Chapter 4
The Interface-Sustained Magnetic Properties Displayed by
the La
2
O
3
-SrO-Co
3
O
4
Nanocomposite

4.1 Introduction
As mentioned in the earlier chapter, partial substitution of La
3+
by Sr
2+
in LaCoO
3

leads to remarkable changes in the properties of the material. With the increase in
dopant concentration, the rhombohedral distortion in the perovskite structure is
reduced, oxygen vacancies are generated and a small fraction of Co
3+
is converted to
Co
4+
. The work of Raccah et al.

and Bhide et al.

showed that the rhombohedral
distortion decreases with the introduction of Sr
2+
until x = 0.5, after which the structure
remains cubic (Raccah and Goodenough, 1968; Bhide, et al., 1975). This is because in
the range of 0 < x < 0.5, the structure responds to strontium substitution by steadily
increasing Co
4+
rather than losing lattice oxygen (Yakel, 1955). Thus, the resultant
perovskite oxides, La
1-x
Sr
x
CoO
3-δ
(x < 0.5), possess remarkably different electric and
magnetic properties from their parental form.

This work is a continuation of the work presented in the previous chapter. As
described in the previous chapter, heterogeneous ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
complex
oxide system is studied. These oxide mixtures exhibit much higher coercivity than the
perovskite oxide with the same composition at low temperatures. Petrov reported a

coercivity of 0.03T at a measured temperature of 4.2K for perovskite La
0.6
Sr
0.4
CoO
3-y

and the coercivity of this material is found to decrease with the increase in temperature
(Petrov, et al., 1995). On the contrary, the complex oxide with the same composition (x

79
= 0.4) exhibits a coercivity of 0.165T at 74 K, which is about 5.5 times that of the
perovskite La
0.6
Sr
0.4
CoO
3
measured at a much lower temperature.

It was reported earlier that ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
complex oxide manifested
room temperature ferromagnetism and attributed it to a special interfacial

phenomenon.

Both La
2
O
3
and SrO phases are considered to cause, via interfacial
induction, distortion of the octahedral coordination sphere in the spinel Co
3
O
4
phase.
This phenomenon is known as the “Jahn-Teller” effect and becomes notable only in a
highly dispersed system where a large extent of interfacial contact exists. In this
chapter, the influence of SrO content and chelating ligands on both the coercivity and
remanence of the complex oxide at low temperature are explored.

4.2 Experimental
4.2.1 Chemicals
Lanthanum nitrate hydrate (La(NO
3
)
3
.yH
2
O, 99.99%, Aldrich), strontium nitrate
(Sr(NO
3
)
2

, >99%, Acros Organic), cobalt (II) nitrate hexahydrate (Co(NO
3
)
2
.6H
2
O,
99%, Acros Organic), glycine (≥98.5%, Fluka), ethylene glycol (Mallinckrodt, AR),
poly(vinylbutyral) resin (Butiva-79, Monsanto), toluene (>99.5%, Merck), 2-butanone
(>99.8%, Fisher Scientific), citric acid (>99.5%, Sigma), DL - malic acid (99%,
Acros), lactic acid (about 90%, Merck) and Ethylene diamine tetraacetic acid (EDTA,
Fluka, ≥ 98%) were used as received.

4.2.2 Preparation of the hydrogel using citric acid-ethylene glycol ligands
Similar to the preparation method depicted in Section 3.2.2, a series of hydrogels
containing various molar ratios of metal ions (La
3+
: Sr
2+
: Co
2+
= 1-x : x : 1 and 0 < x ≤

80
0.95) were prepared by the wet chemistry approach. A typical procedure includes
preparation of an aqueous solution of La(NO
3
)
3
.yH

2
O, Sr(NO
3
)
2
, Co(NO
3
)
2
.6H
2
O,
glycine and citric acid, in which the molar ratio of total metal cations to the total
functional groups (i.e. amino group and carboxyl group) of citric acid and glycine was
maintained at 0.154 (mass ratio of citric acid to glycine is 0.129). Ethylene glycol
(77% by volume) was then added to this solution. The solution was allowed to
concentrate on a hot plate at 200°C to form a gel.

4.2.3 Preparation of the hydrogel by using other types of ligands
Three other types of chelating reagent systems, namely the malic acid/glycine/ethylene
glycol, lactic acid/glycine/ethylene glycol and ethylene diamine tetraacetic acid
(EDTA) were employed respectively to synthesize the hydrogels by using the same
procedure as described in the above section. But only one composition of the
composite, namely x = 0.95, was used in these three types of gels for investigating the
ligand effect.

4.2.4 Pyrolysis and calcination
The preparation of the testing samples is similar to that reported in Section 3.2.3. The
gel, as obtained from Section 4.2.2 was heated to 400°C to execute pyrolysis to yield a
black powder. This black powder was calcined at 600°C for 2h under air purge to

ensure complete removal of carbon residues and growth of the crystal phases in the
three oxides.

The resulting ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
powder was then ground and added to a
polymer solution comprising of poly(vinylbutyral) dissolved in the mixture of toluene

81
and 2-butanone (v/v=1). After evaporation of the solvent, the lumps were ground into
fine powder (<40μ) and pressed into pellets (
1
≅
d
cm). The pellet was then combusted
at 400°C to burn away the polymer binder and calcined at 600°C or higher (e.g. 750°C,
800°C and 875°C) for 1 h to conduct solid phase reaction of the three oxides.

4.2.5 Characterisations
The degradation profile of the dehydrated precursor gel is obtained using the
thermogravimetric analyser (TGA, TA instrument). A heating rate of 10°C/min in
100ml/min of Nitrogen gas was used and the weight against temperature chart
obtained enables the study of segmental degradation of the precursor with temperature.
The hysteresis loops of the samples were obtained from the Vibrating Sample

Magnetometer (Oxford Magnetometer VSM) with an applied field of 10kOe measured
at various temperatures. X-ray Diffractometer with Cu Kα radiation (Philips X’Pert)
was employed to determine the crystalline phases of the samples. The electrical
properties of the complex oxides and their homogeneous counterparts were
investigated using an impedance analyser (Solartron 1226). The morphology of nano-
composite was observed with a Field Emission Scanning Electron Microscope (FE-
SEM, JEOL, JSM-6700F) and the nano-scaled phase domain morphology of the
complex oxides was examined using High-Resolution Transmission Electron
Microscopy (HR-TEM, Philips CM300). The X-ray photoelectron spectrometer (XPS,
Kratos Axis HSi System) is employed to investigate the chemical composition and
environment of the heterogeneous oxides using C 1s peak (284.6 eV) as the internal
reference. The electron paramagnetic resonance spectroscopy (EPR) was obtained
using X-band Elexsys E500 CW-EPR Spectrometer (Bruker BioSpin GMBH).


82
The first four stages (including peaks a-d), which occurred at temperatures below
550°C, are pyrolysis processes which remove the organic components and nitroxides.
The last peak (labelled e) is the solid-phase reaction among the three different oxide
domains with the removal of oxygen and formation of a solid solution at about 610°C.
X-ray diffraction also verifies that the heterogeneous oxide 0.1La
2
O
3
-0.8SrO/⅓Co
3
O
4

is converted to the hexagonal type structure provided that the calcination is carried out

at temperatures above the threshold of 610°C (Figure 4.2).
The precursor of the complex metal oxides ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
is a metallo-
organic hydrogel formed via chelating bonding between the three metal ions (La
3+
,
Sr
2+
and Co
2+
) and the hydrophilic organic ligands (citric acid and glycine). The
conversion process from precursor to ceramic is analysed using the TGA. From the
TGA profile shown in Figure 4.1 for precursor gel with x = 0.8, it can be observed that
the conversion process involves four weight loss stages.

83
⎯⎯→⎯
C
o
610
We have noted that no reaction would happen at this temperature if the three oxides
are prepared separately and blended together using the same stoichiometry. Thus, the
high reactivity of solid phase reaction must have originated from the uniform

distribution of the three metal ions in the hydrogel, which leaves a composite
consisting of highly divided oxide domains after complete combustion of the organic
components.

4.3.1 Nano-scale grain-boundary structure cast by metallo-organic gel
4.3 Results and Discussion
½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
La
1-x
Sr
x
CoO
3-δ



0
20
40
60
80
100
0 100 200 300 400 500 600 700 800

Temperature (
o
C)
Weight (%)
-0.1
0.1
0.3
0.5
0.7
0.9
Deriv. weight (%/
o
C)
a
b
c
d
e














Figure 4.1 Typical TGA profile of preceramic gel with x = 0.80

84

85

















Figure 4.2 X-ray chart of 0.1La
2
O
3
-0.8SrO/⅓Co
3
O

4
calcined at 600°C and 800°C respectively where H, P, s, l and c represent
hexagonal, perovskite, SrO, La
2
O
3
and Co
3
O
4
phase respectively
20 30 40 50 60 70 80
2 theta
Intensity (arbitrary units)
H
H
H
s
l
p
s / c
s
s
s
p
p / c
l
/ s
c
600°C

800°C
As mentioned earlier, if the pyrolytic product (x= 0.8) is calcined at a temperature
below the threshold of solid reaction, e.g. at 600°C, the heterogeneous tri-oxide
composite is resulted (Figure 4.2). By referring to the XRD patterns of the three
pristine oxides (shown previously in Figure 3.4), the composite could basically retain
the properties of the individual oxides (labelled by l – La
2
O
3
, s – SrO, and c – Co
3
O
4
)
despite having a low content of perovskite phase (labelled by P). The formation of
perovskite phase is considered to occur at the location where both La
2
O
3
and Co
3
O
4

oxides are in proximate contact with each other. According to their XRD patterns, this
perovskite phase fades away quickly with the increase in x as shown in Figure 3.5. The
FE-SEM image of the composite reveals a grain-boundary structure with grain sizes in
the range of 20-30 nm (Figure 4.3a). A closer observation of the individual grain by
HR-TEM (Figure 4.3b) shows two different major types of domains in the nanometer
scale, which cannot be specified by energy dispersive X-ray spectroscopy (EDS) due

to their minute sizes. However, they could be assigned approximately using the
respective radii of metal ions (Sr
2+
-1.18 Å; La
3+
-1.36 Å; Co
2+
-0.65 Å) and the lattice
packing density (Co
3
O
4
> SrO) as the basis. We could observe that the assigned Co
3
O
4

domain comprises of more closely assembled unit cells. Since these domains mutually
interpenetrate in the scale of a few nano-meters, the solid reaction among them can
thus take place at the temperature (e.g. 610
o
C) that is substantially lower than that
needed for the powder blend of the three oxides.






86



a










2
Co
3
O
4
1
SrO - La
2
O
3
b











Figure 4.3 (a) FESEM and (b) HR-TEM images of 0.1La
2
O
3
-0.8SrO/⅓Co
3
O
4

calcined at 600°C



87
The maximum weight-loss rate of peak e on the TGA curve is a measure of the activity
of the solid-phase reaction. Table 4.1 lists the reaction rates caused by the variation of
organic chelating ligand (hydroxycarboxylic acid) in the hydrogels with x = 0.95. It is
found that the reactivity correlates with the functionality (f = number of -CO
2
H and -
OH groups per molecule) of the hydroxycarboxylic acid. The details about the role of
organic ligands will be elaborated in Section 4.3.3.

Table 4.1 Solid reaction rates on different chelating systems
Chelating ligand system
Maximum weight-loss rate of peak e

(% / °C)
Citric acid (f=4)/glycine/ethylene glycol 0.05580
Malic acid (f = 3)/glycine /ethylene glycol 0.04540
Lactic acid (f = 2)/glycine/ethylene glycol 0.03403
EDTA 0.04638


4.3.2 The origins of ferromagnetic properties of the heterogeneous tri-oxide
composites
As mentioned in Section 3.3.1, both the calcination temperature and the SrO content
affect the phase structure of the materials formed. Perovskite solid solution is readily
generated when x ≤ 0.5 at 600°C (Figure 4.4). But at this temperature, as elucidated
above, the tri-oxide composites (x ≥ 0.8) are generated. This can be further validated
through the X-ray photoelectron spectra (Figure 4.5). The Sr 3d XPS reveals rather
complicated multiple peaks. The two distinct doublets, 3d
5/2
and 3d
3/2
that are observed
here were also reported by several previous studies (van der Heide, 2002; Vovk, et al.,
2005). As can be seen from the figure, the doublets appear in the XPS spectra of the
three samples (x = 0.5, 0.8 and 0.95) despite the different crystal structures between

88

89
them (perovskite structure is obtained for x = 0.5 while the other two have a complex
mixed tri-oxide composite). Nevertheless, the perovskite sample (x = 0.50) displays a
shoulder peak at low binding energy side (ca. 131.2 eV), which is proposed to be the
result of coupling of the two sets of doublets due to the two distinct chemical

environments (Vovk et al., 2005).

It can be observed that Sr 3d spectrum of the sample
with x = 0.8 also manifests such shoulder peak, though much weaker, and this is in
agreement with the fact that this heterogeneous oxide sample contains minor
perovskite component (Figure 4.4). Hence, it is clear that only one set of the doublet
peaks is present in Sr 3d spectrum of the heterogeneous trioxide composite with x =
0.95 and this is because of the existence of Sr-O phase and negligible perovskite
phase. On the contrary, the corresponding Sr 3d spectrum of the binary SrO/Co
3
O
4

composite oxide displays a severely overlapped Sr 3d doublet (Figure 4.5d). This
indicates that a very low content of La
2
O
3
phase in the former composite made Sr 3d
spectrum different. The electric conductivity measurement (-lg σ) also shows a leap
from the perovskite solid solution to the heterogeneous tri-oxide composite (Figure
4.6) because the former has mixed-conductive structure that contains electronic
conductivity (Petrov et al., 1995).


20 30 40 50 60 70 80
2 theta
Intensity (arbitrary units)
x = 0.4
x = 0.8

x = 0.9
x = 0.95
P
P
P















Figure 4.4 XRD chart of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with x = 0.4, 0.8, 0.9 and 0.95, calcined at 600°C


90

128130132134136138140
Binding energy (eV)
Intensity
x = 0.50
128130132134136138140
Binding energy (eV)
Intensity
x = 0.80
128130132134136138140
Binding energy (eV)
Intensity
x = 0.95
128130132134136138140
Binding energy (eV)
Intensity
SrO/Co
3
O
4
a
c
d
b
3d
3/2

3d
3/2


3d
5/2

3d
3/2

3d
3/2

3d
5/2

3d
5/2

3d
5/2

131.2 eV














Figure 4.5 Sr 3d photoelectron spectra of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with (a) x = 0.5, (b) x = 0.8, (c) x = 0.95 and (d) x = 0 (which is
SrO/Co
3
O
4
)

91















0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
x
-lg(conductivity)
Figure 4.6 Electrical conductivity vs x measured at room temperature

92

93
The two perovskite solid solutions La
1-x
Sr
x
CoO
3-y
(x = 0.4 and 0.5) exhibit
ferromagnetism at low temperatures (Figure 4.7). This is consistent with the earlier

work done by Petrov et al (Petrov et al., 1995). In contrast to the homogeneous
perovskite samples, the heterogeneous trioxide composites (x = 0.8 and 0.9) reveals a
larger coercivity (Figure 4.8). Both figures show strong temperature dependence of
magnetic properties, which is consistent with the characteristic magnetic behaviour of
the perovskite oxide despite the low contents of perovskite phase in the two composite
samples.

In the tri-oxide composites ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with x = 0.8 and 0.9, there are
two phases that possess magnetic response. They are the spinel Co
3
O
4
and the
perovskite La
1-
α
Sr
α
CoO
3-
β
phase. The content of perovskite phase in the composite

with x = 0.9 is lower than that in the composite with x = 0.8 (Figure 4.4), and the
composite with x = 0.95 contains no perovskite phase according to the intensity of the
peak at 2θ = 33°, which is the characteristic perovskite peak. Besides temperature
sensitivity, the coercivity of these three composites is also dependent on the x value.
Looking at Figure 4.8, it can be noted that coercivity for composite with x = 0.9 is
higher than that of the composite with x = 0.8 when the measured temperature is lower
than 214K. For the composite with x = 0.95, its coercivity exhibits an apparent less
declination with increasing temperature than the other two composites and as it does
not possessed perovskite phase, its coercivity is lower than that of x = 0.8 and 0.9.
Since spinel Co
3
O
4
phase is the only magnetic phase in this composite, the faster
decreasing trend of coercivity exhibited by the two lower-x composites should
therefore be accounted to the effect of perovskite component.

0
500
1000
1500
2000
2500
50 70 90 110 130 150 170 190 210
Temperature (K)
Coercivity (Oe)
x = 0.4
x = 0.5














Figure 4.7 Coercivity vs measurement temperature of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with x = 0.4 and 0.5

94

















0
500
1000
1500
2000
2500
3000
50 100 150 200 250 300
Temperature (K)
Coercivity (Oe)
x=0.8
x=0.9
x=0.95
Figure 4.8 Coercivity vs measurement temperature of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with x = 0.8, 0.9 and 0.95


95

96
crye
gg
It is known that pristine spinel Co
3
O
4
is an antiferromagnet with a Néel temperature of
40K. However, it was reported by Makhlouf that the material exhibits weak
ferromagnetism at around 25K. Similar effect was observed from Wang and co-
workers who reported the temperature to be around 40K (Makhlouf, 2002; Wang, et
al., 2005b). Since the measured temperature of 80K is much higher than the Néel
temperature or any transition temperature as mentioned by Makhlouf or Wang, spinel

The three composites presented in Figure 4.8 all possess certain coercivities at 250K
even though the values are rather small. Since the perovskite phase loses
ferromagnetism at 250K, the coercivity above this temperature is due to the
coexistence of spinel Co
3
O
4
and SrO nano-phases, which are considered to possess
ferromagnetic property. The EPR analysis also supports the different structure
backgrounds responsible for the ferromagnetism (Figure 4.9). On the X-band EPR
spectra of composite x = 0.5 (perovskite oxide) at 200K, a multiple-splitting spectrum
is present, which is an indication of the existence of multiple-crystal field that have
different crystal field splitting energies (Δ

cry
). It is known that the Δ
cry
value affects g
factor through spin-orbital coupling (
=
−
/
Δ
αλ
, where
α
is a parameter related
to the orientation of crystal field and type of transition metal). This EPR spectrum
reflects rather intricate chemical environments (due to the participation of other metal
ions) surrounding cobalt ions (Co
3+
and Co
4+
) in the perovskite phase. However, the
EPR spectrum of composite with x = 0.95 exhibits a much simpler X-band with the
magnetic cobalt ions (Co
3+
) in spinel phase surrounded by oxygen ions.
-150
-100
-50
0
50
100

150
3240 3260 3280 3300 3320 3340 3360 3380 3400 3420
H (G)
Intensity
x = 0.5
x = 0.95
Figure 4.9 EPR spectra of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
for x = 0.5 and 0.95 measured at 200K
















97

98
Co
3
O
4
here manifested paramagnetism (as shown in Figure 4.10). However, in di-
oxide composite SrO/Co
3
O
4
, in which both phases are mixed in the nanoscale, the
material is a ferromagnet. This has been regarded as the interface-sustaining induction
through the Sr-O-Co bonds along the phase boundary. Such bonding brings about
Jahn-Teller distortions of the octahedral coordination interstice in the spinel Co
3
O
4
phase and results in the generation of unpaired electrons as mentioned in the earlier
chapter. Moreover, the composite 0.025La
2
O
3
-0.95SrO/⅓Co
3
O
4
exhibits stronger
coercivity and remanence than SrO/Co

3
O
4
(1
st
row in Table 4.2). It has been verified
that this is due to the generation of a La
3+
-doped Co
3
O
4
phase (refer to Section 3.3.2).
There is an increase in the coercive force in the two composites (x = 0.9 and 0.95)
when temperature is raised from 250K to 298K (Figure 4.8). The doped Co
3
O
4
phase
is thus considered to be more susceptible to interfacial induction than their undoped
counterpart.

Although the complex oxides (x = 0.8 and 0.9) possess stronger coercive forces than
the perovskite solid solutions (x = 0.4 and 0.5), the latter group possesses stronger
remanence (Table 4.3). The lower remanence exhibited by the complex oxides reflect
the fact that they contain smaller numbers of magnetic moments. As concluded earlier,
the ferromagnetism of the complex oxides at temperatures below 200K originates
primarily from the SrO-Co
3
O

4
interfacial phase and partially from the perovskite phase
(only in x = 0.8 and 0.9). As an example, Figure 4.11 depicts the hysteresis loop of
composite with x = 0.4 and x = 0.9 measured at 80K. Since when x = 0.4, the structure
obtained is a perovskite structure while that for x = 0.9 is a heterogeneous oxide
mixture of 0.05La
2
O
3
-0.9SrO/⅓Co
3
O
4
, the SrO-Co
3
O
4
interfacial phase that is present
-0.4
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
0.4
-15000 -10000 -5000 0 5000 10000 15000
Magnetic Field (Oe)
Magnetic Moment (emu/g)

SrO/Co
3
O
4
Co
3
O
4
Figure 4.10 Hysteresis loop of di-oxide composite SrO/Co
3
O
4
and pristine Co
3
O
4
measured at 80K
















99
in the heterogeneous oxide mixture could offer a smaller number of magnetic moments
despite being more resilient against demagnetisation by external magnetic field than
the perovskite phase. Similarly, by comparing the magnitudes of remanence of the
three composites, a decreasing trend with increase in x value with respect to each
temperature can be observed. This is in parallel with the reduction of the perovskite
phase content in the composites and is consistent with their XRD patterns.


Table 4.2 Ferromagnetic properties (at 25
o
C) and Co
3
O
4
domain size in
0.025La
2
O
3
-0.95SrO/⅓Co
3
O
4
composites from different precursors
pK values
Hydrocarboxylic acid
La

3+
Sr
2+
Co
2+

Coercivity
(Oe)
Remanence
(emu/g)

104
a

0.025
Citric acid
b
6.65 2.80 4.83
368.0 0.088
Malic acid 4.37 1.45 2.86 130.0 0.017
Lactic acid 2.27 0.53 1.38 153.5 0.008
Ethylenediaminetetraacetic
acid (EDTA)
16.34 8.80 16.31 37.3 0.002
Glycine 11.2 0.91 10.76

a.
The data in this row come from the binary complex oxide SrO/Co
3
O

4
;
b
pK values are
obtained from (Martell and Smith, 1977)








100

101
Table 4.3 Effects of SrO contents and temperature on remanence of the perovskite
solid solution and tri-oxide composite

Remanence (emu/g)
La
1-x
Sr
x
CoO
3-y
½(1-x)La
2
O
3

-xSrO/⅓Co
3
O
4
Measured
Temperature
(K)
x = 0.4 x = 0.5 x = 0.8 x = 0.9 x = 0.95
74 3.00 5.05 1.78 0.94 0.110
80 3.03 4.55 1.64 0.92 0.11
100 2.40 3.93 1.45 0.82 0.09
125 1.67 2.95 1.14 0.73 0.08
150 1.18 2.18 0.87 0.55 0.06
200 0.38 0.86 0.34 0.32 0.03
250 - - 0.026 0.014 0.02


4.3.3 The hydroxycaboxylic ligand and the interface-sustained ferromagnetism
As far as the interface-sustained ferromagnetism is concerned, an interesting question
is on the impact of the organic components used on the composite magnetic properties.
Table 4.2 compares the coercivity and remanence of the four composites with x = 0.95
that were obtained using four respective metal-containing hydrogels. These four gels
contain four different chelating ligands, they are: (1) citric acid/glycine; (2) malic
acid/glycine; (3) lactic acid/glycine; and (4) Ethylenediaminetetraacetic acid (EDTA).
For the first three cases, during the formation of metallo-organic gel, the hydroxyacid
in each group will compete with glycine to associate with Co
2+
and La
3+
ions since

glycine shows much milder affinity with Sr
2+
ion. From the formation constants (pK)
listed in Table 4.2, it is observed that the smaller the pK of a hydroxyacid with
strontium ion, the poorer are the coercivity and remanence of the resulting oxide
composite (Martell and Smith, 1977) . In the previous analysis, it is known that the
-8
-6
-4
-2
0
2
4
6
8
-15000 -10000 -5000 0 5000 10000 15000
Magnetic Field (Oe)
Magnetic Moment (emu/g)
x = 0.4
x = 0.9
Figure 4.11 Hysteresis loop of ½(1-x)La
2
O
3
-xSrO/⅓Co
3
O
4
with x = 0.4 and 0.9, measured at 80K


















102
induction at the interface of the Co
3
O
4
and the SrO domains is solely responsible for
the generation of ferromagnetism at 298K. Hence, interfacial contact between SrO-
Co
3
O
4
affects the extent of this stimulation. Therefore, if more Sr
2+
ions can be brought

into the chelating network in close proximity to Co
2+
ions, a greater degree of SrO-
Co
3
O
4
mixing can be achieved. In the case of using EDTA alone as the chelating
ligand, though EDTA possesses far stronger Sr
2+
-chelating capability than the above
three acids, this pK is still much smaller relative to its own chelating capabilities to
associate with the other two metal ions. As a result, the oxide composite from EDTA-
based precursor ends up with weaker coercivity and remanence.


4.4 Conclusion
This work explores the magnetic properties of a highly mixed ½(1-x)La
2
O
3
-
xSrO/⅓Co
3
O
4
tri-oxide system, which is prepared via pyrolysis of a metal-ion-
containing hydrogel and subsequent calcination to burn out the organic residue and to
convert CoO to spinel Co
3

O
4
. This precursor-to-ceramic pathway casts a considerable
interfacial feature. It is because of this, only at a high SrO content (x ≥ 0.8) and a
relatively low calcination temperature (ca. 600°C) could solid phase reaction of the
three oxide phases, which forms a homogeneous solid solution, be mainly prevented.
There is however certain extent of solid phase reaction occurring in the heterogeneous
composites with x = 0.8 and 0.9 which results in the formation of perovskite phase La
1-
α
Sr
α
CoO
3-
β
(
α
<< x) at the interfacial loci. In addition, the SrO-Co
3
O
4
interfacial
phase is prevalent in this high dispersion system. These two specific phases render the
composites ferromagnetism at low temperatures. When temperature is increased to
200K the interfacial phase still withholds weak coercivity and remanence values but
the perovskite phase becomes paramagnetic. As compared to the two pristine

103