Vacancy mediated room temperature ferromagnetism in Co-doped Dy 2O3

Article abstract of DOI:10.1063/1.4729386

Nanoparticles of Co doped dysprosium oxide [Dy_1.90Co _0.10O₃] were prepared by co-precipitating the precursor salts in presence of air and argon gas. Crystallographic phase and substitution of Co-ion in Dy₂O₃ were confirmed by Rietveld analysis of the x-ray diffraction patterns. Magnetic susceptibility and magnetization as a function of temperature and magnetic field were measured by Faraday and Superconducting quantum inteference device (SQUID) magnetometers, which showed that the sample synthesized in the inert atmosphere is ferromagnetic at room temperature. But no such effect has been observed in the other sample. This observation confirmed that vacancy mediated ferromagnetism can be introduced in the Co-doped dysprosium oxide.

Full text of DOI:10.1063/1.4729386

Vacancy mediated room temperature ferromagnetism in Co-doped Dy2O3
A. Bandyopadhyay, S. Sutradhar, B. J. Sarkar, A. K. Deb, and P. K. Chakrabarti
Citation: Appl. Phys. Lett. 100, 252411 (2012); doi: 10.1063/1.4729386
View online: http://dx.doi.org/10.1063/1.4729386
View Table of Contents: http://apl.aip.org/resource/1/APPLAB/v100/i25
Published by the American Institute of Physics.
Related Articles
Magnetization of 2.6T in gadolinium thin films
Appl. Phys. Lett. 101, 142407 (2012)
Amorphous Slater-Pauling like behaviour in magnetic nanoparticles alloys synthesized in liquids
J. Appl. Phys. 112, 063910 (2012)
On the influence of nanometer-thin antiferromagnetic surface layer on ferromagnetic CrO2
J. Appl. Phys. 112, 053921 (2012)
Thermal stability and the magnetic properties of hybrid vanadium oxide-tetradecylamine nanotubes
J. Appl. Phys. 112, 053912 (2012)
Nanostructured thin manganite films in megagauss magnetic field
Appl. Phys. Lett. 101, 092407 (2012)
Additional information on Appl. Phys. Lett.
Journal Homepage: http://apl.aip.org/
Journal Information: http://apl.aip.org/about/about_the_journal
Top downloads: http://apl.aip.org/features/most_downloaded
Information for Authors: http://apl.aip.org/authors
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

APPLIED PHYSICS LETTERS 100, 252411 (2012)
Vacancy mediated room temperature ferromagnetism in Co-doped Dy2O3
A. Bandyopadhyay,1 S. Sutradhar,1 B. J. Sarkar,1 A. K. Deb,2 and P. K. Chakrabarti1,a)
1Solid State Research Lab., Department of Physics, Burdwan University, Burdwan 713 104, West Bengal,
India
2Department of Physics, Raiganj College (University College), Uttar Dinajpur 733 134, West Bengal, India
(Received 13 May 2012; accepted 26 May 2012; published online 20 June 2012)
Nanoparticles of Co doped dysprosium oxide [Dy1.90Co0.10O3] were prepared by co-precipitating
the precursor salts in presence of air and argon gas. Crystallographic phase and substitution of
Co-ion in Dy2O3 were confirmed by Rietveld analysis of the x-ray diffraction patterns. Magnetic
susceptibility and magnetization as a function of temperature and magnetic field were measured
by Faraday and Superconducting quantum inteference device (SQUID) magnetometers, which
showed that the sample synthesized in the inert atmosphere is ferromagnetic at room temperature.
But no such effect has been observed in the other sample. This observation confirmed that
C
V
vacancy mediated ferromagnetism can be introduced in the Co-doped dysprosium oxide. 2012
American Institute of Physics. [http://dx.doi.org/10.1063/1.4729386]
Diluted magnetic semiconductors (DMS) have been
exhaustively investigated for their promising applications in
various spintronic devices.1,2 In this domain, enhancement
of magnetic property and Curie temperature of DMS materi-
als have been the subject of intense research in recent
times.3–5 In the DMS family, transition metal (TM) ion
doped ZnO, CeO2, SnO2, TiO2, etc. are well known
spintronic materials which were exhaustively studied.6,7
Recently, it is reported that Mn-doped germanium,
Mn5ꢀxFexGe3, etc. are also promising spintronic materials.8,9
But in this context, the TM-ion doped nanocrystalline rare
earth (R ¼ Ce to Lu) oxides were not yet considered largely,
though the rare earth oxides may become an interesting field
for DMS/DMD (dilute magnetic dielectrics) systems as the
free ion magnetic moments of many R3þ ions are quite high.
Besides, the magnetization of TM-ion doped R-oxides would
be high as both the dopant and doping ions have high value
of magnetic moments, which is not the case in the conven-
tional DMS systems.6,7 There are several reports on the
introduction of ferromagnetism in the TM-ion doped DMS
system but till now the reports on the introduction of ferro-
magnetism in the DMD system is very few. Considering the
importance of rare earth oxides, particularly the broad inter-
est in Dy-based oxides,10,11 in the present work, we have
investigated Co-doped Dy2O3. The choice of Dy3þ ion in the
rare earth series is due to its high value of free ion magnetic
moment (10.42lB). Besides, Dy2O3 is also an established
dielectric material12,13 and this system may be a potential
DMD candidate after doping with Co ion. Two nanocrystal-
line samples of Dy1.90Co0.10O3 were prepared by co-
precipitation method. In the first sample (S1), the different
steps in the course of preparation viz., synthesis, filtration,
drying, and annealing were carried out in presence of air,
while in the case of second sample (S2), those operations
were carried out in presence of argon gas to create the defect
by oxygen vacancy. Besides this, the dissolved oxygen gas
of the triple distilled water was also boiled out before its use
in the co-precipitation method for the preparation S2. Inter-
estingly, it has been found that at room temperature (RT), S2
is ferromagnetic with high value of saturation magnetization,
while S1 is a clear paramagnetic at RT. Thus for the first
time, it is observed that the oxygen vacancy mediated ferro-
magnetism can also be introduced in the case of nanopar-
ticles of Co-doped Dy2O3.
The precursor materials used for the preparation of S1
and S2 were cobalt acetylacetonate Co(C5H7O2)2 and
dysprosium chloride. Dysprosium chloride was prepared by
adding concentrated hydrochloric acid in the slurry of dys-
prosium oxide. The newly prepared solution of dysprosium
chloride was slowly evaporated to reduce the extra hydro-
chloric acid and to obtain the microcrystals of dysprosium
chloride. The microcrystals of dysprosium chloride and the
required amount of cobalt acetylacetonate were added in
250 ml alcohol to prepare the salt solution. Diluted sodium
hydroxide solution was slowly added to the salt solution and
the final pH was maintained atꢁ8. The details of the co-
precipitation process were given in our earlier papers.14,15
The as-prepared sample was annealed at TA ꢁ700 ꢂC for 6 h.
The dc magnetic susceptibility was measured by our home-
made Faraday setup in the temperature range of 300-
12 K.14–16 Static magnetic measurements viz., magnetization
vs. field (M-H) loops, field-cooled (FC) and zero FC (ZFC)
curves, etc. were carried out in Quantum Design SQUID
Magnetometer in the temperature range of 300-2 K.
X-ray diffractograms (XRD) of S1 and S2 are shown in
Fig. 1. Here, we have adopted the modified Rietveld method
to derive the structural (atomic coordinates, lattice parame-
ters, etc.) and microstructural parameters (coherent domain
size/crystallite size, micro strain, etc.). The whole pattern fit-
ting has been done with the help of the software MAUD.17
Here each profile was fitted with a pseudo-voigt type func-
tion. These microstructural parameters along with the other
structural parameters were set as refinable parameters. The
accuracy of profile fitting was judged by the goodness of fit
(GOF) and the reliability parameters (Rp, Rw, RB, etc.),
which were defined in our earlier paper.15 The position and
relative intensities of the diffraction peaks are in good
a)Author to whom correspondence should be addressed. Electronic mail:
pabitra_c@hotmail.com. Tel.: þ91 342 2657800. Fax: þ91-342-2534200.
C
V
0003-6951/2012/100(25)/252411/5/$30.00
100, 252411-1
2012 American Institute of Physics
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

252411-2
Bandyopadhyay et al.
Appl. Phys. Lett. 100, 252411 (2012)
tice parameter of each doped sample did not change due to
the doping of Co ion. This is quite likely due to the small dif-
˚
ference in the ionic diameters of Co (ꢁ0.86 A) and Dy
˚
(ꢁ0.91 A) ions. Due to the small difference in the ionic
diameters, it is also expected that the substitution of Co ion
in the lattice of Dy2O3 could not introduce much stress or
strain in the unit cell. This fact is also substantiated by our
Rietveld analysis where we have obtained the low value of
strain (ꢁ2.3 ꢃ 10ꢀ4 for S1 and 2.4 ꢃ 10ꢀ4 for S2). To
explore the role of oxygen vacancy in the introduction of the
observed ferromagnetism of the sample S2 discussed in the
next paragraph, we have estimated the oxygen vacancy in S1
and S2 from the Rietveld refinement and the estimated val-
ues of occupancy factors are, respectively, 0.97 and 0.727
for S1 and S2. Thus the sample prepared in the inert atmos-
phere is more oxygen deficient form (Dy0.95Co0.05)2O3ꢀx
,
with x ¼ 0.819 than S1.
FIG. 1. X ray diffraction patterns along with the Rietveld analysis (a) S1
and (b) S2.
The magnetic susceptibilities (v) measured with the
application of the magnetic field of 850 Oe of S1 and S2 in
the temperature range of 300 to 12 K are shown in Figs. 2(a)
and 2(b). The measured susceptibility recorded at different
temperatures was tried to fit in the well-known Curie-Weiss
ꢀ
agreement with the cubic structure having space group Ia3
(according to JCPDS file No. 79-1722). Interestingly, the
values of GOF in both samples are very close to 1 (1.02 for
S1 and 1.08 for S2), which hint that the observed XRD pat-
terns are well fitted by the derived pattern. Also, it has been
found from the refinement that there is no extra peak due to
any impurity phase in the observed pattern and this finding
ruled out the formation of any unwanted phase in the sam-
ples. The average nanocrystallite sizes estimated from the
Rietveld refinement are 18 nm for S1 and 24 nm for S2. The
lattice parameters of both the samples obtained from the
Rietveld fitting are same (1.069 nm) and this value is very
close to the value of the pure sample of Dy2O3. Thus the lat-
C
law v ¼ ðTꢀhÞ ; where C is the Curie constant and h is the Cu-
rie temperature and the respective fittings along with the
observed data for S1 and S2 are shown in Figs. 2(a) and 2(b).
From Fig. 2(a), it is evident that v vs T curve of S1 is well
fitted by the Curie-Weiss law but the poor fitting of S2 (Fig.
2(b)) indicates that v vs T curve is not like a paramagnetic
type of behavior and from this, it is evident that the sample
S2 is not in the paramagnetic phase. We have also recorded
that M-H (maximum field ꢁ2240 Oe) data in the Faraday
setup. In this region of field, we have observed that the
FIG. 2. Thermal variation of average mass susceptibility of (a) S1, (c) S2, (b) ZFC and FC magnetization of S1, and (d) S2.
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

252411-3
Bandyopadhyay et al.
Appl. Phys. Lett. 100, 252411 (2012)
response of magnetization with the increase of field at RT
deviates strongly from the linear type of variation in case of
S2, but the response curve is linear in case of S1. The above
observation initiated us to measure the ZFC and FC curves
as well as M-H curves in SQUID magnetometer. FC and
ZFC curves of S1 and S2 are displayed in Figs. 2(c) and 2(d)
and from Fig. 2(d), it is clear that the two curves of S2 are
well separated from each other which suggest the presence
of field cooling effect that is the presence of magnetic order-
ing. But in S1, both the curves superimposed each other
which ruled out the presence of any ordering. The nature of
variation of S2 also suggests that the system consists of both
ferromagnetic as well as paramagnetic phase. The bifurca-
tion of FC and ZFC starts above 300 K and hence the block-
ing temperature in case of sample S2 is more than 300 K and
consequently, there is no superparamagnetic effect at and
below 300 K. From the FC and ZFC curves, it is also clear
that the magnetization increases very rapidly below 50 K and
this sharp rise in the magnetization is the characteristics of
DMS/DMD material and related to the weakly coupled and
uncoupled Co ions.18 It is further to be noted that RT mag-
netic susceptibility of S2 (8.65 ꢃ 10ꢀ4) is about 4-times than
that of S1 (2.15 ꢃ 10ꢀ4) but at lower temperature, the values
are nearly equal. This is due to the characteristic features of
ferromagnetic and paramagnetic samples.
introduced in the sample (S2) which is prepared in the inert
atmosphere and this observation also establish the fact that
ferromagnetism can be introduced in the Co-doped rare
earth oxide of Dy2O3 by the creation of oxygen vacancies.
Some selected M-H loops at different temperatures viz.,
10, 30, 100, and 200 K are shown in Figs. 3(b)–3(d) and 3(f),
respectively. From the displayed loops, it is evident that the
maximum magnetization increases with the lowering of tem-
perature. Also, the tendency of saturation increases with the
lowering of temperature. Interestingly, at 10 K, clean satura-
tion was observed in the M-H loop, where the saturation
magnetization is quite high (ꢁ110 emu/g). Also, the coercive
field increases from 770 Oe to 1449 Oe down to 100 K, but
below this temperature its value decreases steadily and
reaches 310 Oe at10 K. Lowering of coercive field indicates
the lowering of magneto-crystalline anisotropy with the low-
ering of temperature below 100 K.
It is well known that Dy2O3 is a paramagnetic and
shows no signature of saturation up to the maximum field of
30 T.19 In the present case, no ferromagnetic ordering has
been introduced in S1 and this may be due to the fact that in
case of Co-ions, the exchange coupling is short ranged and
the sign of the exchange interaction depends on the arrange-
ment of the concerned ions in the crystalline arrangement.
But strong ferromagnetic ordering can be introduced through
defect/oxygen vacancy but the interaction is still short
ranged.20 This condition is achieved by synthesizing the
sample in the inert atmosphere where the concentration of
Co-ions and, eventually, the defect created in the system are
equal or greater than the percolation limit.20
Fig. 3 represents the RT M-H curves of S1 and of S2.
The M-H curve of S1 (Fig. 3(a)) recorded at RT (300 K) is
a clear straight line with no signature of nonlinearity or
any coercivity up to the maximum applied field of ꢁ3 T.
But the M-H curve of S2 recorded at RT (Fig. 3(b)) is a
clear hysteresis loop where the extracted values of the co-
ercive field and magnetization are ꢁ770 Oe and 6.45 emu/g,
respectively. Thus ferromagnetic ordering has been
The origin of the observed FM at RT in the sample S2
could arise from a number of sources viz., intrinsic property
of the doped ions, formation of some nanoscale “Co” related
FIG. 3. (a) Room temperature M-H curves of (a) S1 (b) S2 and M-H curves of S2 observed at (c) 10 K, (d) 30 K, (e) 100 K, and (f) 200 K.
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

252411-4
Bandyopadhyay et al.
Appl. Phys. Lett. 100, 252411 (2012)
FIG. 4. Observed initial magnetization
vs field (M-H) curve of sample S2 fitted
with BMP model.
secondary phase viz., precipitation of Co and/or CoO, etc.
However, the cause of the introduction of the ferromagnet-
ism by the CoO phase can be easily ruled out, since CoO is
antiferromagnetic with a Nee´l temperature of 293 K. Sec-
ondly, the cause of the ferromagnetism observed in the sam-
ple S2 by metallic Co is also an unlikely source of this
ferromagnetism, as XRD and Rietveld analysis show no me-
tallic Co-clusters in the sample of S2. In DMS/DMD sys-
tems, different mechanisms of ferromagnetism have been
reported. Hou et al.21 reported that the carrier-induced ferro-
magnetism (RKKY mechanism) might be applied to explain
the ferromagnetism, in which the free carrier concentration
is vital to determine whether the material is paramagnetic or
ferromagnetic. But this is not applicable in the present case.
Recently, Coey et al.,22 reported that bound magnetic po-
laron (BMP)-mediated ferromagnetic exchange may be re-
sponsible for the introduction of ferromagnetism in the
systems which have oxygen vacancy. Among the possibility
of the different causes behind the introduction of ferromag-
netism, the most plausible explanation for the present case is
the defect/oxygen vacancy created by maintaining the entire
preparation condition in the inert atmosphere. This oxygen
vacancy in the sample S2 is also in good agreement with the
results of Rietveld analysis. The other explanation cannot
substantiate the occurrence of ferromagnetism in the present
case. We have tried to explain the occurrence of ferromag-
netism in our present case by BMP model. The magnetic
exchange interaction between O vacancy and Co ions align
all the Co spins around the O vacancy, forming BMPs. With
the filling up of oxygen vacancies (S1), neighboring Co ions
which were coupled via an oxygen vacancy (ferromagnetic
exchange) are now being coupled by an oxygen bond (super
exchange interaction), which were responsible for the drastic
destruction of ferromagnetism in case of S1.
where localized charge carriers strongly interact with the
doped transition metal ions over the Bohr radius and the para-
magnetic matrix. The spontaneous moment of the system
M0 ¼ Nms, where ms is the actual spontaneous moment of
each BMP and N is the number of BMPs per unit volume,
1
LðxÞ ¼ cothxꢀ x is the Langevin function with x ¼ Hmeff =TK
,
B
where meff is the effective spontaneous magnetic moment of
each BMP. Here, we have neglected the interaction between
the BMP and thus fitting procedure is done by assuming
ms ¼ meff . We have analyzed the M-H curve recorded at 10
and 2 K, by BMP model as in these cases, magnetic moments
were more-or-less saturated and the respective fittings are
shown in Fig. 4. From Fig. 4, it is clear that the observed M-H
curves are well fitted by the BMP model. The number of BMP
and effective magnetic moment per BMP estimated from the
fittings are, respectively, 1.56ꢃ 1021, 6.05ꢃ 10ꢀ20 (2 K),
6.82ꢃ 1020, 1.60ꢃ 10ꢀ19 (10 K). The estimated number of
BMP is almost ten times to that of percolation in Co doped
ZnO (Ref. 24) and this is quite likely due to the high value of
magnetization of the present sample.
Though various works regarding the introduction of
ferromagnetism in ZnO, SnO2, CeO2, etc. after doping with
transition metal ions (Fe, Co, Ni, etc.) have been
reported,25,26 but for the first time here, we report the intro-
duction of RT ferromagnetism of dysprosium oxide after
doping with Co ion. The present work showed that the doped
system is not only a new DMD system but also confirmed
the fact that oxygen vacancy is the main key factor for the
introduction of exchange interaction in the system. Rietveld
refinement ruled out the formation of CoO and/or Co phase
and from the detailed Rietveld analysis of the samples, we
may also conclude that the Co ions have been successfully
entered in the Dy2O3 lattice and no other phase except
Dy2O3 has been formed. Thus, the observed ferromagnetism
is not due to the other phase. The saturation magnetization is
remarkably high which suggests that Co doped Dy2O3 may
be a potential DMD candidate for applications.
In order to explain the ferromagnetism by BMP model
in our case, we have fitted the measured initial M-H curves
in terms of the bound magnetic polaron (BMP) model like
the previous work.23 According to this approach,
Authors wish to acknowledge the financial support
obtained from the DST, Govt. of India (File No. SR/S2/
CMP/0058-2010).
M ¼ M0LðxÞ þ vmH:
(1)
The first term is the contribution coming from BMP and sec-
ond term is due to the paramagnetic matrix. According to the
above equation, the sample is visualized as a mixture of BMPs
1H. Ohno, Science 281, 951 (1998).
2A. Stroppa, S. Picozzi, A. Continenza, M. Y. Kim, and A. J. Freeman,
Phys. Rev B 77, 35208 (2008).
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions

252411-5
Bandyopadhyay et al.
Appl. Phys. Lett. 100, 252411 (2012)
3K. Ueda, H. Tabata, and T. Kawai, Appl. Phys. Lett. 79, 988 (2001).
4A. M. Nazmul, T. Amemiya, Y. Shuto, S. Sugahara, and M. Tanaka, Phys.
Rev. Lett. 95, 17201 (2005).
16J. Mondal, S. Acharya, D. Bisui, K. N. Chattopadhyay, and P. K. Chakra-
barti, J. Appl. Phys. 105, 063921 (2009).
17L. Lutterotti, S. Matthies, and H.-R. Wenk, “MAUD: a friendly Java pro-
gram for material analysis using diffraction,” IUCr: Newsl. CPD 21, 14
(1999).
5A. Stroppa, X. Duan, M. Peressi, D. Furlanetto, and S. Modesti, Phys.
Rev. B 75, 195335 (2007).
6A. Tiwari, V. M. Bhosle, S. Ramachandran, N. Sudhakar, J. Narayan, S.
Budak, and A. Gupta, Appl. Phys. Lett. 88, 142511 (2006).
7Z. W. Zhao, B. K. Tay, J. S. Chen, J. F. Hu, B. C. Lim, and G. P. Li, Appl.
Phys. Lett. 90, 152502 (2007).
18D. A. Schwartz, K. R. Kittilstved, and D. R. Gamelin, Appl. Phys. Lett.
85, 1395 (2004).
19N. Małgorzata, E. Kampert, U. Zeitler, and J. A. Peters, J. Am. Chem.
Soc. 130, 5335 (2008).
8A. Stroppa, G. Kresse, and A. Continenza, Appl. Phys. Lett. 93, 092502
(2008).
20R. Hanafin, C. D. Pemmaraju, T. Archer, and S. Sanvito, Acta Phys. Pol.
A 115, 263 (2009).
9A. Stroppa, G. Kresse, and A. Continenza, Phys. Rev. B 83, 085201 (2011).
10Y. Tokunaga, S. Iguchi, T. Arima, and Y. Tokura, Phys. Rev. Lett. 101,
097205 (2008).
21D. L. Hou, X. J. Ye, X. Y. Zhao, H. J. Meng, H. J. Zhou, X. L. Li, and C.
M. Zhen, J. Appl. Phys. 102, 033905 (2007).
22J. M. D. Coey, M. Venkatesan, and C. B. Fitzgerald, Nature Mater. 4, 173
(2005).
11A. Stroppa, M. Marsman, G. Kresse, and S. Picozzi, New. J. Phys. 12,
93026 (2010).
23C. Chiorescu, J. L. Cohin, and J. J. Neumeier, Phys. Rev. B 76, 020404(R)
(2007).
12A. Goswami and R. R. Verma, Thin Solid Films 28, 2 (1975).
13A. A. Dakhel, J. Alloys Compd. 422, 1 (2006).
24B. K. Santara, B. Pal, and P. K. Giri, J. Appl. Phys. 110, 114322
(2011).
14S. Modak, M. Ammar, F. Mazaleyrat, S. Das, and P. K. Chakrabarti,
J. Alloys Compd. 473, 15 (2009).
25X. F. Liu and R. H. Yu, J. Appl. Phys. 102, 083917 (2007).
26D. Y. Inamdar, A. K. Pathak, I. Dubenko, N. Ali, and S. Mahamuni,
J. Phys. Chem. C 115, 23671 (2011).
15A. Bandyopadhyay, A. K. Deb, K. Mukhopadhyay, S. K. Roy, and P. K.
Chakrabarti, J. Mater. Sci. 47, 2284 (2012).
Downloaded 09 Oct 2012 to 128.143.23.241. Redistribution subject to AIP license or copyright; see http://apl.aip.org/about/rights_and_permissions