Defence Technology 13 (2017) 22e32

Contents lists available at ScienceDirect

Defence Technology
journal homepage: www.elsevier.com/locate/dt

Thermal decomposition and kinetics of plastic bonded explosives

based on mixture of HMX and TATB with polymer matrices
Arjun Singh a, b, *, Tirupati C. Sharma a, Mahesh Kumar a, Jaspreet Kaur Narang a,
Prateek Kishore a, Alok Srivastava b
a
Terminal Ballistics Research Laboratory, Defence Research & Development Organisation, Ministry of Defence, Explosive Division, Sector 30, Chandigarh
(UT), 160030, India
b
Department of Chemistry & Centre for Advanced Studies in Chemistry, Panjab University, Chandigarh, 160 014, India

a r t i c l e i n f o a b s t r a c t

Article history: This work describes thermal decomposition behaviour of plastic bonded explosives (PBXs) based on
Received 21 August 2016 mixture of l,3,5,7-tetranitro- 1,3,5,7-tetrazocane (HMX) and 2,4,6- triamino-1,3,5-trinitrobenzene (TATB)
Received in revised form with Viton A as polymer binder. Thermal decomposition of PBXs was undertaken by applying simulta-
13 November 2016
neous thermal analysis (STA) and differential scanning calorimetry (DSC) to investigate influence of the
Accepted 28 November 2016
Available online 23 December 2016
HMX amount on thermal behavior and its kinetics. Thermogravimetric analysis (TGA) indicated that the
thermal decomposition of PBXs based on mixture of HMX and TATB was occurred in a three-steps. The
first step was mainly due to decomposition of HMX. The second step was ascribed due to decomposition
Keywords:
Plastic bonded explosives
of TATB, while the third step was occurred due to decomposition of the polymer matrices. The thermal
Thermogravimetric analysis decomposition % was increased with increasing HMX amount. The kinetics related to thermal decom-
Differential scanning calorimeter position were investigated under non-isothermal for a single heating rate measurement. The variation in
Thermal decomposition the activation energy of PBXs based on mixture of HMX and TATB was observed with varying the HMX
Kinetics amount. The kinetics from the results of TGA data at various heating rates under non-isothermal con-
ditions were also calculated by FlynneWalleOzawa (FWO) and Kissinger-Akahira-Sunose (KAS)
methods. The activation energies calculated by employing FWO method were very close to those ob-
tained by KAS method. The mean activation energy calculated by FWO and KAS methods was also a good
agreement with the activation energy obtained from single heating rate measurement in the first step
decomposition.
© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction homogeneity, better dimensional integrity and higher thermal

stability than TNT based melt cast compositions. The role of poly-
Plastic bonded explosives (PBXs) based on l,3,5,7-tetranitro- meric matrices is to minimise their sensitivity, improved mechan-
1,3,5,7-tetrazocane (HMX) or 2,4,6-triamino-1,3,5- trinitrobenzene ical and high thermal properties [17].
(TATB) with various polymer matrices have been formulated in the TATB based PBXs formulations such as LX-17 and PBX 9502
literature [1e16]. Polymer matrices; Viton A; a vinylidene fluoride - [18e21] have been developed for nuclear bomb, missiles and space
hexafluoropropylene copolymer, Kel-F 800; a vinylidene fluoride - applications. TATB has a high thermal stability, insensitive in terms
chlorotrifluoroethylene copolymer, polytetrafluoroethylene, Estane of impact and friction but poor performance. The performance of
5703; a poly(ester urethane) block copolymer etc. are mainly used formulations has been enhanced either to use new explosive
for PBXs formulation due to their higher loading density, molecule having better performance than TATB or admixture of
high energetic materials which has a high performance and
comparatively less sensitive to ensure the safety parameters
[22e24]. Therefore, PBXs based on HMX and TATB have been
* Corresponding author. Terminal Ballistics Research Laboratory, Defence formulated with polymer matrices; Estane, Viton A and Kel-F 800
Research & Development Organisation, Ministry of Defence, Explosive Division,
Sector 30, Chandigarh (UT), 160030, India.
to some extent compromise with insensitivity [25,26]. These for-
E-mail address: arjunsngh@yahoo.com (A. Singh). mulations have been characterized in terms of density, detonation
Peer review under responsibility of China Ordnance Society. velocity, ignition temperature and other explosive properties which

http://dx.doi.org/10.1016/j.dt.2016.11.005
2214-9147/© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
 A. Singh et al. / Defence Technology 13 (2017) 22e32 23

are covered under few reports and in paper also [27e37]. The 90
C to evaporate the solvent from the solution. Then, the polymer
thermal decomposition behaviour and kinetics are very important was precipitated on the surface of energetic materials. Finally,
because it ensures safety parameters during handling, processing, filtered and dried to get PBXs. In this way, PBXs from a mixture of
production and storage [38e40]. PBXs based on HMX or TATB have HMX and TATB at varying mass ratios (10:80, 20:70, 30:60, 40:50,
been extensively investigated for the thermal decomposition 50:40, 60:30, 70:20 and 80:10 respectively) with 10 weight percent
behaviour and its kinetics [41e44] by means of non-isothermal of Viton A were formulated and designated as HT1080, HT2070,
thermogravimetry (TGA) and differential scanning calorimetry HT3060, HT4050, HT5040, HT6030, HT7020 and HT8010 respec-
(DSC). The kinetic parameters of HMX based PBXs with Viton A tively. For comparison, HMX/Viton A and TATB/Viton A formula-
[45], C4 [46], Formex [47] and Semtex [48,49] have been investi- tions (90:10 by weight percent of HMX or TATB with Viton A) were
gated and published. Brunham and Weese [50] have investigated also formulated.
the kinetics of PBXs with three endothermic binders; Estane 5703,
Viton A and Kel-F 800 by performing TG measurements at different 2.3. Characterization of PBXs
heating rates, exhibited that Viton A and Kel-F 800 were more
thermally stable than HMX and TATB. Craig et al. [51] have been Non-isothermal thermo-gravimetric (TG) analyses of PBXs were
studied thermal behaviour of PBXs based on HMX or TATB with carried out by Simultaneous Thermal Analyzer (STA), manufactured
same endothermic binders, exhibited longer times to thermal ex- and supplied by METTLER TOLEDO, Model Mettler Toledo 851e. The
plosion than those of pure HMX or TATB in the one-dimensional samples were subjected to heating from 25
C to 600
C at heating
time to explosion and in other thermal experiments [52,53]. rate 10
C/min under nitrogen atmosphere. HT4050 sample was
Moreover, the decomposition kinetic models with different also subjected to heating from 25
C to 600
C at different heating
polymeric matrices have been published as both kinetic parameters rates 10, 20 and 30
C/min under nitrogen atmosphere.
and reaction models are the key factor for the prediction of the DSC analyses were carried out in sealed standard 40 mL
thermal hazard properties. It has been reported that the effect of aluminum crucible using a Differential Scanning Calorimeter (DSC),
the polymer matrices on the decomposition mechanism has been manufactured and supplied by METTLER TOLEDO, Model DSC 823e.
significantly observed and resulting in very different reaction 4.0 ± 0.5 mg weight of the sample was used for each experiment.
models. Tarver and Tran have also been measured the decompo- The samples were scanned from 25
C to 600
C for each experi-
sition models to predict of explosion and the locations within the ment at heating rate 10
C/min under nitrogen atmosphere. TG and
explosive charges [54]. However, the thermal decomposition DSC analyses of pure HMX, TATB and Viton A were also monitored
behaviour and kinetics of PBXs based on mixture of TATB and HMX under similar conditions.
with Viton A are less addressed in an open literature. In our pre-
vious study [55], the mechanical and explosive properties of PBXs 3. Results and discussion
based on mixture of HMX and TATB have been investigated and
published. 3.1. TG/DTG studies
In the present work, the thermal decomposition behaviour of
PBXs based on mixture of HMX and TATB with Viton A as polymeric Fig. 1 shows TG thermograms of HMX, TATB, Viton A and PBXs
matrices is studied by employing Simultaneous Thermal Analysis namely HT1080, HT2070, HT3060, HT4050, HT5040, HT6030,
(STA) and DSC. The kinetic parameters namely the activation en- HT7020, HT8010, HMX/Viton A and TATB/Viton A. TG thermogram
ergy and pre-exponential factor are determined under non- of HMX shows that 99.9% weight loss occurs in a single step as a
isothermal conditions for a single heating rate measurement. The sharp thermal decomposition strips at 280
C [18,56]. TG thermo-
three dynamic TGA measurements at different heating rates are gram of TATB shows 87.7% weight loss in a single step over a wide
also used to investigate activation energy as a function of reaction range of the temperature (300
C e450
C). The main weight loss
conversion through non-isothermal condition by employing occurs within a range of 350
C e385
C due to the thermal
FlynneWalleOzawa (FWO) and Kissinger-Akahira-Sunose (KAS) decomposition of TATB [52]. TG thermogram of Viton A shows that
methods. weight loss occurs in a single step at a high temperature within a
range of 458
C e504
C due to the thermal decomposition of
2. Experimental polymer matrices.
TG thermograms of PBXs; HT1080, HT2070, HT3060, HT4050,
2.1. Materials HT5040, HT6030, HT7020 and HT8010 show that the weight loss
occurs in a three steps, but the thermal stability causes a decay of a
HMX which is prepared in-house by nitration of hexamine by first one. The weight loss occurs in the first step at 278
Ce287
C
Bechmann reaction was used for PBXs formulation. Other ingre- due to thermal decomposition of HMX; while weight loss occurs in
dient TATB obtained from High Energy Material Research Labora- a second step in a range of 307
Ce374
C due to the thermal
tory Pune, India was used as received. Viton A; copolymer of decomposition TATB. The weight loss occurs in a third step in a
vinylidene fluoride and hexafluoropropylene manufactured by 3M range of 456
Ce498
C due to thermal decomposition of polymer
DuPont Corporation was served as a polymer binder as received. matrices. HMX/Viton A and TATB/Viton A samples also show two
Methyl ethyl ketone procured from SD Chem. Ltd. Pvt. India was steps weight loss; the weight loss in the first step is due to
used as received. decomposition of the HMX or TATB, whereas weight loss in the
second step is due to decomposition of polymer matrices. Thermal
2.2. Preparation of PBXs decomposition % obtained at different stages during decomposition
process from TG thermograms is summarized in Table 1.
PBXs formulations were prepared from a mixture of TATB and As far as thermal stability is concerned, it is reported that small
HMX with Viton A by slurry coating process as described in our weight loss of a materials at a certain temperature, more is ther-
previous work [55]. In this process, a mixture of HMX and TATB mally stable. The results indicate that 9e72% weight loss of PBXs
was dispersed with 3000 ml of a water with continuous stirring at based on mixture of HMX and TATB is occurred in the first step due
60
C. Then, polymer solution prepared in a methyl ethyl ketone to thermal decomposition of HMX. It is also observed that the
was added to the above mixture. The temperature was raised to decomposition % in the first step is increased with increasing HMX
24 A. Singh et al. / Defence Technology 13 (2017) 22e32

Fig. 1. TG thermograms of HMX, TATB, Viton A and PBXs obtained at heating rate 10
C/min under nitrogen atmosphere.

Table 1
TG data of PBXs along with HMX, TATB and Viton A obtained from TG thermograms at heating rate 10
C/min under nitrogen atmosphere.

Sample Multisteps thermal decomposition Decomposition

SDTa FDTb SDTa FDTb SDTa FDTb /%

(270
C e290
C) (290
C e450
C) (450
C e550
C)

/wt.% /wt.% /wt.%

HT1080 7.5 71.1 12.5 91.1

HT2070 13.4 67.5 11.2 92.2
HT3060 17.5 63.6 11.7 92.8
HT4050 40.4 44.7 11.2 96.4
HT5040 54.1 31.0 11.3 96.5
HT6030 69.9 15.8 10.6 96.6
HT7020 74.9 11.9 10.7 97.4
HT8010 79.1 8.9 8.1 97.8
HMX/Viton A 87.5 1.7 10.1 99.4
TATB/Viton A 0.5 73.2 10.7 83.7
HMX 99.7 e e 99.9
TATB 1.0 85.8 0.9 87.9
Viton A e 0.4 97.9 98.6
a
Starting decomposition temperature.
b
Final decomposition temperature.

amount. This implies that PBXs are comparatively exhibited to amount of sample used for TG analysis.
better thermal stability with less HMX amount. The decomposition Fig. 2 shows that the derivative thermogravimetric (DTG) ther-
% in the first step is not exactly found to same as the initial weight of mograms of HMX, TATB and Viton A show a single peak of
HMX due to a high thermal conductivity of TATB causes the maximum thermal decomposition. The HT1080, HT2070, HT3060,
neighbouring HMX molecules to heat up and fast burning more HT4050, HT5040, HT6030, HT7020 and HT8010 show three peaks;
quickly than pure HMX [51]. Other reason for discrepancy in the the first peak appears due to the thermal decomposition of HMX,
results may be due to different particle size i.e. coarse HMX and fine the second peak appears due to the thermal decomposition of TATB
TATB crystals [49] used for formulations as well as the small and the third peak appears due to the thermal decomposition of
 A. Singh et al. / Defence Technology 13 (2017) 22e32 25

Fig. 2. DTG thermograms of HMX, TATB, Viton A and PBXs obtained at heating rate 10
C/min under nitrogen atmosphere.

polymer matrices. HMX/Viton A and TATB/Viton A samples show 3.2. DSC studies
two peaks; the first peak attributes to the thermal decomposition of
HMX or TATB, while the second peak ascribes to the thermal DSC curves of HMX, TATB, Viton A and PBXs based on mixture of
decomposition of polymer matrices. DTG thermograms can be HMX and TATB are shown in Fig. 3. DSC curve of HMX shows three
easily distinguished as when the thermal decomposition ends and endo-/exo-thermic peaks in the DSC curve. The first endothermic
the next decomposition begins. Thermal data in terms of TG onset peak appears at 189
C due to b to d-HMX polymorph trans-
temperature (Tonset), endset temperature (Tendset) and peak tem- formation. The second endothermic peak appears at 278
C which
perature for maximum thermal decomposition (Tmax) are listed in corresponds to melting point (Tm), followed by the third
Table 2. It is observed that Tonset and Tmax values are slightly exothermic peak at 287
C due to the thermal decomposition of
increased with increasing HMX amount from 10 to 80 wt. %. The HMX [54,57]. DSC curve of TATB exhibits a single exothermic peak
results also suggest that the HMX, TATB and Viton A are compatible at 388
C due to the thermal decomposition of TATB. The thermal
with each other. properties such as glass transition temperature (Tg) and softening

Table 2
Thermal data in terms of Tonset, Tendset and Tmax obtained from TG/DTG thermograms of PBXs along with HMX, TATB and Viton A.

Sample Thermal decomposition

Tonset Tendset Tmax Tonset Tendset Tmax Tonset Tendset Tmax

(270
C e290
C) (290
C e450
C) (450
C e550
C)

HT1080 278.1 385.7 280.2 348.4 373.9 369.7 467.3 495.6 482.9
HT2070 278.3 282.4 280.7 346.5 373.6 369.6 470.7 486.9 478.5
HT3060 279.2 281.7 280.7 340.5 363.9 359.4 471.0 495.6 480.3
HT4050 279.7 281.9 280.8 340.2 363.9 358.8 474.2 495.3 481.1
HT5040 279.5 281.4 280.8 337.5 363.7 357.3 473.2 493.7 486.6
HT6030 279.0 283.0 281.3 329.1 353.9 344.4 473.3 495.7 486.8
HT7020 279.4 286.1 281.5 309.1 326.4 335.2 472.3 495.7 487.7
HT8010 279.9 286.0 283.7 321.4 321.4 473.3 497.2 486.8
HMX/Viton A 279.9 286.8 281.0 e e e 470.1 497.5 487.4
TATB/Viton A e e e 361.3 393.2 375.5 441.2 480.9 476.2
HMX 282.3 292.6 285.3 e e e e e e
TATB e e e 352.7 383.4 376.7 e e e
Viton A e e e e e e 458.8 503.5 493.3
26 A. Singh et al. / Defence Technology 13 (2017) 22e32

Fig. 3. DSC curves of HMX, TATB, Viton A and PBXs samples at heating rate 10
C/min under nitrogen atmosphere.

point of Viton A are measured where the sample was subjected to of TATB. The second exothermic peak is not clearly seen in the DSC
heating from e50
C to 100
C at heating rate of 10
C/min under curves because of overlapping of the multiple DSC curves into in a
nitrogen atmosphere. Fig. 3 shows DSC curve of Viton A. It is single curve. The third exothermic peak appears within a range of
observed that the Tg and softening point are found to be e18
C and 484
C e498
C due to thermal decomposition of polymer matrices.
60
C respectively. The second DSC curve of Viton A shows a single DSC curves also show that there is no change in thermal decom-
endothermic peak at 493
C due to the thermal decomposition of position behaviour or a new peak formation for PBXs formulations
polymer matrices. compared to those individual materials. This confirms that the
DSC curves of PBXs; HT1080, HT2070, HT3060, HT4050, HT5040, HMX, TATB and Viton A are compatible with each other. The results
HT6030, HT7020 and HT8010 show two endothermic and three obtained by DSC are fully supported to TG thermal data where PBXs
exothermic peaks in the DSC curves. The first two endothermic exhibit almost the similar thermal decomposition profiles.
peaks appear at 190 ± 1.5
C and 278 ± 0.6
C due to b to d-HMX The thermal data in terms of Tonset, Tendset, Tmax and enthalpy of
phase transformation and melting point respectively. Some of the decomposition process interpreted from DSC curves are listed
endothermic peaks due to melting point for HT1080, HT2070 and in Table 3. It can be seen from Table that the melting point of PBXs is
HT3060 samples is not clearly detected in DSC curves, while the found to very close to the HMX. The Tmax values are shifted towards
first exothermic peak appears within a range of 282
C e286
C due a high temperature with increasing the HMX amount, indicating
to thermal decomposition of HMX and the second exothermic peak that the thermal stability is slightly decreased with decreasing
appears in a range of 370
C e380
C due to thermal decomposition HMX amount and lower than pure HMX. The enthalpy of
 A. Singh et al. / Defence Technology 13 (2017) 22e32 27

Table 3
The thermal data in terms of Tm, Tmax and enthalpy of PBXs along with HMX, TATB and Viton A.

Sample Tm The exo- and endo-thermic peaks for decomposition process

/(
C)
1st step 2nd step 3rd step

Tmax DH Tmax DH Tmax DH

/(
C) /(J.g1) /(
C) /(J.g1) /(
C) /(J.g1)

HT1080 e 282.1 142 380.2 772 489.1 102

HT2070 e 282.4 252 378.8 318 484.4 161
HT3060 e 282.9 361 378.2 342 484.9 194
HT4050 278.0 283.0 788 376.3 179 496.1 149
HT5040 278.1 284.8 1050 374.9 80 497.7 169
HT6030 277.8 285.1 1190 374.0 43 489.2 112
HT7020 277.9 285.7 1264 370.1 14 493.0 120
HT8010 278.6 286.1 1345 e e 494.5 141
HMX 278.0 287.2 1921 e e e e
TATB e 387.8 1145 e e e e
Viton A 55.9 493.4 (endothermic) 79 e e e e

exothermic decomposition process is found to increase with Table 4

increasing HMX amount. Table 3 shows that PBXs mass-specific The Tonset and Tg of HT4050, HT5040, HT6030, HT7020, and HT8010 samples from
DSC curves under nitrogen atmosphere.
enthalpy of decomposition is comparable with the decomposition
enthalpy of pure HMX. The Tmax value for the second exothermic Sample Tonset Tg
decomposition process is reduced, while the enthalpy of decom- /(
C) /(
C)

position process is decreased with increasing the HMX amount. The HT4050 16.4 15.9
enthalpy for the third exothermic decomposition process is found HT5040 16.3 15.8
HT6030 16.2 15.7
in a wide range of 102e194 kJ/mol. The difference in the results
HT7020 16.2 15.7
obtained are probably due to a broadening of the decomposition HT8010 16.4 15.8
peaks and different mass of Viton A in PBXs formulations which is
confirmed in TGA profiles during pyrolysis process.
To investigate the effect of HMX amount on Tg, some PBXs
as it is heated dynamic or isothermal conditions. Generally, the rate
samples; HT4050, HT5040, HT6030, HT7020 and HT8010 were
of reaction is described as a linear function of two functions i.e. rate
subjected to heating from e50
C to 100
C at heating rate 10
C/
constant (k) and function of the conversion f(a)
min under nitrogen atmosphere. DSC curves HT4050, HT5040,
HT6030, HT7020 and HT8010 samples are shown in Fig. 4. The da
thermal data in terms of Tonset and Tg are listed in Table 4. It is ¼ k f ðaÞ (1)
dt
observed that Tg is found to close with each other as e15.7 ± 0.2
C
which is approximately 3
C higher than Viton A. where k is a reaction rate constant, a is a extent of conversion, and
f(a) is a differential form of the kinetic model, which is a function of
a. According to Arrhenius Equation, rate constant which is tem-
3.3. Kinetics studies perature dependent is described by following Equation (2)

3.3.1. The theoretical background Ea

k ¼ AeRT (2)
Solid-state thermal decomposition kinetics can be investigated
by TGA or DSC methods [58e60] by measuring a sample property
where A is a pre-exponential factor, Ea is activation energy, T is
absolute temperature and R is the gas constant. For non-isothermal
TGA, the conversion (a) at any time T is

m0  mT
a¼ (3)
m0  m∞

where, mo is an initial weight of sample, mT is a sample weight at

temperature T, and m∞ is a final weight of sample. By substituting of
Equation (2) into Equation (1) gives

da Ea
¼ AeRT f ðaÞ (4)
dt
For a dynamic TGA in a non-isothermal experimental condition,
introducing the heating rate f ¼ dT/dt into the Equation (4) gives

da A Ea=RT
¼ e f ðaÞ (5)
dT f
Equations (4) and (5) are fundamental expressions of analytical
Fig. 4. The Tg and Tonset of HT4050, HT5040, HT6030, HT7020 and HT8010 samples at methods used to investigate kinetic parameters on the basis of the
heating rate 10
C/min under nitrogen atmosphere. TGA data [61,62]. TG measurements consist of performing the
28 A. Singh et al. / Defence Technology 13 (2017) 22e32

kinetics analysis which include weight loss curves obtained under for comparing with PBXs and it is found to be 1295 ± 27.5,
dynamic condition. 158.5 ± 2.1 and 377.5 ± 3.5 kJ/mol respectively.
For comparison, the activation energy of HT1080 sample is
3.3.2. Flynn-Wall-Ozawa (FWO) method registered close to the literature value of HMX/Viton A formulation
In general, the activation energies are calculated by model-free [61]. Shi et al. [61] are reported that the activation energy of HMX/
methods such as Kissinger-Akahira-Sunose (KAS), Flynn-Wall- Viton A formulation at different heating rates by DSC is
Ozawa (FWO) and Friedman methods. The FWO method is a 482.7 ± 17.0 kJ/mol using Kissinger method. But the activation
model-free method which is obtained from integral isoconver- energy of the subsequent PBX formulations in our results is
sional method that Flynn, Wall [63] and Ozawa [64] proposed for increased with increasing the HMX amount. The thermal decom-
calculation of activation energy by Doyle's approximation of the position process after a certain amount of TATB may be accompa-
temperature integral [65,66]. nied by fast burning more quickly or deflagration process which can
greatly promote the activation energy of PBXs in the first step
Aa Eaa Eaa thermal decomposition.
logðfÞ ¼ log  2:315  0:457 (6)
gðaÞR RTa The activation energy of PBXs based on mixture of HMX and
TATB in the second step thermal decomposition is found to be
where (f) corresponds to the heating rate, Ta is temperature in slightly lower than that of pure TATB. The thermal decomposition
Kelvin at conversion a, Eaa is activation energy in kJ/mol at different kinetics of TATB [54] is studied and the activation energy is found to
conversion a. This is equation of straight line in which log(f) and 1/ be 42.1 kcal/mol which is close to our results. The activation energy
Ta, are two variables from which activation energy at given con- of PBXs in the third step is also observed to be lower than that of
version can be obtained from slope of plot log(f) against 1/Ta. Viton A. It is stated that activation energy of Viton A is more varied
in the literature. The activation energy of the thermal decomposi-
3.3.3. Kissinger-Akahira-Sunose (KAS) method tion derived from TG and decomposition endotherm is 293 and
This method is also model-free kinetics based on an 350 kJ/mol using single heating rate data [72]. Burnham et al. [50]
isoconversional method [67e69] where activation energy is a are reported that the activation energy of thermal decomposition of
function of the extent of conversion. KAS method [70] consists of Viton A is 217 kJ/mol using model fitting to correlate reaction data
extending the Kissinger's method [71] which is given by Equation from Freidman isothermal method. Moreover, Papazian [73] is
(7) investigated the thermal kinetic of Viton A from single heating
experiment and is found the activation energy of 356 kJ/mol.
4 AR Eaa To evaluate activation energy at different conversion, the
ln ¼ ln  (7)
Ta 2 Ea g ðaÞ RTa

The plot ln(f/Ta 2) against 1=Ta at a constant value of a should be

a straight line and activation energy is calculated from slope of KAS
plots.

3.3.4. Calculation of kinetic parameters

The activation energy and pre-exponential factor (A) of thermal
decomposition of PBXs are determined by single heating rate
measurement and the results are listed in Table 5. The purpose of
the present work is to demonstrate the effect of HMX amount on
the thermal kinetics. The results show that the activation energy is
varied from 524 to 1219 kJ/mol when HMX amount is increased
from 10 to 80 percent by wt. It is initially low and then it is started
to increase with increasing HMX amount in the first step thermal
decomposition. In the second step of decomposition, it is found
within a range of 140e153 kJ/mol, while it is observed within a
range of 304e366 kJ/mol in the third step thermal decomposition. Fig. 5. TG thermograms of HT4050 sample at different heating rates under nitrogen
The activation energy of HMX, TATB and Viton A is also calculated atmosphere.

Table 5
The kinetic parameters for thermal decomposition of PBXs along with HMX, TATB and Viton A.

Sample Kinetic parameters

1st step 2nd step 3rd step

1 1 1 1
Ea/(kJ$mol ) ln(A) (s ) Ea/(kJ$mol ) ln(A) (s ) Ea/(kJ$mol1) ln(A) (s1)

HT1080 524.4 ± 22.3 104.4 140.4 ± 2.8 21.2 306.4 ± 8.5 44.7
HT2070 781.8 ± 32.1 166.5 148.7 ± 2.7 22.7 317.9 ± 7.0 46.6
HT3060 1118.3 ± 21.3 242.7 144.6 ± 1.3 22.1 317.3 ± 4.5 46.3
HT4050 1142.3 ± 17.5 244.0 145.7 ± 1.6 22.7 302.8 ± 4.3 43.7
HT5040 1079.4 ± 23.9 242.5 146.6 ± 1.4 23.4 319.4 ± 5.4 46.9
HT6030 1119.6 ± 22.9 243.3 144.2 ± 2.3 23.5 337.2 ± 5.1 49.2
HT7020 1142.3 ± 19.9 244.6 148.9 ± 1.9 23.7 324.6 ± 5.9 47.0
HT8010 1219.7 ± 16.8 261.3 153.5 ± 2.5 24.2 366.8 ± 5.1 56.3
HMX 1295.4 ± 27.5 278.0 e e e e
TATB e e 158.5 ± 2.1 25.9 e e
Viton A e e e e 377.5 ± 3.5 57.0
 A. Singh et al. / Defence Technology 13 (2017) 22e32 29

thermal decomposition measurement for HT4050 (only one sam- in the second step due to thermal decomposition of TATB. Finally,
ple) was carried out at 25
Ce600
C under nitrogen atmosphere. 9.7e9.8% weight loss attributes to thermal decomposition of Viton
TG thermograms recorded at different heating rates of 10, 20 and A. It can be seen from Fig. 5 that the initial thermal decomposition
30
C/min are shown in Fig. 5. TG thermograms show three steps is shifted towards a high temperature with increasing heating rate.
weight loss; 45e55% weight loss occurs in the first step due to The activation energy at different conversion (a ¼ 0.1e0.9) is
thermal decomposition of HMX, while 28e34% weight loss occurs calculated by using FWO Equation (6). Fig. 6 shows FWO plots at
different extent of conversion. The activation energy calculated at
different conversion and their correlation co-efficient constant are
listed Table 6. The activation energies in the first step are found to
be 1290.4, 1260.3, 1178.5, 1075.2 and 1035.4 kJ/mol at conversion of
0.1, 0.2, 0.3, 0.4 and 0.45 respectively. The linear correlation co-
efficient for each activation energy is high than 0.9909. It is also
observed that activation energy is reduced with extent of conver-
sion. The activation energies in the second step are found to 81.9,
108.7, 134.2 and 156.0 kJ/mol at different conversion of 0.65, 0.7,
0.75 and 0.8 respectively, indicating that the activation energy is
increased with extent of conversion, while the activation energy in
the third step is 186.1 kJ/mol at conversion of 0.9.
According to recommendation of kinetic committee of the In-
ternational Confederation for Thermal Analysis and Calorimetry
(ICTAC) [74], KAS method offers a significant improvement in the
accuracy to reliable evaluation of the activation energy by means of
TGA compared to FWO method. The activation energy is calculated
as a function of conversion by using isoconversional method of KAS
Equation (7). The KAS plots at different extent of conversion are
shown in Fig. 7. The activation energy obtained from the slope at
different extent of conversion (a ¼ 0.1e0.9) are calculated and
summarized in Table 7. It can be seen from Table that the activation
energy is not same at different extent of conversion indicating
multiple reaction mechanisms.
Fig. 8 shows the dependence of the activation energy on extent
of conversion. The results show that the activation energy declines
about 254.9 kJ/mol when the extent of conversion is changed from
0.1 to 0.45 in the first step, then in the second step the activation
energy arises from 71.5 to 145.1 kJ/mol at extent of conversion of
0.65e0.8, and in the third step it arises subsequently from 112.4 to
172.3 kJ/mol as the conversion is changed from 0.85 to 0.9 close to
the end of reaction. The mean activation energy calculated by FWO
and KAS methods are close and in good agreement to those ob-
tained from single heat rate measurement (1142 ± 17.5 kJ/mol) in
the first step decomposition, while in the second and the third
steps, there are found difference in the results which can be further
improved by calculating the activation energy between 0.8 and 0.85
in the second step as well as 0.9 and 0.95 in the third step where
Fig. 6. FWO plots of HT4050 sample at conversion (a) 0.1e0.45, (b) 0.6e0.9.

Table 6
Kinetic parameters for thermal decomposition of HT4050 sample by FWO method.

Conversion (a) Activation energy and correction coefficient

1st step 2nd step 3rd step

1 1
Ea/(kJ$mol ) ra Ea/(kJ$mol ) ra Ea/(kJ$mol1) ra

0.1 1290.4 0.9909 e e e e

0.2 1260.3 0.9999 e e e e
0.3 1117.9 0.9943 e e e e
0.4 1075.2 0.9967 e e e e
0.45 1035.4 0.9941 e e e e
0.6 e e 51.2 0.9981 e e
0.65 e e 81.9 0.9855 e e
0.7 e e 108.9 0.9884 e e
0.75 e e 134.2 0.9971 e e
0.8 e e 156.0 0.9987 e e
0.85 e e e e 130.0 0.9251
0.9 e e e e 186.1 0.9924

Mean 1168.0 ± 45.4 120.3± 13.8 158.1 ± 19.9

ra ¼ correction coefficient for linear regression.

30 A. Singh et al. / Defence Technology 13 (2017) 22e32

Fig. 8. Dependence of activation energy against extents of conversion.

is slightly lower than our values. The kinetics by a single heating

measurement is meticulously matched with the reliable kinetic
evaluations obtained at multiple heating rates in aforementioned
FWO and KAS methods.

4. Conclusions

In the present study, the thermal properties such as Tg, Tm and

thermal decomposition were investigated for PBXs based on
mixture of HMX and TATB with Viton A using STA and DSC. The
results indicate that 9e72% weight loss was occurred in the first
step due to thermal decomposition of HMX. It was also found that
weight loss was increased with increasing HMX amount. TGA
indicated that Tonset and Tmax values were slightly increased with
increasing HMX amount. These results suggested that the HMX,
TATB and Viton A were thermally stable and compatible with each
Fig. 7. KAS plots of HT4050 sample at conversion (a) 0.1e0.45, (b) 0.65e0.9. other. The activation energies were ranged from 524 to 1219 kJ/mol,
and the natural logarithm of pre-exponential factor were ranged
from 104.4 to 261.4 s1 for a single heating rate measurement
there may chance to get maximum value of activation energy. The under non-isothermal condition. The activation energy was
high values of activation energy at low conversion are elucidated significantly increased with increasing the HMX amount in the first
due to a high thermal conductivity of TATB which causes the step thermal decomposition. The kinetics were also investigated by
neighbouring HMX molecules to heat up and fast burning more using FWO and KAS methods. The TGA indicated that the decom-
quickly than pure HMX [51]. However, the differences observed in position of HT4050 was occurred in distinct steps at multiple
the literature data can also be attributed to the fact that the kinetics heating rates under non-isothermal condition. The results indi-
for decomposition of HMX depend on the heating rate [75]. cated that activation energies were 1290 to 1050 kJ/mol at con-
Makasher et al. [62] are investigated kinetics of TATB by TGA version of 0.1e0.45, whereas at conversion of 0.65e0.8, the
method and the activation energy is found to 150.58 kJ/mol which activation energy was ranged from 81.4 to 156.3 kJ/mol, and the

Table 7
Kinetic parameters for thermal decomposition of HT4050 sample by KAS method.

Conversion (a) Activation energy and correction coefficient

1st step 2nd step 3rd step

1 1
Ea/(kJ$mol ) ra Ea/(kJ$mol ) ra Ea/(kJ$mol1) ra

0.1 1282.3 0.9924 e e e e

0.2 1251.9 0.9989 e e e e
0.3 1109.2 0.9932 e e e e
0.4 1065.5 0.9960 e e e e
0.45 1027.4 0.9923 e e e e
0.6 e e 41.1 0.9981 e e
0.65 e e 71.5 0.9855 e e
0.7 e e 98.2 0.9884 e e
0.75 e e 123.4 0.9971 e e
0.8 e e 145.1 0.9987 e e
0.85 e e e e 112.4 0.9251
0.9 e e e e 172.3 0.9924

Mean 1147.3 ± 45.5 109.6 ± 13.8 142.4± 21.1

 A. Singh et al. / Defence Technology 13 (2017) 22e32 31

third step decomposition the activation energy was attributed to 2009.

[22] Gibbs TR, Popolato A. LASL high explosive property data. California, Berkeley:
130 and 186.1 kJ/mol at conversion of 0.85 and 0.9 respectively by
University of California Press; 1980. p. 1e30.
FWO method. The activation energy calculated by both FWO and [23] Maienschein JL, Garcia F. Thermal expansion of TATB based explosives from
KAS methods was close to each other. The mean activation energy 300 K to 566 K. Thermochim Acta 2002;384:71.
was also a good agreement with a single heating rate measurement [24] Ninan KN, Krishnan K, Rajeev R, Viswanathan G. Thermo analytical in-
vestigations on the effect of atmospheric oxygen on HTPB resin. Propel Explos
in the first step decomposition. Pyrotech 1996;21:199.
[25] Voreck W, James B, John E, Hooshang R. Shaped charge for a perforating gun
having a main body of explosive including TATB and a sensitive primer.
Acknowledgements United States Patent 5597974; 1997.
[26] Voreck W, James B, John E, Hooshang R. Shaped charge containing tri-
The authors would like to express their sincere thanks to DRDO aminotrinitrobenzene (TATB), European Patent, EP-0794163 1997.
[27] Cooper PW. Explosive engineering. New York: Wiley-VCH; 1996.
(TBR-1251) for funding and awarding the Project. The authors are [28] Aggarwal P, Singh V, Singh A, Scherer UW, Singh T, Singla ML, et al. Physico-
expressed their sincere thanks to Sh. C. Sarkar, Sh. N. Mukherjee, Dr. chemical transformation in swift heavy ion modified poly(-
P. K. Soni and Dr. Manjit Singh, Distinguished Scientist/Director etheneterephthalate). Radiat Phys Chem 2012;81:284.
[29] Qiu L, Xiao H. Molecular dynamics study of binding energies, mechanical
TBRL for their motivation, guidance and constant help provided properties, and detonation performances of bicyclo-HMX-based PBXs.
during the execution of this work. J Hazard Mater 2009;164:329e36.
[30] Xiao JJ, Fang GY, Ji GF, Xiao HM. Simulation investigation in the binding en-
ergy and mechanical properties of HMX-based plastic- bonded explosives
References (PBXs). Chin Sci Bull 2005;50(1):21e26.9.
[31] Xiao JJ, Huang YC, Hu YJ, Xiao HM. Molecular dynamics simulation of me-
[1] Mattos EC, Moreira ED, Diniz MF, Dutra RCL, Silva G, Iha K, Teipel U. Charac- chanical properties of TATB/Flourine-polymer PBXs along different surfaces.
terization of polymer-coated RDX and HMX particles. Propellants Explos Sci China Ser B-Chem 2005;48:504e10.
Pyrotech 2008;33(1):44e50. [32] Manaa MR, Fried LE, Melius CF, Elstner M, Frauenheim T. Decomposition of
[2] Yan QL, Zeman S, Elbeih A. Recent advances in thermal analysis and stability HMX at extreme conditions: a molecular dynamics simulation. J Phys Chem A
evaluation of insensitive plastic bonded explosives (PBXs). Thermochim Acta 2002;106:9024e9.
2012;537:1e12. [33] Ma XF, Xiao JJ, Huang H, Ju XH, Li JS, Xiao HM. Simulative calculation on
[3] Agrawal JP. 2005. Some new high energy materials and their formulations for mechanical property, binding energy and detonation property of TATB/
specialized applications. Propellants Explos Pyrotech 2005;30(5):316e28. fluorine-polymer PBX. Chin J Chem 2006;24:473e7.
[4] Boddu VM, Viswanath DS, Ghosh TK, Damavarapu R. 2,4,6-Triamino-1,3,5- [34] Song XL, Wang Y, An CW, Guo XD. Dependence of particle morphology and
trinitrobenzene (TATB) and TATB-based formulations-A Review. J Hazard size on the mechanical sensitivity and thermal stability of octahydro-1,3,5,7-
Mater 2010;181(1):1e8. tetranitro-1,3,5,7-tetrazocine. J Hazard Mater 2008;159(2e3):222e9.
[5] Schaffer CL. Compositional analysis of PBX 9503-A TATB/HMX/Kel-F800 [35] Tarver CM. Modeling detonation experiments on triaminotrinitrobenzene
formulation. Report MHSMP-81e55, DE 82006183. USA: Department of En- (TATB)-based explosives LX-17, PBX 9502, and Ultrafine TATB. J Energ Mater
ergy; 1981. 2012;30(3):220e51.
[6] Provatas A. Formulation and performance studies of polymer bonded explo- [36] Elbeih A, Zeman S, Jungova M, Akstein Z. Effect of different polymeric matrices
sives (PBX) containing energetic binder systems. Part 1, Report DSTO-TR- on the sensitivity and performance of interesting cyclic nitramines. Cent Eur J
1397. Edinburgh, Australia: Weapons Systems Division, Systems Sciences Energ Mater 2012;9(2):131e8.
Laboratory; 2003. [37] Elbeih A, Pachman J, Zeman S, Va vra P, Trzcin
 ski WA, Akstein Z. Detonation
[7] Nandi AK, Ghosh M, Sutar V, Pandey RK. Surface coating of cyclo- characteristics of plastic explosives based on attractive nitramines with pol-
tetramethylenetetranitramine (HMX) crystals with the insensitive high yisobutylene and poly(methyl methacrylate) binders. J Energ Mater
explosive1,3,5-triamino-2,4,6-trinitrobenzene (TATB). Central Eur J Energetic 2012;30(4):358e71.
Mater 2002;9(2):119e30. [38] Lee JS, Hsu CK, Chang CLA. Study on the thermal decomposition behaviors of
[8] Kaur J, Arya VP, Kaur G, Lata P. Evaluation of the thermo-mechanical and PETN, RDX, HNS and HMX. Thermochim Acta 2002;392e393:173e6.
explosive properties of bimodal and hybrid polymer bonded explosive (PBX) [39] Fu XL, Fan XZ, Wang BZ. Thermal behavior, decomposition mechanism and
compositions based on HNS and HMX. Central Eur J Energetic Mater thermal safety of 5,7-diamino-4,6-dinitrobenzenfuroxan (CL-14). J Therm
2013;10(3):371e91. Anal Calorim 2016;124:993.
[9] Guo H, Luo J-R, Shi P-A, Xu J-G. Research on the fracture behavior of PBX under [40] Yan Q-L, Zeman S, Sa nchez Jime nez PE, Zhao F-Q, Perez-Maqueda LA, Ma lek J.
static tension. Def Technol 2014;10:154e60. The effect of polymer matrices on the thermal hazard properties of RDX-based
[10] Tian Y, Xu R, Zhou Y, Nie F. Study on formulation of CL-20. In Proceedings of PBXs by using model-free and combined kinetic analysis. J Hazard Mater
the 4th International Autumn Seminar on Propellants, Explosives and Pyro- 2014;271:185e95.
technics, Shaoxing, China 2001 pp. 43e47. [41] Singh G, Felix SP, Soni P. Studies on energetic compounds Part 31: thermolysis
[11] An C, Wang J, Xu W. Preparation and properties of HMX coated with a and kinetics of RDX and some of its plastic bonded explosives. Thermochim
composite of TNT/energetic material. Propel Explos Pyrotech 2010;35: Acta 2005;426:131e9.
365e72. [42] Singh G, Felix SP, Soni P. Studies on energetic compounds part 28: thermolysis
[12] Nair UR, Asthana SN, Rao AS, Gandhe BR. Advances in high energy materials. of HMX and its plastic bonded explosives containing Estane. Thermochim
Def Sci J 2010;60(2):137e51. Acta 2003;399:153e65.
[13] Turker L. Thermobaric and enhanced blast explosives (TBX and EBX). Def [43] Zeman S, Elbeih A, Akstein Z. Preliminary study on several plastic bonded
Technol 2016;12(6):423e35. explosives based on cyclic nitramines. Chin J Energ Mater 2011;19(1):8e12.
[14] Elbeih A, Husarova A, Zeman S. Path to ε-HNI W with reduced impact sensi- [44] Yan Q-L, Zeman S, Zhang X-H, Ma lek J, Xie W-X. The mechanisms for
tivity. Cent Eur J Energ Mater 2011;8(3):173e82. desensitization effect of synthetic polymers on BCHMX: physical models and
[15] Agrawal JP. High energy materials, propellants, explosives and pyrotechnics. decomposition pathways. J Hazard Mater 2015;294:145e57.
Wiley- VCH; 2010. p. 69e150. Chapter 2. [45] Yan Q-L, Zeman S, Elbeih A. Thermal behavior and decomposition kinetics of
[16] Sorescu DC, Boatz JA, Thompson DL. Classical and quantum mechanical Viton A bonded explosives containing attractive cyclic nitramines. Thermo-
studies of crystalline FOX-7 (1, 1 diammino- 2, 2- dinitroethylene). J Phys chim Acta 2013;562(20):56e64.
Chem 2001;105(20):5010e21. [46] Yan Q-L, Zeman S, Zhao F-Q, Elbeih A. Non-isothermal analysis of C4 bonded
[17] Steele RD, Stretz LA, Taylor GW, Rivera T. Effects of Binder Concentration on explosives containing different cyclic nitramines. Thermochem Acta
the Properties of Plastic Bonded Explosives. Proceedings of the Ninth Sym- 2013;556:6e12.
posium (International) on Detonation, Office of the Chief of Naval Research [47] Yan Q-L, Zeman S, Zhang T-L, Elbeih A. Non-isothermal decomposition
OCNR 113291e7, Portland, OR 1989 August 28, pp. 1014e1018. behavior of Fluorel bonded explosives containing attractive cyclic nitramines.
[18] Talwar MB, Agrawal AP, Anniyappan M, Gore GM, Asthana SN, Venugopalan S. Thermochim Acta 2013;574:10e8.
Method for preparation of fine TATB (2e5 mm) and its evaluation in plastic [48] Yan Q-L, Zeman S, Zhang T-L, Elbeih A, Akstein Z. The Influence of the Semtex
bonded explosive (PBX) formulations. J Hazard Mater 2006;137:1848e52. matrix on the thermal behavior and decomposition kinetics of cyclic nitr-
[19] Boddu VM, Viswanath DS, Ghosh TK, Damavarapu R. 2,4,6-Triamino-1,3,5- amines. Central Eur J Energetic Mater 2013;10(4):509e28.
trinitrobenzene (TATB) and TATB-based formulations-a review. J Hazard [49] Clements BE, Mas EM. A theory for plastic-bonded materials with a bimodal
Mater 2010;18:1e8. size distribution of filler particles. Modell Simul Mater Sci Eng 2004;12:
[20] Dobratz BM. The insensitive high explosive Triaminotrinitrobenzene (TATB): 407e21.
development and characterization-1888 to 1994. report LA-13024H. Los [50] Burnham AK, Weese RK. Kinetics of thermal degradation of explosive binders
Alamos: National Laboratory; August 1994. Viton A, Estane and Kel-F. Thermochim Acta 2005;426:85e92.
[21] Willey TM, DePiero SC, Hoffman DM. A Comparison of New TATBs, FK-800 [51] Craig M, Tarver J, Koerner G. Effects of endothermic binders on times to ex-
binder and LX-17-like PBXs to Legacy Materials, 1st Korean International plosion of HMX- and TATB-based plastic bonded explosives. J Energetic Mater
Symposium on High Energy Materials Incheon, South Korea October 6-9, 2007;26(1):1e28.
32 A. Singh et al. / Defence Technology 13 (2017) 22e32

[52] McGuire RR, Tarver CM. Chemical Decomposition Models for the Thermal 1996;46(1):225e36.
Explosion of Confined HMX, TATB, RDX, and TNT Explosives. Proceedings of [63] Flynn JH, Wall LA. General treatment of the thermogravimetry of polymers.
the Seventh Symposium (International) on Detonation, Naval Surface J Res Natl Bur Std, A Phys Chem 1966;70:487e523.
Weapons Center NSWC MP 82e334, Annapolis MD, 1981; 16, pp. 56e65. [64] Ozawa T. A new method of analyzing thermogravimetricdata. Bull Chem Soc
[53] Lin CP, Chang YM, Tseng JM, Shu C-M. Comparisons of nth-order kinetic al- Jpn 1965;38:1881e6.
gorithms and kinetic model simulation on HMX by DSC tests. J Therm Anal [65] Zoglio MA, Windheuser JJ, Vatti R, Maulding HV, Kornblum Jr SS, Jacobs AL,
Calorim 2010;100:607. et al. Pharm Sci Linear nonisothermal Stab Stud 1968;57:2080e5.
[54] Tsyshevsky RV, Sharia O, Kuklja MM. Molecular theory of detonation initia- [66] Doyle CD. Estimating isothermal life from thermogravimetric data. J Appl
tion: insight from first principles modelling of the decomposition mechanisms Polym Sci 1962;6. 639e632.
of organic nitro energetic materials. Molecules 2016;21:236. [67] Vyazovkin S, Wight CA. Model-free and model-fitting approaches to kinetic
[55] Singh A, Kumar M, Soni PK, Singh M, Srivastava A. Mechanical and explosive analysis of isothermal and nonisothermal data. Thermochim Acta 1999;340:
properties of plastic bonded explosives based on mixture of HMX and TATB. 53e68.
Def Sci J 2013;63(6):622e9. [68] Vyazovkin SV, Goryachko VI, Lesnikovich AI. An approach to the solution of
[56] Ger MD, Hwu WH, Huang CC. A study on the thermal decomposition of the inverse kinetic problem in the case of complex processes. Part III. Parallel
mixtures containing an energetic binder and a nitramine. Thermochim Acta independent reactions. Thermochim Acta 1992;197:41e51.
1993;224:127e40. [69] Vyazovkin S. Conversion dependence of activation energy for model DSC
[57] Chakraborty D, Muller RP, Dasgupta S, Goddard WA. Mechanism for unim- curves of consecutive reactions. Thermochim Acta 1994;236:1e13.
olecular decomposition of HMX (1,3,5,7-tetranitro-1, 3, 5, 7-tetrazocine), an [70] Poletto M, Zattera AJ, Santana MC. Thermal decomposition of wood: kinetics
ab initio study. J Phys Chem A 2001;105(8):1302e14. and degradation mechanisms. Bioresour Technol 2012;126:7e12.
[58] Brown ME. Introduction to thermal analysis: techniques and applications. [71] Kissinger HE. Reaction kinetics in differential thermal analysis. Anal Chem
London: Chapman and Hall; 1988. p. 127e51. 1957;29:1702e6.
[59] Brown ME. Introduction to thermal analysis: techniques and applications. 2nd [72] Cuccuru P, Manfredini S, Argiero L. La Chim e Indus 1973;55:6.
ed. Amsterdam: Kluwer; 2001. Chapter 10. [73] Papazian HA. Prediction of polymer degradation kinetics at moderate tem-
[60] Bianchi O, Martins J, De N, Fiorio R, Oliveira RVB, Canto LB. Changes in acti- perature from TGA measurements. J Appl Polym Sci 1972;16:2503.
vation energy and kinetic mechanism during EVA crosslinking. Polym Test [74] Vyazovkina S, Burnham AK, Criado JM, Pe rez-Maqueda LA, Popescu C,
2011;30:616e24. Sbirrazzuoli N. ICTAC Kinetics Committee recommendations for performing
[61] Shi X, Wang J, Li X, An C, Wang J, Ji W. Preparation and properties of kinetic computations on thermal analysis data. Thermochim Acta 2011;520:
1,3,5,7,tetranitro- 1,3,5,7 tetraazocane based nanocomposite. Def Sci J 1e19.
2015;65(2):131e4. [75] Pinheiro GFM, Lourenço VL, Iha K. Influence of the heating rate in the thermal
[62] Makashir PS, Kuriam EM. Spectroscopic and thermal studies of the decom- decomposition of HMX. J Therm Analysis Calorim 2002;67:445e52.
position of 1,3,5 triammino 2,4,6ntrinitrobenzene (TATB). J Anal Calorim