hono K 2005

Appl. Phys. Lett. 86 122509[6] Tokita M 1993 J. Soc. Powder Technol. Japan 30 790[7] Saito T, Tomonari T and Kageyama H 2004 IEEE Trans.

Magn. 40 2880[8] Ono H, Waki N, Shimada M, Sugiyama T, Fujiki A and

Yamamoto Hand Tani M 2001 IEEE Trans. Magn.37 2552

[9] Yue M, Zhang J, Tian M and Liu X B 2006 J. Appl. Phys.99 08B502

[10] Rama Rao N V, Gopalan R, Manivel Raja M,Chandrasekaran V, Chakravarty D, Sundaresan R,Ranganathan R and Hono K 2007 J. Magn. Magn. Mater.312 252

[11] Grosse G 1993 PC-MOS II Manual and ProgramDocumentation for Mossbauer Spectra Analysis (Germany:FAST Comtec GmbH)

[12] Schneider G, Henig E-T H, Lukes H L and Petzow G 1985J. Less-Common Met. 110 159

[13] Meacham B E and Branagan D J 2003 J. Appl. Phys. 93 7963[14] Hadjipanayis G C and Gabay A M 2004 Proc. 18th Int.

Workshop on High Performance Magnets and theirApplications (Annecy, France) p 590

[15] Yermolenko A S, Zabolotskiy YE I and Korolev A V 1976Phys. Met. Metallogr. 41 52

[16] Vincze I, Campbell I A and Meyer A J 1974 Solid StateCommun. 15 1495

[17] Pop V, Isnard O, Chicinas I, Givord D and Le Breton J M 2006J. Optoelectr. Adv. Mater. 8 494

[18] Cullity B D 1972 Introduction to Magnetic materials(Reading, MA: Addison-Weseley)

[19] Forker M, Julius A, Schulte M and Best D 1998 Phys. Rev. B18 11565

[20] Zhang J, Zhang S Y, Zhang H W and Shen B G 2001 J. Appl.Phys. 89 5601

[21] Ito M, Majima K, Umemoto T, Katsuyama S and Nagai H2001 J. Alloys Compounds 329 272

[22] Goll D, Seeger M and Kronmuller H 1998 J. Magn. Magn.Mater. 185 49

7