Enhanced magnetoresistance and surface state of CrO2 particles improved by chemical process

Article abstract of DOI:10.1016/j.jmmm.2006.03.057

The paper presents an approach to enhance a magnetoresistance (MR) effect in CrO₂ powder compact by an oxidization reaction process. An aqueous potassium permanganate (KMnO₄) was used to react with the CrO₂ particles coated naturally with Cr₂O₃ layer. The experiment indicates that the strong oxidant can effectively adjust thickness of the natural Cr₂O₃ layer, and thereby change the surface state of the CrO₂ particles. Structural and magnetic properties for the improved CrO₂ particles have been characterized by X-ray diffractometer (XRD), X-ray photoelectron spectroscopy (XPS) and SQUID magnetometer. The results exhibit that the magnetotransport behavior of CrO₂ particles depends sensitively on the chemical reaction time. An optimal reaction process yields an obvious increase up to -33% in magnetoresistance at a temperature of 5 K for the chemical treated CrO₂ powder, compared to MR=-27% for the original CrO₂ powder. The mechanism of magnetotransport is assumed to originate from the spin-dependent tunneling in the granular system, which is consistent with our experimental results. The simple chemical approach has a potential to achieve an enhanced magnetoresistance in a metallic particle system by adjusting the surface state of the magnetic nanoparticle.

Full text of DOI:10.1016/j.jmmm.2006.03.057

ARTICLE IN PRESS
Journal of Magnetism and Magnetic Materials 307 (2006) 134–138
www.elsevier.com/locate/jmmm
Enhanced magnetoresistance and surface state of CrO2
particles improved by chemical process
a,
Xiaoyu Zhang , Yajie Chen , Zhenya Li
Ã
b
c
a
Department of Physics and Jiangsu Key Laboratory of Thin Films and Department of Physics, Suzhou University, Suzhou 215006, China
b
Center for Microwave Magnetic Materials and Integrated Circuits and the Department of Electrical and Computer Engineering,
Northeastern University, Boston, MA 02115, USA
CCAST (World Laboratory), P.O. Box 8730, Beijing 100080, China and Department of Physics, Suzhou University, Suzhou 215006, China
c
Received 11 February 2006; received in revised form 16 March 2006
Available online 25 April 2006
Abstract
The paper presents an approach to enhance a magnetoresistance (MR) effect in CrO powder compact by an oxidization reaction
2
process. An aqueous potassium permanganate (KMnO
The experiment indicates that the strong oxidant can effectively adjust thickness of the natural Cr
surface state of the CrO particles. Structural and magnetic properties for the improved CrO particles have been characterized by X-ray
4
) was used to react with the CrO
2
particles coated naturally with Cr
2 3
O layer.
2
O
3
layer, and thereby change the
2
2
diffractometer (XRD), X-ray photoelectron spectroscopy (XPS) and SQUID magnetometer. The results exhibit that the magnetotran-
sport behavior of CrO particles depends sensitively on the chemical reaction time. An optimal reaction process yields an obvious
increase up to ꢀ33% in magnetoresistance at a temperature of 5 K for the chemical treated CrO powder, compared to MR ¼ -27% for
2
2
2
the original CrO powder. The mechanism of magnetotransport is assumed to originate from the spin-dependent tunneling in the
granular system, which is consistent with our experimental results. The simple chemical approach has a potential to achieve an enhanced
magnetoresistance in a metallic particle system by adjusting the surface state of the magnetic nanoparticle.
r 2006 Elsevier B.V. All rights reserved.
PACS: 72.25.ꢀb; 73.43.Qt; 73.43.Jn
Keywords: Magnetoresistance; Spin-dependent tunnelling; Potassium permanganate
1
. Introduction
tunneling barrier in the intergranular tunneling process. As
a tunneling barrier, the native and stable Cr O phase on
the surface of CrO , has been characterized by Hwang et al.
2
3
Recently, chromium dioxide (CrO ) has received much
2
2
attention due to its half-metallic properties [1,2]. The
nearly perfect spin polarization of CrO has been proposed
to be an ideal material for the magnetic tunneling devices
[8] and Dai et al. [9] Since a surface state/layer on the
metallic particles plays a crucial role in the electronic
tunneling process, many emphases have been placed on
2
[
polycrystalline film and powder compact of CrO can yield
a large negative magnetoresistance (MR) at low tempera-
ture [5,6]. The large MR is attributed to spin dependent
intergranular tunneling effect [7,8]. Here, either the grain
boundary or surface layer is believed to play a role of
3,4]. Previous investigations have indicated that the
surface natures of CrO particles in the past decade [10,11].
2
If the tunneling barrier is too thick, tunneling MR will
decrease dramatically. Since Cr O is a stable thermo-
2
2
3
dynamic oxide, it is difficult to reduce the antiferromag-
netic Cr O by physical process. In this paper, we introduce
a simple and effective method to change Cr O oxide layer,
2
3
2
3
i.e. a strong oxidant potassium permanganate (KMnO ) is
4
Ã
used to react with Cr O . The electrical and magnetotran-
3
sport property has been systemically studied with different
reaction times. Here, an enhanced MR in the improved
2
Corresponding author.
E-mail addresses: ys0108012@suda.edu.cn, yachen@ece.neu.edu
X. Zhang).
(
0304-8853/$ - see front matter r 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmmm.2006.03.057

ARTICLE IN PRESS
X. Zhang et al. / Journal of Magnetism and Magnetic Materials 307 (2006) 134–138
135
CrO particles has been observed at low temperature. It is
2
believed that the increase in MR is associated with an
optimal thickness of Cr O surface layer, depending on the
chemical reaction time and temperature.
2
3
2
. Experimental
In this experiment, we used an aqueous potassium
permanganate (KMnO ) to react an original CrO powder
4
2
that was supplied by Micro Magnetics, USA, and was
acicular single-domain particles with an average length of
400 nm and an aspect ratio of about 9:1. The reaction was
carried out in 0.01 mol/l mol concentration and at bath
temperature of 50 1C.
Firstly, we prepared 0.01 mol/l KMnO solution 100 ml
4
and dipped 1.5 g CrO powders carefully into the initial
2
solution while stirring continuously at a bath temperature
of 50 1C. A 1–3 nm thick Cr O layer on CrO particles
Fig. 1. X-ray diffraction patterns of CrO
2
particles for the sample t0,
2
3
2
sample t15 and t30. Inset shows the detailed scan over the range
32 1o2yo35 1.
corresponds approximately to about 5% amount of the
pure CrO powders sample [9]. Namely, there is about
2
0
.000493 mol Cr O in 1.5 g CrO powders. Subsequently,
2 3 2
the reacted CrO2 powders were washed with deionized
water several times and then heated to 50 1C in order to
evaporate the solvent. Finally, the prepared powders were
pressed into pellets with a diameter of 10 mm and a
thickness of 2 mm.The samples were coded with t0, t5, t, t,
t20, t25 and t30, corresponding to different reaction time:
0
, 5, 10, 15, 20, 25, and 30 min. The crystalline structure of
the samples was determined using Rigaku D/MAX 2500/C
X-ray diffractometer (XRD). Magnetic measurements were
carried out using a SQUID magnetometer (Quantum
Design MPMS XL). In the temperature range of 4 KoT
o300 K, the electrical resistivity was measured by a
standard four-probe method using Keithley 2400 Source-
Meter and 2182 NanovoltMeter. Chemical element and
valence state were determined using PHI-550 X-ray
photoelectron spectroscopy (XPS).
Fig. 2. Variation of resistivity with the chemical reaction time for CrO
particles at 290 K.
2
3
. Results and discussion
In order to adjust the stable trivalent Cr O surface
2
3
scan, i.e. a small scan step (0.002 1) and long scan
time (0.1 1/min). The results are presented in the inset of
Fig. 1. A very clear signal of Cr O (1 0 4) peak has
layer, a strong oxidant (KMnO ) is used to react with
4
Cr O . The chemical reaction can be expressed as follows:
2
3
2
3
MnO þ Cr2O3 þ H2O ¼ 2MnO2 þ 2CrO þ 2Hþ.
ꢀ
2ꢀ
been observed in the samples t0 and t15, not in the sample
t30. This means that the Cr O on surface of the CrO
2
2
4
4
2
3
(1)
particle has been completely removed after an etching of
15 min. Fig. 2 presents the variation of electrical resistivity
of the samples with reaction time at 290 K. Since the
measured resistivity arises mainly from the insulating
Cr O layer on surface of CrO particles, the resistivity
This chemical reaction indicates that the KMnO4
oxidant can react effectively with Cr O , and finally
3
convert Cr to Cr
solution. In this experiment, the solution showed a weak
acidity due to a pH value of 6 after reaction.
Fig. 1 shows the XRD patterns for the samples t0, t15
and t30, respectively. A weak Cr O peak (1 0 4) appears
near 2yꢁ33 1 in the sample t0 and t15, whereas the sample
t30) does not exhibit this peak. Therefore, the original
CrO2 powders are analyzed by XRD in a precision
2
3
+
6+
ion that ultimately remains in the
2
3
2
decreases rapidly with the reaction, especially during initial
reaction of 10 min. However, the change in resistivity is
insensitive to the reaction time for a longer reaction time.
As a result of the reaction, a desired thickness of the
insulating Cr O layers may be achieved by control of the
2
3
(
2
3
reaction time.

ARTICLE IN PRESS
X. Zhang et al. / Journal of Magnetism and Magnetic Materials 307 (2006) 134–138
136
effectively removed from surface of the original CrO2
particles. Furthermore, a small amount of Mn element
(
a)
(
ꢁ3 at%) is found in the sample t15 and t30, which is due
to an incomplete washing in the two samples. The binding
energy of Mn 2p state is also presented in Fig. 3(c). The Mn
2p state in the sample t15 and t30 are in good agreement
with MnO2.
The specific magnetization M as a function of chemical
reaction time is shown in Fig. 4. The measurement was
performed at a temperature of 5 K and external magnetic
field of 50 kOe. Clearly, there is a significant increase in the
specific magnetization for the reaction time of 10 min. It is
assumed that the enhanced M is strongly dependent on the
decreased thickness of Cr O surface layer. It is worth
(
b)
2
3
noting that a tendency of the decrease in the specific
magnetization M appears in t410 min. Since the mol
concentration ratio of KMnO to Cr O is 2:1, the mol
4
2
3
concentration of KMnO is very small after a reaction of
4
30 min.KMnO4 is unable to react with CrO2 drasti-
cally after Cr O has been etched completely. At the same
2
3
time, XPS of Cr valence state and resistivity measurements
are hardly sensitive to etching time for t410 min. These
results indicate that it is unlikely to produce any other
phases after long time etching. We consider that the
possible reason for decrease in magnetization M is that a
(
c)
small amount of KMnO destroys spin arrangement on
4
surface of CrO . This leads to a disorder or spin canting in
2
surface spin. Furthermore, the disorder or spin canting of
surface spin may induce the large coercivity H . The
c
increase in H can be observed in sample t30, as presented
c
in the inset of Fig. 4.
Fig. 5 gives the dependence of MR on the chemical
reaction time at a low temperature of 5 K and applied field
of 50 kOe. Due to a thick and natural Cr O layer on the
Fig. 3. Specific magnetization M as a function of chemical reaction time
at a low temperature of 5 K and applied field of 50 kOe. Inset shows the
low field hysteresis loops for the CrO2 powder compacts (H ¼ 4000 Oe).
2
3
CrO particle surface, the original CrO particles has a
2
2
MR ¼ ꢀ27%, which is comparable to previous reported
X-ray photoelectron spectroscopy (XPS) patterns for the
samples t0, t15 and t30 are illustrated in Fig. 3(a). Except
for Cr, O and C element, no other elements have been
found in the original CrO particles. It is believed that the
2
small amount of carbon C may arise from the contamina-
tion on the surface of CrO . Fig. 3(b) shows a magnified
2
pattern of binding energy corresponding to Cr 2p3/2 state
for the three samples. With the increase of the reaction
time, the binding energy of Cr 2p3/2 gives rise to a drop
from 576.8 eV (Sample t0) to 576.3 eV (Sample t15) and
5
2
76.2 eV (Sample t30). Obviously, the binding energy of Cr
p3/2 for the sample t0 is in good agreement with previous
reported (576.7-576.8 eV) of Cr O [12,13]. Since XPS is
2
3
sensitive with 1–2 nm into the particle surface, this is a
convincing evidence that the Cr O layer on the original
2
3
CrO2 particle has more than 2 nm in thickness. More
importantly, both the sample t15 and t30 exhibit a almost
same binding energy of Cr 2p, which is consistent with the
Fig. 4. (a) X-ray photoelectron spectroscopy of the sample t0, t15, and
t30, (b) A magnified pattern of the binding energy for Cr 2p state, and
(c) Binding energy for Mn 2p state.
binding energy (576.3–576.4 eV) of CrO [12,14]. Remark-
2
ably, these results imply that Cr O layer has been
2
3

ARTICLE IN PRESS
X. Zhang et al. / Journal of Magnetism and Magnetic Materials 307 (2006) 134–138
137
Fig. 5. Dependence of MR ratio for CrO particle on chemical reaction
time at T ¼ 5 K and applied field H ¼ 50 kOe.
2
ꢀ1/2
Fig. 7. The Ln R as a function of T for the sample t0, t10, t15, t20 and
t30 in the temperature range from 4 to 110 K. Inset presents the resistance
R as a function of temperature for the sample t0.
Fig. 6. MR ratio as a function of magnetic field H at T ¼ 5 K and applied
field up to 50 kOe.
Fig. 8. Dependence of D on chemical reaction time for these CrO
2
samples.
results [7,9]. Apparently, the sample t10 exhibits a large
MR (ꢀ33%) that increases by ꢁ22%, compared to MR of
the original particles. Fig. 6 shows the negative MR as a
function of applied magnetic field for these samples. It is
clear that low-field MR contributes predominately to
overall magnetoresistance of the samples.
where D is proportional to the Coulomb charging energy
E , tunneling barrier thickness s and barrier height k. D can
c
be determined from the slopes of the linear region in Fig. 7.
The dependence of D on the reaction time is illustrated in
Fig. 8. Actually, the difference in the value of D reflects a
difference in the barrier nature on the surface of CrO . For
the original CrO , a large D value of 20.2 K reveals a thick
Previous research work has indicated that the low
temperature magnetotransport mechanism in CrO powder
2
2
sample stems from spin-dependent tunneling effect [7,9].
ꢀ
1/2
2
1/2
Fig. 7 shows ln R as a function of temperature T
range from 4 to 110 K. The ln R varies linearly with T
in the
ꢀ
Cr O layer on the CrO particles, whereas the D exhibits a
2
3
2
minimum value (D ¼ 5.48 K) when the reaction time is
10 min. At the same time, we have noticed that a long time
reaction (t410 min) do not give rise to an obvious change
in D value. The spin-dependent tunneling conduction
process is dominant at low temperature in these samples,
which thereby is sensitive to the barrier thickness [17,18].
If the barrier thickness is too large, the tunneling
at low temperatures. This is a typical temperature
dependence of resistance in the materials in which the
conduction is dominated with a tunneling process. There-
fore, the resistance as a function of temperature can be
expressed as [15,16]
R ¼ R0 expðD=TÞ ¼ R0 exp 2ð2ksEc=TÞ1=2,
1=2
(2)
2
2
2
conductance G ¼ e Z /h [19], which is proportional to

ARTICLE IN PRESS
X. Zhang et al. / Journal of Magnetism and Magnetic Materials 307 (2006) 134–138
138
2
transmission coefficient Z , will decrease due to a drop of
10474069 and 10547128 and Natural Science Foundation
of Jiangsu Education Committee of China under the grant
No. 05KJB140112.
2
Z . This leads to a decrease in the tunneling magnetoresis-
tance. As a result of the chemical treatment, a suitable
Cr O layer between CrO particles yields an enhanced
2
3
2
magnetoresistance for t ¼ 10 min. But a long time reaction
results in a reduction in MR, which is associated with the
remove of Cr O layer.
References
2
3
[
[
[
1] R.J. Soulen, J.M. Byers, M.S. Osofsky, B. Nadgorny, et al., Science
82 (1998) 85.
2] K.P. Kamper, W. Schmitt, G. Gu
1987) 2788.
2
4
. Conclusions
¨
¨
ntherodt, et al., Phys. Rev. Lett. 59
(
3] S. Sundar Manoharan, D. Elefant, G. Reiss, J.B. Goodenough, Appl.
Phys. Lett. 72 (1998) 984.
We have demonstrated a chemical approach to adjust/
remove a natural Cr O oxide layer on the surface of CrO
2
2
3
[
[
4] A. Gupta, X.W. Li, Gang Xiao, J. Appl. Phys 87 (2000) 6073.
5] X.W. Li, A. Gupta, G. Xiao, G.Q. Gong, Appl. Phys. Lett. 71 (1997)
1124.
particle. The approach is based on the reaction between a
strong oxidant KMnO and Cr O at 50 1C. The experi-
4
2
3
mental results exhibit that magnetic and transport property
relies strongly on the reaction time. A negative MR of
[6] X.W. Li, A. Gupta, G. Xiao, Appl. Phys. Lett. 75 (1999) 713.
[7] J.M.D. Coey, A.E. Berkowitz, Ll. Balcells, et al., Phys. Rev. Lett. 80
(
8] H.Y. Hwang, S.W. Cheong, Science 278 (1997) 1607.
1998) 3815.
ꢀ
33% is measured at a temperature of 5 K and an applied
field of 50 kOe for an improved CrO sample. This results
[
[
2
9] J. Dai, J. Tang, H. Xu, L. Spinu, Appl. Phys. Lett. 77 (2000) 2840.
in an increase of 22% in MR, compared to the original
CrO2 powder sample with a thick Cr O3 layer. The
[
10] Y.J. Chen, X.Y. Zhang, Z.Y. Li, Chem. Phys. Lett. 375 (2003) 213.
[11] L.N. Fan, Y.J. Chen, X.Y. Zhang, D.L. Yao, J. Magn. Magn. Mater.
85 (2005) 353.
2
2
enhancement in MR is attributed to an appropriate barrier
nature in the CrO powder compacts. The assumption has
[
[
12] H.A. Bullen, S.J. Garrett, Chem. Mater. 14 (2002) 243.
13] G.C. Allen, M.T. Curtis, A.J. Hooper, P.M. Tucker, J. Chem. Soc.,
Dalton Trans 16 (1973) 1675.
2
been supported by XRD and XPS measurement. We
believe that the work is significance of the control of oxide
barrier layer on the surface of metallic particles.
[14] I. Ikemoto, K. Ishi, K. Shuichi, H. Kuroda, M.A. Alario Franco,
J.M. Thomas, J. Solid State Chem. 17 (1976) 425.
[
[
[
[
15] T. Zhu, Y.J. Wang, Phys. Rev. B 60 (1999) 11918.
16] P. Sheng, Philos. Mag. B 65 (1992) 357.
Acknowledgements
17] J.C. Slonczewski, Phys. Rev. B 39 (1989) 6995.
18] J.M. MacLaren, X.-G. Zhang, W.H. Butler, Phys. Rev. B 56 (1997)
This work was supported by National Natural Science
Foundation of China under the grant Nos.10174049,
1
1827.
[19] J. Inoue, S. Maekawa, Phys. Rev. B 53 (1996) R11927.