On the observation of negative magnetization under zero-field-cooled process

Article abstract of DOI:10.1016/j.ssc.2010.03.016

We have addressed problems associated with the measurement of zero-field-cooled (ZFC) magnetization and its interpretation in the light of negative magnetization reported in certain ferrimagnetic materials such as CoCr₂O₄. We demonstrate that a small negative trapped field in the sample space as well as large coercive fields are responsible for the observed negative magnetization. The problem is commonly encountered while working with magnetometers and a superconducting magnet where the sign of the trapped field can be positive or negative depending on the way the field is reduced to zero.

Full text of DOI:10.1016/j.ssc.2010.03.016

Solid State Communications 150 (2010) 1162–1164
Contents lists available at ScienceDirect
Solid State Communications
journal homepage: www.elsevier.com/locate/ssc
On the observation of negative magnetization under zero-field-cooled process
∗
Nitesh Kumar, A. Sundaresan
Chemistry and Physics of Materials Unit, Jawaharlal Nehru Centre for Advanced Scientific Research, Jakkur P.O., Bangalore 560 064, India
a r t i c l e i n f o
a b s t r a c t
Article history:
We have addressed problems associated with the measurement of zero-field-cooled (ZFC) magnetization
and its interpretation in the light of negative magnetization reported in certain ferrimagnetic materials
such as CoCr2O4. We demonstrate that a small negative trapped field in the sample space as well as
large coercive fields are responsible for the observed negative magnetization. The problem is commonly
encountered while working with magnetometers and a superconducting magnet where the sign of the
trapped field can be positive or negative depending on the way the field is reduced to zero.
Received 23 December 2009
Received in revised form
1
7 February 2010
Accepted 10 March 2010
by D.D. Sarma
Available online 17 March 2010
©
2010 Elsevier Ltd. All rights reserved.
Keywords:
A. Ferrimagnets
D. Magnetization reversal
E. Magnetization
The phenomenon of ferrimagnetism is well known since its
discovery in spinel ferrites where the spontaneous magnetization
arises due to uncompensated magnetic moment resulting from
magnetic interactions among the magnetic ions at two different
crystallographic sites. In 1948, Néel predicted [1] that certain fer-
rimagnetic materials will exhibit spontaneous magnetization that
changes sign with temperature or magnetic field due to different
temperature dependence of sublattice magnetization associated
with the two different crystallographic sites. This phenomenon
was indeed experimentally observed in mixed spinels, for exam-
ple, in Li0.5Fe2.5−xCrxO4 and Co2VO4 [2,3]. Later, many other fer-
rimagnetic systems such as garnets, molecular magnets and even
antiferromagnetic [4–7] materials were shown to exhibit negative
magnetization below a compensation temperature. In the case of
orthovanadates RVO3 (R =La, Nd, Sm, Gd, Er and Y), various models
including antisymmetric Dzyaloshinsky–Moriya interactions have
been proposed to account for the negative magnetization [8]. Ex-
perimentally, there are two ways of measuring magnetization as a
function of temperature; these are field-cooled (FC) and zero-field-
cooled (ZFC) processes which are based on whether the material
is cooled under an applied magnetic field or not, respectively. The
former represents the equilibrium magnetization and the latter the
non-equilibrium time dependent magnetization.
on bulk and nanoparticles of CoCr2O4 have shown that these
samples exhibit negative magnetization just below TC under
ZFC measurement condition with low applied field (∼100 Oe),
while the magnetization remains positive in the case of FC
measurement condition for the same applied field [11,12]. This
behaviour is remarkably different from what is seen in the above
mentioned materials where the magnetization changes sign at the
compensation temperature which lies well below the magnetic
ordering temperature. The observed negative magnetization in
CoCr O4 has been attributed to uncompensated spins at the
2
grain boundaries [11]. But, it is difficult to conceive the negative
magnetization based on the uncompensated spins. It is most likely
arising from an artefact that there could be a small trapped field
in the superconducting magnet during cooling that can affect the
magnetization drastically under ZFC condition as observed in RVO3
system [13,14]. However, if the origin of negative magnetization
is intrinsic, such studies would be interesting from physics point
of view as well as possibility of exploiting the phenomenon for
device applications such as magnetic memory and switching. It
is for these reasons that we have reinvestigated the magnetic
properties of CoCr2O4 and demonstrate that the observed negative
magnetization under ZFC condition is an artefact due to trapped
field in the superconducting magnet. Although we have shown
that the trapped fields can influence the magnetization behaviour
of CoCr2O4, it is common for ferro- and ferrimagnetic materials,
particularly with significant magnetic anisotropy and coercive
field.
Recently, there has been a lot of interest in the ferrimagnetic
(
TC ∼ 93 K) spinel CoCr2O4 because of its multiferroic properties
with a magnetoelectric coupling below the spiral magnetic
ordering at ∼26 K [9,10]. A few reports of magnetization studies
The sample CoCr2O4 was prepared by solid state reaction of
thoroughly mixed powders of stoichiometric amount of Co3O4
∗
and Cr O3 at 1350 °C in air for several hours with intermittent
Corresponding author. Tel.: +91 80 22082824; fax: +91 80 22082766.
E-mail address: sundaresan@jncasr.ac.in (A. Sundaresan).
2
grindings. Phase purity was checked by the analysis of powder
0
038-1098/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ssc.2010.03.016

N. Kumar, A. Sundaresan / Solid State Communications 150 (2010) 1162–1164
1163
Fig. 2. Magnetization as a function of temperature measured under positive and
negative trapped fields. Inset shows magnetization in the paramagnetic region in
expanded scale.
Fig. 1. ZFC and FC magnetization versus temperature curves of CoCr2O4 measured
with an applied field of 100 Oe (See text). Inset shows ZFC curves at different applied
fields.
sign from negative to positive because of decrease in magnetic
anisotropic constant and coercive field which results in rotation
of spins towards applied field direction. This explains the observed
negative magnetization in ZFC measurement. In order to confirm
this, we measured magnetization under NTF without applying any
external magnetic field and the results are shown in Fig. 2. It is seen
that the magnetization in the entire temperature range including
the paramagnetic state is negative. Similarly, we measured the
magnetization under positive trapped field (PTF) after reducing
the field to zero from a large negative field in oscillation mode.
As expected, the magnetization in the entire temperature range
is positive and mirror image of that measured under NTF. The
positive and negative values of magnetization in the paramagnetic
state are clearly seen in the inset of Fig. 2. The values of NTF
and PTF obtained by equating the susceptibility values in the
paramagnetic state are −2.1 Oe and 3.6 Oe, respectively. These
results further confirm that the negative magnetization observed
in CoCr2O4 under ZFC condition is an artefact arising from the
trapped magnetic field in the superconducting magnet. Similar
behaviour has been observed in CoFe2O4, CoFeCrO4 and other
intermediate compositions under ZFC condition with NTF which
support the non-intrinsic characteristic of negative magnetization
in some spinel systems. In fact, it was shown earlier that
even earth’s magnetic field can influence the magnetization
behaviour of certain magnetic materials [15]. It is noteworthy
that even in the case of materials which exhibit intrinsic negative
magnetization, the ZFC magnetization measured under negative
trapped field does not represent the intrinsic magnetic property
of the material [7,16]. Therefore, we suggest that one has to be
cautious while measuring and interpreting magnetization under
ZFC condition. Unless the trapped field is removed completely
by some means, the magnetization behaviour will be dominated
by the trapped fields depending on the intrinsic magnetic
properties of the material. In order to observe intrinsic negative
magnetization under ZFC condition, the field should be reduced to
zero from a negative field in oscillation mode so that the trapped
field is positive [14]. Since this measurement condition is nothing
but field cooling process under small positive trapped field, the ZFC
magnetization will resemble the behaviour of FC magnetization as
X-ray diffraction (XRD) pattern recorded with a Bruker D8
Discover diffractometer. Magnetic measurements were carried out
with Vibrating Sample Magnetometer (VSM) option in Physical
Properties Measurement System (PPMS), Quantum Design, USA.
Analysis of XRD pattern of CoCr2O4 confirmed that the sample
was single phase having the normal spinel structure (space group
´
Fd3m) with lattice parameter, a
=
8.3343(1) Å. Shown in
Fig. 1 is the magnetization as a function of temperature measured
under ZFC and FC conditions with the same applied field (100 Oe)
used in the literature and the observed magnetization behaviour
is very similar to what is reported earlier [10,11]. There are
two magnetic anomalies, one at 97 K which corresponds to
ferrimagnetic ordering and the other at 27 K is due to noncollinear
conical spiral ordering. More importantly, just below TC , the ZFC
curve shows a peak in magnetization below which it crosses
zero magnetization and becomes negative at low temperatures.
With increasing applied magnetic field, the peak in magnetization
moves to low temperature and becomes broad. Further, the
negative magnetization observed at lower fields become positive
at large enough fields as shown in the inset of Fig. 1.
In order to understand this magnetization behaviour, it is
important to rule out extrinsic origin of such unusual magnetism.
In this respect, it is essential to look at the history of the
superconducting magnet just before doing the magnetization
measurement because the ZFC measurements are very sensitive
to the remanent field present in most of the magnetometers.
As our PPMS is not equipped with low-field option to reduce
the magnitude of the remanent field below 0.1 Oe, we normally
minimize the remanence down to a few Oersteds by setting
the field to zero (from an initial field >1 T) in oscillation
mode. This small field is the result of trapped magnetic flux
inside the superconducting material. Consequently, in the ZFC
measurement, the sample was cooled under the trapped field.
The other important parameter is the sign of the trapped field
that is opposite in sign when reducing the field to zero from
a positive or negative field. In the present case, the field was
reduced from a positive value and therefore the trapped field is
negative. Thus, the sample was field cooled (FC) under negative
trapped field (NTF) to the lowest temperature and then a positive
field of 100 Oe was applied and the magnetization was measured
while warming. During cooling, the NTF forces the moments to
align in the negative direction and the applied positive field
reported for YVO [14]. We also suggest that the negative trapped
3
field should be avoided for all magnetic measurements.
In conclusion, we have demonstrated various artefacts associ-
ated with ZFC magnetization measurement with magnetometers
and a superconducting magnet. Furthermore, we have suggested
magnetization measurement protocol to identify materials ex-
hibiting intrinsic negative magnetization, particularly under ZFC
process.
(
100 Oe) cannot change the magnetization direction to positive
presumably because of magnetic anisotropy and large coercive
field. With increase of temperature, the magnetization changes its

1
164
N. Kumar, A. Sundaresan / Solid State Communications 150 (2010) 1162–1164
References
[9] Y. Yamasaki, S. Miyasaka, Y. Kaneko, J.P. He, T. Arima, Y. Tokura, Phys. Rev. Lett.
6 (2006) 207204.
9
[
10] Y.J. Choi, J. Okamoto, D.J. Huang, K.S. Chao, H.J. Lin, C.T. Chen, M.V. Veenendaal,
T.A. Kaplan, S.W. Cheong, Phys. Rev. Lett. 102 (2009) 067601.
[
[
[
[
1] L. Néel, Ann. phys. 3 (1948) 137.
2] E.W. Gorter, J.A. Schulkes, J.S. van Wieringen, Phys. Rev. 89 (1953) 487.
3] N. Menyuk, K. Dwight, D.G. Wickham, Phys. Rev. Lett. 4 (1960) 119.
4] L. Néel, R. Pauthenet, B. Dreyfus, Prog. Low temp. Phys., vol. 4, North Holland
Publishing Company, Amsterdam, 1965, p. 344.
[11] G. Lawes, B. Melot, K. Page, C. Ederer, M.A. Hayward, T. Proffen, R Seshadri,
Phys. Rev. B 74 (2006) 024413.
[12] D.P. Dutta, J. Manjanna, A.K. Tyagi, J. Appl. Phys. 106 (2009) 043915.
[13] L.D. Tung, Phys. Rev. B 73 (2006) 024428.
[
[
[
5] S. Ohkoshi, Y. Abe, A. Fujishima, K. Hashimoto, Phys. Rev. Lett. 82 (1999) 1285.
6] A. Kumar, S.M. Yusuf, L. Keller, J.V. Yakhmi, Phys. Rev. Lett. 101 (2008) 207206.
7] H. Kageyama, D.I. Khomskii, R.Z. Levitin, A.N. Vasil’ev, Phys. Rev. B 67 (2003)
[14] L.D. Tung, M.R. Lees, G. Balakrishnan, D.M. Paul, Phys. Rev. B 75 (2007)
104404.
224422.
[15] T.V.C. Rao, P. Raj, S.M. Yusuf, L.M. Rao, A. Sathyamoorthy, V.C. Sahni, Phil. Mag.
B 74 (1996) 275.
[
8] Y. Ren, T.T.M. Palstra, D.I. Khomskii, A.A. Nugroho, A.A. Menovsky, G.A.
Sawatzky, Phys. Rev. B 62 (2000) 6577.
[16] R.P. Singh, C.V. Tomy, J. Phys.: Condens. Matter. 20 (2008) 235209.