A comparative study of the effects of decomposition rate control and mechanical grinding on the thermal decomposition of aluminum hydroxide

Article abstract of DOI:10.1007/s10973-005-0830-3

Two different processes of the thermal decomposition of synthetic bayerite, i.e., the non-isothermal decomposition of mechanically ground sample in flowing N₂ and the controlled rate thermal decomposition of crystalline bayerite under vacuum, were investigated comparatively. In comparison with the conventional non-isothermal decomposition of crystalline bayerite in flowing N₂, the reaction temperature of the thermal decomposition was lowered by the individual effects of mechanical grinding of the sample and the reaction rate control. These decomposition processes indicated similar behavior characterized by the restricted changes of the specific surface area during the course of decomposition reaction and the formation of an amorphous alumina as the decomposition product. Different thermal behaviors were observed for those amorphous Al₂O₃ produced by the respective decomposition processes.

Full text of DOI:10.1007/s10973-005-0830-3

Journal of Thermal Analysis and Calorimetry, Vol. 81 (2005) 595–601
A COMPARATIVE STUDY OF THE EFFECTS OF DECOMPOSITION
RATE CONTROL AND MECHANICAL GRINDING ON THE THERMAL
DECOMPOSITION OF ALUMINUM HYDROXIDE
N. Koga*
Chemistry Laboratory, Department of Science Education, Graduate School of Education, Hiroshima University, 1-1-1
Kagamiyama, Higashi-Hiroshima 739-8524, Japan
Two different processes of the thermal decomposition of synthetic bayerite, i.e., the non-isothermal decomposition of mechanically
ground sample in flowing N2 and the controlled rate thermal decomposition of crystalline bayerite under vacuum, were investigated
comparatively. In comparison with the conventional non-isothermal decomposition of crystalline bayerite in flowing N2, the reaction
temperature of the thermal decomposition was lowered by the individual effects of mechanical grinding of the sample and the reaction
rate control. These decomposition processes indicated similar behavior characterized by the restricted changes of the specific surface
area during the course of decomposition reaction and the formation of an amorphous alumina as the decomposition product. Different
thermal behaviors were observed for those amorphous Al2O3 produced by the respective decomposition processes.
Keywords: amorphous aluminum oxide, controlled rate thermal analysis, mechanochemical effect, synthetic bayerite, thermal de-
composition
Introduction
surface area during the course of reaction, producing
an amorphous alumina as the decomposition product.
It was found that those observations are very similar
to the reported reaction behavior of the non-isother-
mal decomposition of amorphous aluminum hydrox-
ide obtained by the mechanical grinding of the crys-
talline bayerite [16]. A comparative study of the
effects of decomposition rate control under vacuum
and mechanical grinding of the sample on the reaction
pathway and kinetic behavior of the thermal decom-
position processes was carried out in the present
work, in order to clarify the reasons for the similari-
ties of the respective effects on the decomposition
processes. The thermal behaviors of the amorphous
alumina produced by the respective decomposition
processes were also compared.
Because the kinetic behavior and the properties of the
solid products of the thermal decomposition of solids
are influenced largely by the reaction atmosphere ap-
plied and generated by the reaction itself [1, 2], precise
control of the reaction atmosphere including the self-
generated reaction condition is desired for the material
synthesis through the thermal decomposition processes.
Application of the principle of controlled rate thermal
analysis (CRTA) [3–5] to the material synthesis through
the thermal decomposition processes is one of the possi-
ble solutions to control the reaction atmosphere includ-
ing the self-generated atmosphere, where the reaction
atmosphere during the whole course of the reaction can
be regulated precisely by controlling the decomposition
rate. Effects of the decomposition rate control on the ki-
netic behaviors and the properties of the solid products
have been examined as exemplified by various pro-
cesses of thermal decomposition [5–15], providing fun-
damental understandings on the usefulness of the CRTA
as an intelligent tool for controlling the process of
material synthesis.
Experimental
Sample preparation
Aluminum hydroxide was prepared by the procedure
reported previously [17]. Carbon dioxide was intro-
duced, at a rate of 100 cm3 min–1, into 0.9 M sodium
aluminate solution at room temperature until the pH
value of the solution being down to 12.3. The precipi-
tate produced during the carbonation process was
aged in mother solution with mechanical stirring for
12 h. The precipitate was filtered and washed with
In our previous work [15], we have examined the
effect of decomposition rate control under vacuum on
the reaction pathway of the thermal decomposition of
synthetic bayerite, β-Al(OH)3. The decomposition
process was characterized by the lowered reaction
temperature and a very limited change of the specific
*
nkoga@hiroshima-u.ac.jp
1388–6150/$20.00
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands
© 2005 Akadémiai Kiadó, Budapest

KOGA
Results and discussion
water, and dried in vacuum desiccator. The precipi-
tate was identified, as Al(OH)3 with crystal structure
of bayerite, by chemical analysis for Al3+, FTIR
spectroscopy, powder X-ray diffraction (XRD) and
TG-DTA.
Thermal decompisition processes
Figure 1 shows typical traces of reaction temperature
and residual pressure during the controlled rate thermal
decomposition of synthetic bayerite under vacuum.
Initiating the decomposition reaction, the residual
pressure in the reaction atmosphere increased due to
the evolution of water vapor as the gaseous product.
Through a feedback control of the reaction temperature
by monitoring the residual pressure, the residual pres-
sure during the course of reaction is controlled success-
fully at a constant value of 7·10–3 Pa. At the end of the
reaction, the reaction temperature increases rapidly to
Controlled rate decomposition
The sample was weighed into an alumina boat
(30 mm in length, 6 mm in width, and 4 mm in depth)
and was sat in a tube furnace. The pressure in the tube
furnace was reduced to 1.0·10–3 Pa by diffusion and
rotary pumps. Using an instrument of controlled rate
evolved gas detection (CREGD) [15], the sample was
heated up to 775 K by controlling the sample temper-
ature so as to maintain the pressure to be 7.0·10–3 Pa
during the course of thermal decomposition.
Mechanical grinding
Mechanical grinding of the sample was carried out in
line with the procedure reported by Tsuchida and
Ichikawa [16]. The sample of 5 g was weighed into an
alumina jar of 45 cm3, together with seven alumina
balls of 12 mm in diameter. Using a planetary mill
(Fritsch P-7), the sample was ground at a rotation
speed of 700 rpm for 30 h, where every one hour the
grinding was stopped to scrape the sample from the
balls and jars. The ground sample indicated no
distinguished XRD peak with the specific surface
area of 35.8 m2 g–1.
Fig. 1 Typical traces of reaction temperature and residual
pressure during the controlled rate thermal decomposi-
tion of crystalline bayerite (300 mg) at a controlled
pressure of 7.0 10–3 Pa
Measurements
Using an instrument of ULVAC DTG-9600, TG-DTA
curves for the as-prepared crystalline bayerite and the
ground sample were measured at a heating rate of
30 K min–1 under flowing N2 (100 cm3 min–1). The
changes of the XRD patterns during the thermal de-
composition of these samples and the subsequent
phase changes of the decomposition products were
traced by heating at a rate of 2 K min–1 using an instru-
ment of high temperature XRD (Rigaku RINT2200;
monochrome CuKα, 40 kV, 20 mA). The specific sur-
face areas of the samples decomposed partially by heat-
ing at a rate of 2 K min–1 in flowing N2 (100 cm3 min–1)
were measured by the single point method of BET us-
ing FlowSorbII-2300 (Micromeritics). DSC measure-
ments for the decomposition products, obtained thro-
ugh the non-isothermal decomposition of ground
sample in flowing N2 and controlled rate thermal de-
composition of crystalline bayerite under vacuum by
heating up to 775 K, were carried out using a
Rigaku DSC8160 at various heating rates under flow-
ing N2 (50 cm3 min–1).
Fig. 2 Typical TG-DTA curves for the non-isothermal decom-
positions of crystalline bayerite and mechanically
ground sample (ca. 20 mg), together with the reaction
temperature of controlled rate thermal decomposition
of synthetic bayerite (300 mg) by CREGD at a con-
trolled pressure of 7.0 10–3 Pa
596
J. Therm. Anal. Cal., 81, 2005

THERMAL DECOMPOSITION OF ALUMINUM HYDROXIDE
the limit temperature of measurement and the residual
pressure decreases to the basal pressure of 1·10–3 Pa.
Because only water vapor is evolved through the
present reaction as the gaseous product, the evolution
rate of water vapor is also regulated at a constant rate.
It has already been reported in our previous study [15]
that an amorphous Al2O3 is produced as the decompo-
sition product of the conrolled rate thermal decompo-
sition of synthetic bayerite under vacuum.
Figure 2 compares TG-DTA curves for the ther-
mal decompositions of crystalline bayerite and
ground sample. As has been reported previously [17],
the thermal decomposition of crystalline bayerite ini-
tiates at around 450 K and proceeds in two distin-
guished reaction steps via an intermediate product at
around 600 K. The gradual mass loss is observed until
950 K. On the other hand, the thermal decomposition
of ground sample takes place in a wide temperature
range from room temperature to 950 K with smooth
mass-loss trace and a broad endothermic DTA peak.
At the higher temperature region, i.e., >950 K, the de-
composition products indicate apparently different
thermal behaviors. The synthetic bayerite indicate no
distinguished thermal effect in the DTA curve at the
higher temperature region. As for the ground sample,
two exothermic DTA peaks are observed at around
1100 and 1350 K. Unexpectedly, apparent mass loss
is observed at the temperature range between these
two exothermic DTA peaks. By the first identification
of the evolved gases at this stage by means of TG-MS,
only water vapor was detected during the temperature
region. The similar mass-loss behavior of the ground
bayerite at the same temperature region has also been
reported by Tsuchida and Ichikawa [16]. For a com-
parison, the trace of reaction temperature measured
by CREGD under vacuum was also shown in Fig. 2.
The controlled rate decomposition of synthetic
bayerite under vacuum takes place at the lower tem-
perature region from 370 to 600 K, even comparing
with that for the thermal decomposition of the ground
sample under linear heating. The respective influences
of the lowered reaction temperature, induced by the
mechanical grinding and controlled rate decomposi-
tion under vacuum, on the decomposition kinetics and
the thermal behavior of the decomposition products
were compared throughout the present study.
Fig. 3 Changes of XRD patterns during heating a – crystalline
bayerite, b – ground sample, and c – amorphous de-
composition product obtained by the controlled rate
thermal decomposition of bayerite
guished XRD peak can be found at the early stage of
decomposition until 500 K, where the decomposition
reaction is already achieved to the fractional conver-
sion of ca. 85%. At the higher temperature region cor-
responding to the temperature range of non-isothermal
decomposition of crystalline bayerite, i.e. 500–900 K,
a slight but detectable diffraction peak of η-Al2O3 at
45.7° can be observed. Accordingly, the decomposi-
tion product of amorphous Al(OH)3 at 950 K is as-
sumed as the amorphous Al2O3 mixed partially with
poorly crystalline η-Al2O3. It has been reported [15]
that, through the controlled rate decomposition of
crystalline bayerite under vacuum, the diffraction
peaks ascribed to β-Al(OH)3 decrease the intensities
monotonously without indicating any other diffrac-
tion peaks of crystalline product, as can be seen from
the XRD pattern in Fig. 3c at 775 K.
Variations in the specific surface areas of the sam-
ples during the thermal decomposition were compared
in Fig. 4. During the non-isothermal decomposition of
Figure 3 compares the changes of the XRD pat-
terns during heating the synthetic bayerite and ground
sample, together with those for the decomposition
product of controlled rate decomposition under vac-
uum (heated up to 775 K). The crystalline bayerite de-
composed in two distinguished reaction steps via the
intermediate compound of boehmite AlO(OH) pro-
ducing finally η-Al2O3 at around 950 K [17]. As for
the amorphous Al(OH)3 of ground sample, no distin-
Fig. 4 Changes of specific surface area during the non-isother-
mal decompositions of crystalline bayerite and its
ground sample in flowing N2 and during the controlled
rate thermal decomposition of crystalline bayerite at a
controlled pressure of 7.0 10–3 Pa
J. Therm. Anal. Cal., 81, 2005
597

KOGA
crystalline bayerite, the specific surface area increases
as reaction proceeds and indicates a maximum at
around the fractional decomposition of 80%, which
corresponds to the formation of intermediate product,
AlO(OH). The remarkable increase in the specific sur-
face area during the thermal decomposition has been
explained by the formations of micropores of the prod-
uct [16]. In contrast, no distinguished variation of the
specific surface area can be observed during the ther-
mal decomposition of ground sample, i.e., amorphous
Al(OH)3. The similar behavior can be seen for the con-
trolled rate thermal decomposition of crystalline
bayerite under vacuum. Two different interpretations
of the phenomena have been made from the view
points of a mechanochemical effect and a significant
feature of the controlled rate decomposition under vac-
uum. From a mechanochemical viewpoint, Tsuchida
and Ichikawa [16] assumed a probable facilitation of
sintering during the course of decomposition even at the
lowered reaction temperature due to the energies accu-
mulated by the sample during grinding. As for the iden-
tical phenomena observed for the controlled rate ther-
mal decomposition of crystalline gibbsite, α-Al(OH)3,
under vacuum, Rouquerol and Ganteaume [6] gave an
interpretation concerning the mechanism of micropore
formation during the thermal decomposition where the
final micropore diameter depends on the mobilities of
the Al3+ and H+ ions. Because the mobilities of these
ions at the reaction interface decrease by lowering the
reaction temperature, size of the micropore is expected
to be reduced under the controlled rate thermal decom-
position in vacuum. The latter interpretation is also ac-
count for the phenomena observed for the thermal de-
composition of ground sample, because the mobilities
of the ions at the reaction interface seems to be de-
creased with the lowered reaction temperature.
Fig. 5 A series of kinetic rate data measured at different heat-
ing rates for the non-isothermal decomposition of
ground sample in flowing N2
by Friedman method [18–20], according to the fol-
lowing equation.
dα
dt
Ea
⎛
⎜
⎝
⎞
⎟
⎠
ln
= ln[Af(α)]–
(1)
RT
where A and f(α) is the pre-exponential factor of the
Arrhenius equation and the kinetic model function in
differential form. Other symbols have their usual
meanings. By plotting ln(dα/dt) vs. reciprocal tem-
perature T –1 using the data points read from the series
of kinetic rate data at a selected α, the value of Ea is
evaluated from the slope of the Friedman plot. Fig-
ure 7 shows the values of Ea at various α. Quite differ-
ent behaviors of the α-dependences of Ea values were
observed between the processes under comparison.
The Ea values for the controlled rate thermal decom-
position of crystalline bayerite under vacuum indicate
a constant value of 44 2 kJ mol–1 in a wide range of
reaction, 0.05≤α≤0.80. By contrast, for the non-iso-
thermal decomposition of ground sample in flowing
Decomposition kinetics
For characterizing the apparent kinetic behaviors of
the thermal decomposition processes, the respective
series of kinetic rate data for the non-isothermal de-
composition of ground sample in flowing N2 and for
the controlled rate thermal decomposition of crystal-
line bayerite under vacuum were collected. Figure 5
shows a series of kinetic rate data for the non-isother-
mal decomposition of ground sample under flowing
N2 at various heating rates. As shown in Fig. 6, a se-
ries of the kinetic rate data for the controlled rate ther-
mal decomposition of crystalline bayerite under vac-
uum were measured by changing the sample masses
under the conditions otherwise identical, i.e., basal
pressure of 1.0·10–3 Pa and controlled pressure of
7.0·10–3 Pa. The value of apparent activation energy
Ea at a selected fractional reaction α was calculated
Fig. 6 A series of kinetic rate data measured using different
sample masses for the controlled rate decomposition of
crystalline bayerite at a controlled pressure of
7.0 10–3 Pa
598
J. Therm. Anal. Cal., 81, 2005

THERMAL DECOMPOSITION OF ALUMINUM HYDROXIDE
Fig. 7 Apparent values of Ea at different α evaluated for the
Fig. 8 The experimental master plot of (dα/dθ)/(dα/dθ)0.5 vs. α
for the controlled rate decomposition of crystalline
bayerite under vacuum, together with theoretical mas-
ter plots drawn by assuming the R1.1 and A1.1 laws
non-isothermal decomposition of ground sample in
flowing N2 and for the controlled rate decomposition
(CRD) of crystalline bayerite at a controlled pressure
of 7.0 10–3 Pa
master plot of simulated rate data at infinite temperature,
(dα/dθ)/(dα/dθ)0.5 vs. α [23, 24], for the controlled rate
thermal decomposition of crystalline bayerite in vac-
uum. The experimental master curve is decelerating and
convex slightly in the shape. According to Eq. 2, the ap-
parent fitting of the formal kinetic model function was
compared by plotting dα/dθ vs. various f(α). Rather
good linear fittings were observed when using an
Avrami-Erofeev law (Am), f(α) =m(1–α)[–ln(1–α)]1–1/m
with m=1.1, and a phase boundary controlled law
(Rn), f(α)=n(1–α)1–1/n with n=1.1, that correspond
very closely to the first order and zero order equa-
tions, respectively. As was shown in Fig. 8, the appar-
ent fitting to those rate laws can be confirmed by com-
paring the experimental master plot with the theoreti-
cal curves of f(α)/f(0.5) vs. α drawn by assuming R1.1
and A1.1 laws. The apparent values of A evaluated from
the slopes of (dα/dθ) vs. f(α) plots were 24.3 and
5.1 s–1 for the R1.1 and A1.1 laws, respectively.
N2, the Ea values increase gradually from about 70 to
140 kJ mol–1 as reaction advances. The observed vari-
ation of the Ea values during the course of non-iso-
thermal decomposition of ground sample implies the
possible change of the reaction mechanism depending
on the reaction temperature distributed widely from
room temperature to 950 K. It has been shown by
27Al MASS NMR for the effect of mechanical grind-
ing of gibbsite [21] that the originally octahedral co-
ordination structure of aluminum ions and layered
structure of hydroxyl groups are destructed by the me-
chanical grinding by showing partially the lower co-
ordination in aluminum ions, i.e., tetrahedral and
5-fold coordination structures. Similar structural dis-
tribution induced by mechanical grinding, expected
also in the ground sample of bayerite, accounts for the
widened range of reaction temperature and systematic
variation in the decomposition mechanism. Because
direct application of the fundamental kinetic equa-
tion, based on an invariant reaction mechanism dur-
ing the course of reaction, to such complicated pro-
cess with systematic variation in the Ea values as
reaction proceeds is not allowed, the further detailed
kinetic characterization for the non-isothermal de-
composition of ground sample in flowing N2 was dis-
couraged at present.
Crystallization of amorphous alumina
Figure 9 compares typical DSC curves for the amor-
phous decomposition products obtained through the
non-isothermal decomposition of ground sample in
flowing N2 and controlled rate thermal decomposition
of crystalline bayerite under vacuum. Both the amor-
phous decomposition products indicate exothermic
peaks at the same temperature range of 900–1150 K.
By comparing with the changes of XRD patterns dur-
ing heating the samples shown in Fig. 3, these exo-
thermic peaks observed for both the samples are as-
cribed to the crystallization of η-Al2O3 from amorp-
hous Al2O3. The enthalpy changes during the crystal-
lization of η-Al2O3 differ largely between the dehy-
drated samples, i.e. –18.8 0.7 and –31.4 1.1 kJ
(mol Al2O3)–1 for the amorphous decomposition prod-
ucts obtained from the non-isothermal decomposition
By taking the constant value of Ea observed in a
wide range of α as the evidence of the applicability of
the fundamental kinetic equation, a simulated kinetic
curve at infinite temperature was drawn for the con-
trolled rate thermal decomposition of crystalline bay-
erite in vacuum by assuming the following equation:
dα dα
dθ dt
E
a
⎛
⎜
⎝
⎞
⎟
⎠
exp
=
= Af(α)
(2)
RT
where θ is the generalized time proposed originally
by Ozawa [22]. Figure 8 shows the experimental
J. Therm. Anal. Cal., 81, 2005
599

KOGA
Fig. 9 Typical DSC traces for the amorphous decomposition
products produced through the non-isothermal decom-
position of the ground sample in flowing N2 and the
controlled rate decomposition (CRD) of crystalline
bayerite under vacuum
Fig. 11 The experimental master plot of (dα/dθ)/(dα/dθ)0.5
vs. α for the crystallization process of η-Al2O3
identical. From the slope of the Kissinger plots, the
values of Ea were calculated as 481 24 and 488 16 kJ
mol–1 for the crystallization of amorphous decomposi-
tion products obtained from the non-isothermal de-
composition of ground sample in flowing N2 and con-
trolled rate thermal decomposition of crystalline
bayerite under vacuum, respectively. Using the values
of Ea, experimental master curves, (dα/dθ)/(dα/dθ)0.5
vs. α were drawn as shown in Fig. 11. Both the experi-
mental master curves are concave in the shape, indicat-
ing the overall kinetic behavior controlled by diffusion
of chemical species [23, 24].
Within the temperature region of the present
DSC measurements, an alternative exothermic peak at
the higher temperature was observed only for the de-
composition product of the ground sample at around
1300–1400 K, Fig. 9. The secondary exothermic peak
corresponds to the transformation of η-Al2O3 to
α-Al2O3, as can be seen from the changes of XRD
patterns shown in Fig. 3b. As for the decomposition
product of controlled rate decomposition, the diffrac-
tion peaks of α-Al2O3 is observed at around 1475 K,
Fig. 3c. Because the formation of α-Al2O3 during lin-
ear heating of crystalline bayerite can not be detected
clearly even heating up to 1475 K, Fig. 3a, it is appar-
ent that the formation temperature of α-Al2O3 is low-
ered by both the effects of mechanical grinding of the
sample and the controlled rate decomposition under
vacuum. As is deduced from the differences in the in-
tensities and half-width of the diffraction peaks of
η-Al2O3, the variation in the formation temperature of
α-Al2O3 is likely due to the different crystallinities of
η-Al2O3 produced through different reaction pathways.
of ground sample in flowing N2 and controlled rate
thermal decomposition of crystalline bayerite under
vacuum, respectively. Partial crystallization of
η-Al2O3 during the non-isothermal decomposition of
ground sample is assigned to one of the reasons for
the differences of the enthalpy changes.
On the basis of a series of DSC traces at different
heating rates β, the apparent values of Ea for the over-
all crystallization processes were determined by the
Kissinger plots [25] according to
⎛
⎞
β
AR
Ea
Ea
ln
=ln – f ' α
−
(3)
⎜
⎟
⎟
p
( )
Tp2
RTp
⎜
⎝
⎠
where subscript p indicates the peak maximum. Be-
cause the logarithmic term in the right hand side of
Eq. 3 can be approximated as constant, the value of Ea
can be calculated from the slope of the plot of
2
ln(β/Tp ) vs. Tp–1. Figure 10 compares the Kissinger
plots for the overall crystallization processes of the
amorphous decomposition products. It is apparent
that the slopes of the Kissinger plots for the respective
crystallization processes of η-Al2O3 are practically
Conclusions
The temperature range of the thermal decomposition
of synthetic bayerite is apparently lowered by the re-
Fig. 10 The Kissinger plots for the crystallization of η-Al2O3
600
J. Therm. Anal. Cal., 81, 2005

THERMAL DECOMPOSITION OF ALUMINUM HYDROXIDE
5 O. T. SÝrensen and J. Rouquerol, Sample Controlled Ther-
spective effects of mechanical grinding of the sample
and the reaction rate control under vacuum. Both the
decomposition processes, i.e. the non-isothermal de-
composition of ground sample in flowing N2 and the
controlled rate thermal decomposition of crystalline
bayerite under vacuum, are characterized by the very
limited change of the specific surface area of the sam-
ple during the course of the thermal decomposition
and by the formation of amorphous Al2O3 as the de-
composition product. These decomposition pro-
cesses, however, indicated different kinetic behav-
iors. The controlled rate decomposition of crystalline
bayerite under vacuum was characterized kinetically
by a monotonously decelerate process with the con-
stant value of Ea=44 2 kJ mol–1, whereas the values
of Ea for the non-isothermal decomposition of ground
sample increased gradually from 70 to 140 kJ mol–1 as
reaction advances due to the possible variation of re-
action mechanism.
The decomposition products, amorphous Al2O3,
produced by the respective decomposition processes
were crystallized to η-Al2O3 at around 900–1150 K,
with a comparable kinetic behavior of diffusion con-
trolled transformation with Ea≈480 kJ mol–1. A differ-
ence is again appeared in the transformation tempera-
ture of η-Al2O3 to α-Al2O3, where the transformation
of η-Al2O3 produced through heating the ground sam-
ple took place at the lower temperature by about 150 K.
mal Analysis, Kluwer, Dordrecht, 2003.
6 J. Rouquerol and M. Ganteaume, J. Thermal Anal.,
11 (1977) 201.
7 N. Koga and J. M. Criado, Int. J. Chem. Kinet.,
30 (1998) 737.
8 N. Koga, J. M. Criado and H. Tanaka, Thermochim. Acta,
340/341 (1999) 387.
9 L. A. Perez-Maqueda, J. M. Criado, C. Real, J. žubrt and
J. Bohacek, J. Mater. Chem., 9 (1999) 1839.
10 M. D. Alcala, F. J. Gotor, L. A. Perez-Maqueda, C. Real,
M. J. Dianez and J. M. Criado, J. Therm. Anal. Cal.,
56 (1999) 1447.
11 G. S. Chopra, C. Real, M. D. Alcala,
L. A. Perez-Maqueda, J. žubrt and J. M. Criado, Chem.
Mater., 11 (1999) 1128.
12 N. Koga, J. M. Criado and H. Tanaka, J. Therm. Anal.
Cal., 60 (2000) 943.
13 J. M. Criado, J. Therm. Anal. Cal., 72 (2003) 1097.
14 E. A. Fesenko, P. A. Barnes, G. M. B. Parkes,
D. R. Brown and N. Naderi, J. Therm. Anal. Cal., 72
(2003) 1103.
15 N. Koga and S. Yamada, Solid State Ionics, 172 (2004) 253.
16 T. Tsuchida and N. Ichikawa, React. Solids, 7 (1989) 207.
17 N. Koga, T. Fukagawa and H. Tanaka, J. Therm. Anal. Cal.,
64 (2001) 965.
18 H. L. Friedman, J. Polym. Sci., Part C, 6 (1964) 183.
19 T. Ozawa, J. Thermal Anal., 31 (1986) 547.
20 N. Koga, Thermochim. Acta, 258 (1995) 145.
21 K. J. D. MacKenzie, J. Temuujin and K. Okada,
Thermochim. Acta, 327 (1999) 103.
22 T. Ozawa, Thermochim. Acta, 100 (1986) 109.
23 F. J. Gotor, J. M. Criado, J. Malek and N. Koga, J. Phys.
Chem. A, 104 (2000) 10777.
References
24 J. M. Criado, L. A. Perez-Maqueda, F. J. Gotor, J. Malek
and N. Koga, J. Therm. Anal. Cal., 72 (2003) 901.
25 H. E. Kissinger, Anal. Chem., 29 (1957) 1702.
1 A. K. Galway and M. E. Brown, Thermal Decomposition
of Ionic Solids, Elsevier, Amsterdam 1999.
2 N. Koga and H. Tanaka, Thermochim. Acta, 288 (2002) 41.
3 J. Rouquerol, Thermochim. Acta, 144 (1989) 209.
4 J. Rouquerol, Thermochim. Acta, 300 (1997) 247.
DOI: 10.1007/s10973-005-7088-7
J. Therm. Anal. Cal., 81, 2005
601