The effect of specific surface area of TiO2 on the thermal decomposition of ammonium perchlorate

Article abstract of DOI:10.1007/s10973-009-0462-0

The thermal decomposition of ammonium perchlorate (AP) is considered to be the first step in the combustion of AP-based composite propellants. In this report, the effect of the specific surface area of titanium oxide (TiO ₂) catalysts on the thermal decomposition characteristics of AP was examined with a series of thermal analysis experiments. It was clear that the thermal decomposition temperature of AP decreased when the specific surface area of TiO₂ increased. It was also possible that TiO₂ influences the frequency factor of AP decomposition because there was no observable effect on the activation energy. Akademiai Kiado, Budapest, Hungary 2009.

Full text of DOI:10.1007/s10973-009-0462-0

J Therm Anal Calorim (2010) 99:27–31
DOI 10.1007/s10973-009-0462-0
The effect of specific surface area of TiO2 on the thermal
decomposition of ammonium perchlorate
Kaori Fujimura Æ Atsumi Miyake
Japan Symposium 2008
´
´
Ó Akademiai Kiado, Budapest, Hungary 2009
Abstract The thermal decomposition of ammonium
perchlorate (AP) is considered to be the first step in the
combustion of AP-based composite propellants. In this
report, the effect of the specific surface area of titanium
oxide (TiO2) catalysts on the thermal decomposition
characteristics of AP was examined with a series of thermal
analysis experiments. It was clear that the thermal
decomposition temperature of AP decreased when the
specific surface area of TiO2 increased. It was also possible
that TiO2 influences the frequency factor of AP decom-
position because there was no observable effect on the
activation energy.
control the burning rate of composite solid propellants used
to date include adding a combustion catalyst [1–10] and
controlling the particle size of the solid oxidizer [11, 12].
It is well known that the burning rate of these composite
propellants which consist of ammonium perchlorate (AP) is
enhanced when some transition metal oxides (TMOs) are
added to the compositions [1–3, 6, 7]. For example, ferric
oxide (Fe2O3), copper oxide (CuO or Cu2O), copper
chromate (CuCrO4), copper chromite (CuCr2O4), and
manganese oxide (MnO2) are very effective in enhancing
the burning rate. Although many researchers have proposed
various theories as to how TMOs enhance the burning rate,
the mechanism remains unclear. It also has been observed
that the specific surface area of the combustion catalysts
greatly affects the burning rate enhancement. However,
there have been few studies in which the effect of specific
surface area was confirmed experimentally. Recently,
nano-sized particles have become available at a relatively
low price, but there have been few reports of these nano-
sized particles applied to composite propellants as com-
bustion catalysts [8–10].
Keywords Ammonium perchlorate ꢀ
Specific surface area ꢀ Thermal decomposition ꢀ
Titanium oxide
Introduction
Burning rate and specific impulse are representative
parameters that describe the burning characteristics of solid
propellants. In order to design rocket motors, the basic
method to satisfy the required relationship between
impulse and combustion time of the rocket motor is the
adjustment of the burning area by tailoring the grain con-
figuration of the propellant. However, controlling the
burning rate more widely would be very advantageous in
designing rocket motors because the restriction on grain
configuration could be lessened. Representative methods to
In a previous report, the effect of the particle size of
ferric oxide (Fe2O3) catalysts on the thermal decomposition
of AP-based composite propellants and pure AP was
examined with a series of thermal analysis experiments [1].
It was clear that the thermal decomposition of the propel-
lant samples and pure AP was greatly enhanced by
reducing the size of Fe2O3 from the submicron to nano-
meter size. Recently, it has been reported that titanium
oxide (TiO2) nanoparticles enhanced the burning rate of
AP-based solid composite propellants [10]. In the present
study, the thermal decomposition characteristics of AP,
which has been considered to be the first step of AP-based
propellant decomposition, with and without TiO2 catalysts
were determined, and the effect of the specific surface area
K. Fujimura (&) ꢀ A. Miyake
Yokohama National University, 79-7 Tokiwadai, Hodogayaku,
Yokohama, Kanagawa 240-8501, Japan
e-mail: d07tf008@ynu.ac.jp
123

28
K. Fujimura, A. Miyake
of the TiO2 catalysts on the AP decomposition was
investigated by thermal analyses.
inert environment. Each sample was measured three times.
For both instruments, the Stare software (Mettler-Toledo K.
K.) was used to control the temperature program and obtain
thermograms for analysis.
Experimental
Materials
Results and discussion
The particle size of AP used in this study was about 10 lm.
Five types of commercial TiO2 particles were used as
catalysts and their specifications are listed in Table 1. The
specific surface areas of TiO2 in Table 1 were experi-
mentally confirmed by the BET method with an automatic
surface area analyzer, Betasorb model 4200 (Nikkiso Co.,
Ltd.).
Thermal analysis
The DSC curves of the samples at a heating rate of
10 °C min-1 are shown in Fig. 1. Two thermal events are
observed in each curve: a small endothermic peak near
240 °C and a large exothermic peak around 450 °C. The
peak temperatures were measured at least three times per
sample and then averaged. The endothermic peak near
240 °C is attributed to the phase transition of AP from the
orthorhombic to cubic form [13, 14]. The relationship
between the peak temperature Tp of the large exothermic
peak around 450 °C and the specific surface area of the
TiO2 catalyst is summarized in Fig. 2. The peak tempera-
tures of the samples including TiO2 were nearly equivalent
to that of pure AP, showing the peak temperature did not
depend on the surface area of the TiO2 catalyst. One
explanation for these results is that the reaction between
AP and TiO2 catalyst is depressed, because the sample pans
used in the DSC experiment were sealed completely,
compressing the reaction gas inside the pan.
In order to prepare the samples for thermal analysis, AP
and each TiO2 powder were heated separately at 120 °C for
3 h in a furnace to eliminate moisture. The designated
weights of AP and TiO2 were measured, and the samples
were thoroughly mixed with a wooden stick in a mortar.
The ratio of AP and TiO2 in the samples was 97:3.
Thermal analysis
Differential scanning calorimetry (DSC) measurements
were carried out with a DSC 822e differential scanning
calorimeter (Mettler-Toledo K. K.). The sample was placed
in an aluminum pan and then the pan was sealed to with-
stand an internal pressure of about 3 MPa. The weight of
the sample was about 1.0 mg. Dynamic scans were per-
formed at a heating rate of 10 °C min-1 in a temperature
range from 20 to 500 °C. During the measurement, nitro-
gen gas was purged throughout the system at 50 mL min-1
to maintain an inert environment.
The TG and DTG curves at a heating rate of 10 °C
min-1 are shown in Fig. 3. a is the conversion degree as
represented below [15]:
mðt ¼ 0Þ ꢁ mðtÞ
a ¼
ð1Þ
mðt ¼ 0Þ ꢁ mðt ! 1Þ
Thermogravimetric (TG) analyses and differential ther-
mogravimetry (DTG) measurements were carried out
with a TGA/SDTA 851e thermogravimetric analyzer
(Mettler-Toledo K. K.). The sample (about 3.0 mg) was
placed in an open-type alumina pan. The dynamic scans
were performed at a heating rate of 3, 5, 8, and
10 °C min-1, in a temperature range from 30 to 600 °C.
During the measurement, nitrogen gas was purged
throughout the system at 200 mL min-1 to maintain an
All DTG curves consisted of a small peak and a large
peak. It has been pointed out that AP decomposition occurs
in two steps: low temperature decomposition (LTD)
(\350 °C) and high temperature decomposition (HTD)
([350 °C) [16–18]. The first peak indicates the LTD of AP.
For every sample, the LTD reaction starts near 300 °C and
then shows a shoulder near 340 °C in the DTG curve. This
peak cannot be observed in the DSC curve. The reaction is
Table 1 Specifications of TiO2 particles
TiO2(IV)
Manufacturer
Particle size (from catalogs)/nm
Specific surface area (BET method)/m2 g-1
(A)
(B)
(C)
(D)
(E)
Wako Pure Chemical Industries, Ltd.
Kanto Chemical Co., Inc.
\5000
19.7–101.0
200
55
35
18
7
Kanto Chemical Co., Inc.
Wako Pure Chemical Industries, Ltd.
Kanto Chemical Co., Inc.
\5000
100–300
6
123

The effect of specific surface area of TiO2 on the thermal decomposition of ammonium perchlorate
29
nucleation followed by proton transfer as shown below
[16],
AP
AP+(A)
AP+(B)
AP+(C)
AP+(D)
AP+(E)
NH4ClO4ðsÞ ꢀ NH3ðgÞ þ HClO4ðgÞ
ð2Þ
From the TG results, it was concluded that the LTD
reaction was not affected by TiO2 catalysts.
The next large peak indicates the HTD of AP. Boldyrev
showed that the HTD starts from the dissociation of AP
into NH3 and HClO4 on the surface of AP, then secondary
reactions between the thermal decomposition products of
HClO4 and NH3 occur either on the surface of AP or in the
gas phase above the surface [16]. In the presence of TiO2
catalyst, it can be seen that both onset temperature and
peak temperature are lower than those of pure AP. The
relationship between the peak temperature of HTD of AP
and the specific surface area of TiO2 catalyst is summa-
rized in Fig. 4. The temperature shift of this peak increases
as the specific surface area of TiO2 gets larger and the plot
is found to be linear except for pure AP (the specific sur-
face area of TiO2 = 0). Therefore, it can be concluded that
this HTD reaction of AP takes place on the surface of the
TiO2 catalyst.
200
250
300
350
400
450
500
Temperature/°C
Fig. 1 DSC curves of pure AP and mixtures of AP with TiO2
catalysts (heating rate: 10 °C min-1
Kinetic analysis
)
Kinetic analyses were carried out by the Ozawa method,
which is known as a model-free approach [19]. It is also
known as an iso-conversional method. In order to deter-
mine the activation energy of HTD of AP using TG data, it
is assumed that each reaction follows the Arrhenius
equation,
470
460
450
k ¼ A expðꢁEa=RTÞ
ð3Þ
where k is the rate constant, Ea is the activation energy, R is
the gas constant, T is the absolute temperature, and A is the
frequency factor. g(a) can be defined as the integral
function of conversion degree as shown below,
a
ꢀ
ꢁ
Z
da
da
dt
gðaÞ ¼
¼ kt
*
¼ kfðaÞ
ð4Þ
fðaÞ
0
In this case, the thermal stability of the samples at
various heating rates can be analyzed using a Ozawa plot in
which log b versus 1/T shows a linear relationship, as
shown in the following equation:
20
40
60
0
Specific surface area/m2 g–1
Fig. 2 The relationship between the peak temperature Tp of the large
exothermic peak around 450 °C and the specific surface area of TiO2
catalyst
AEa
Ea
log b ¼ log
ꢁ 2:315 ꢁ 0:4567
ð5Þ
gðaÞ
RT
depressed by the pressurization from the sealing of the
sample pan in the DSC experiment, so that the peak may be
shifted to higher temperature and overlapped with the next
peak. Boldyrev proposed that this LTD reaction starts with
where b is the heating rate [19]. When log b is plotted
against 1/T for each of a = constant, a straight line is
obtained with a slope equal to -0.4567Ea/RT, and then Ea
can be calculated.
123

30
K. Fujimura, A. Miyake
Fig. 3 a TG curves of pure AP
and mixtures of AP with TiO2
catalysts. b DTG curves of pure
AP and mixtures of AP with
TiO2 catalysts (heating rate:
(a)
1
pure AP
AP+(A)
AP+(B)
AP+(C)
AP+(D)
AP+(E)
0.5
0
10 °C min-1
)
280
300
320
340
360
380
400
420
Temperature/°C
(b)
pure AP
AP+(A)
AP+(B)
AP+(C)
AP+(D)
AP+(E)
280
300
320
340
360
380
400
420
Temperature/°C
either the specific surface area of TiO2 or the existence of
TiO2.
410
400
390
The frequency factor A could not be determined, because
the method used above is the model-free approach and no
kinetic model is taken into account. Given the fact that the
TiO2 catalyst does not influence the activation energy but
decreases the reaction temperature with increasing specific
surface area, there is a possibility that TiO2 affects the fre-
quency factor. This warrants further study.
← pure AP
Conclusions
20
40
0
60
Specific surface area of TiO2/m2g–1
It was clear that the thermal decomposition of AP was
enhanced by increasing the specific surface area of TiO2. In
particular, the effect on the HTD of AP is apparent and
there is a linear relationship between the peak temperature
of HTD of AP in DTG curves and the specific surface area
of the TiO2 catalyst. It is also possible that the TiO2
influences the frequency factor of AP decomposition
because there was no observable effect on the activation
energy.
Fig. 4 The relationship between the peak temperature of HTD of AP
in DTG curves and the specific surface area of TiO2 catalyst
The activation energy for a = 0.5 which represents the
HTD reaction calculated by Ozawa plots and its correlation
coefficient (r) are summarized in Table 2. The difference
in the activation energy between pure AP and AP? various
TiO2 is negligible regardless of the specific surface area.
Therefore, we conclude that the specific surface area of the
TiO2 catalyst does not influence the activation energy of
the HTD reaction of AP. In other words, the energy
required for the HTD reaction to occur is not affected by
Acknowledgements This research was supported by the Technical
Research & Development Institute at the Ministry of Defense. We
would also like to acknowledge Professor Kohga of the National
Defense Academy for his assistance and advice on performing spe-
cific surface area measurements.
Table 2 Activation energy of HTD reaction of AP calculated with
the Ozawa method
Sample
Ea/kJ mol-1
r
References
Pure AP
124
115
123
139
133
123
0.957
0.929
0.970
0.978
0.965
0.919
1. Fujimura K, Miyake A. Sci Technol Energ Mater. 2008;69:149.
2. Krishnan S, Jeenu R. Combustion characteristics of AP/HTPB
propellants with burning rate modifiers. J Propuls Power. 1992;8:
748–55.
3. Chakravarthy SR, Price EW, Sigmant RK. Mechanism of burning
rate enhancement of composite solid propellants by ferric oxide. J
Propuls Power. 1997;13:471–80.
AP ? (A)
AP ? (B)
AP ? (C)
AP ? (D)
AP ? (E)
123

The effect of specific surface area of TiO2 on the thermal decomposition of ammonium perchlorate
31
4. Fong CW, Hamshere BL. The mechanism of burning rate catal-
ysis in. Composite HTPB-AP propellant combustion. Combust
Flame. 1986;65:61–9.
12. van der Heijden AEDM, Leeuwenburgh AB. HNF/HTPB pro-
pellants: influence of HNF particle size on ballistic properties.
Combust Flame. 2009;156:1359–64.
5. Fong CW, Hamshere BL. The mechanism of burning rate catalysis
incomposite propellants by transition-metal complexes. Combust
Flame. 1986;65:71–8.
13. Shen SH, Chen SI, Wu BH. The thermal decomposition of
ammonium perchlorate (AP) containing a burning-rate modifier.
Thermochim Acta. 1993;223:135–43.
6. Dubey BL, Nath N, Tripathi A, Tiwari N. Catalysed. combustion
of ammonium perchlorate, polystyrene and their composite pro-
pellants. Indian J Eng Mater Sci. 1994;1:341–9.
7. Pearson GS. Composite propellant catalysts: copper chromate
and chromite. Combust Flame. 1970;14:73–84.
8. Patil PR, Krishnamurthy VN, Joshi SS. Differential scanning
calorimetric study of HTPB based composite propellants in pres-
ence of nano ferric oxide. Propellants Explos Pyrotech. 2006;6:
442–6.
14. Rocco JAFF, Lima JES, Frutuoso AG, Iha K, Ionashiro M, Matos
JR, et al. Thermal degradation of a composite solid propellant
examined by DSC. J Therm Anal Calorim. 2004;75:551–7.
15. Longuet B, Gillard P. Experimental investigation on the hetero-
geneous kinetic process of the low thermal decomposition of
ammonium perchlorate particles. Propellants Explos Pyrotech.
2009;34:59–71.
16. Boldyrev VV. Thermal decomposition of ammonium perchlorate.
Thermochim Acta. 2006;443:1–36.
9. Li W, Cheng H. Cu-Cr-O nanocomposites: synthesis and char-
acterization as catalysts for solid state propellants. Solid State
Sci. 2007;9:750–5.
17. Halawy SA, Al-Shihry SS. Role of the acidic-basic characters of
some metal oxides in the pyrolysis of ammonium perchlorate.
J Therm Anal Calorim. 1999;55:833–40.
10. Small JL, Stephens MA, Deshpande S, Peterson EL, Seal S.
Proceedings of 20th International Colloquium on the Dynamics
of Explosions and Reactive Systems, Montreal, July 31–August
5, 2005.
11. Pivkina A, Frolov Yu, Ivanov D. Nanosized components of
energetic systems: structure, thermal behavior, and combustion.
Combust Expl Shock Waves. 2007;43:51–5.
18. Chen L-J, Li G-S, Qi P, Li L-P. Thermal decomposition of
ammonium perchlorate activated via addition of NiO nanocrys-
tals. J Therm Anal Calorim. 2008;92:765–9.
19. Ozawa T. Kinetic analysis of derivative curves in thermal anal-
ysis. J Therm Anal Calorim. 1970;2:301–24.
123