Thermite reactions of Al/Cu core-shell nanocomposites with WO3

Article abstract of DOI:10.1016/j.tca.2007.07.017

The Al/Cu core-shell nanocomposites are fabricated via a facile displacement method and characterized by scanning electron microscopy (SEM) and energy dispersion spectrometry (EDS). In comparison with the thermite reactions of bulk Al powders with WO₃, the thermite reactions of Al/Cu core-shell nanocomposites with WO₃ as a function of the equivalence ratio of Al to WO₃ (Φ) and reaction environment are investigated by differential scanning calorimetry (DSC) and dealt with in terms of reaction performance of thermite including onset and peak temperatures, maximum reaction rate (Q_max) and reaction heat (ΔH_r) based on DSC data. The results indicate that the reactivity of Al/Cu core-shell nanocomposites with WO₃ is higher than that of bulk Al powders but depends strongly on the equivalence ratio, which is optimal when the fuel in the thermites is slightly rich, i.e. Φ = 1.2. It is significant that in air, ΔH_r of WO₃ and core-shell Al/Cu (Φ = 1.2) is higher than that of WO₃ and bulk Al by 49%. In addition, it is also observed that the solid phase reaction 2Al(s) + WO₃(s) = Al₂O₃(s) + W(s) may be irritated prior to Al melting when heating Al/Cu-WO₃ in argon, which is confirmed by DSC and X-ray diffractometry (XRD). Finally, the reasons of improving the thermal performance of thermites have been discussed.

Full text of DOI:10.1016/j.tca.2007.07.017

Thermochimica Acta 463 (2007) 69–76
Thermite reactions of Al/Cu core-shell nanocomposites with WO3
Yi Wanga, Wei Jianga, Zhipeng Chenga, Weifan Chenb, Chongwei Ana,
Xiaolan Songa, Fengsheng Lia,∗
a National Special Superfine Powder Engineering Research Center,
Nanjing University of Science & Technology, Nanjing 210094, China
b School of Materials Science & Engineering, Nanchang University, Nanchang 330031, China
Available online 1 August 2007
Abstract
The Al/Cu core-shell nanocomposites are fabricated via a facile displacement method and characterized by scanning electron microscopy (SEM)
and energy dispersion spectrometry (EDS). In comparison with the thermite reactions of bulk Al powders with WO3, the thermite reactions of
Al/Cu core-shell nanocomposites with WO3 as a function of the equivalence ratio of Al to WO3 (Φ) and reaction environment are investigated
by differential scanning calorimetry (DSC) and dealt with in terms of reaction performance of thermite including onset and peak temperatures,
maximum reaction rate (Qmax) and reaction heat (ꢀHr) based on DSC data. The results indicate that the reactivity of Al/Cu core-shell nanocomposites
with WO3 is higher than that of bulk Al powders but depends strongly on the equivalence ratio, which is optimal when the fuel in the thermites is
slightly rich, i.e. Φ = 1.2. It is significant that in air, ꢀHr of WO3 and core-shell Al/Cu (Φ = 1.2) is higher than that of WO3 and bulk Al by 49%.
In addition, it is also observed that the solid phase reaction 2Al(s) + WO3(s) = Al2O3(s) + W(s) may be irritated prior to Al melting when heating
Al/Cu–WO3 in argon, which is confirmed by DSC and X-ray diffractometry (XRD). Finally, the reasons of improving the thermal performance of
thermites have been discussed.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Thermite reactions; Al/Cu core-shell nanocomposites; Al powders; WO3; Thermal reactivity
1. Introduction
getic materials, the preparation of thermites is distinct from the
monomolecular energetic materials. For the thermites in specific
application, desired thermal properties can be readily tailored
by varying ratios of oxidizer to fuel. However, due to the gran-
ular nature of composite energetic materials, reaction kinetics
largely depends on mass transport rates between reactants [7,8].
Although the composites may have extremely high energy den-
sity, the released energy is generally below their theoretical
values. It is well known that the initiation abilities and detona-
tion performance of energetic materials are strongly influenced
by their microstructures [9,10]. For example, when the size of Al
ormetallic-oxidespowdersfallwithinnanoscale, thesethermites
will have higher reactivity and burning rates than those con-
taining micron-sized components due to more intimate contact
between nanometer components [11,12]. However, it is very dif-
ficult to fabricate nanometer Al powders on a large scale because
of cost, safety, process, etc. In our previous study [13], we dealt
with the fabrication of the Al/Cu core-shell nanocomposites via
a displacement method, which is facile and productive. Here we
investigate the thermite reactions of Al/Cu core-shell nanocom-
posites with WO3 and compare them with those of micron-sized
Thermite reactions can be described as exothermic reactions
involving the reduction of metallic oxides with aluminum or
other activated metal powders. This kind of reaction can be
generalized into:
M + AOx = MOx + A + ΔH
(1)
where M is a metal and A is either a metal or a non-metal,
MOx and AOx are their corresponding oxides, and ꢀH is the
heat generated by the reaction. Because of the large exother-
mic heat, a thermite reaction can generally be initiated locally
and then become self-sustaining, a feature, which makes full
use of their energy. These reactions have been utilized in reac-
tive fill for enhanced warhead projectiles, high performance
additives for solid rocket propellants, lead-free electric matches
and a host of other application [1–6]. As composite ener-
∗
Corresponding author. Tel.: +86 25 84315942.
E-mail address: wangyi528528@yahoo.com.cn (F. Li).
0040-6031/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2007.07.017

70
Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
Al powders with WO3 with respect to the thermal performance.
As partial film of Al2O3 on the surface of micron-sized Al par-
ticles is replaced by layers of Cu nanoparticles in the synthesis
of Al/Cu nanocomposites, a kind of novel thermites of Al/Cu
core-shell nanocomposites and WO3 possessing high reactivity
can be obtained.
2.3. Sample characterization
The morphology and surface element of the samples were
examined by a LEO-1530VP field-emission scanning elec-
tron microscope coupled with energy dispersive spectroscopy
(OXFORD, INCA-300). Differential scanning calorimetry of
the thermites was performed on a TA Model Q600 differential
scanning calorimeter from room temperature to 1100 ◦C under
the air or argon gas flowing at 10 mL/min. The thermites of
about 3 mg were contained in open alumina pans. Heating rate
of 5, 10 and 20 ◦C/min was adopted. The phase variations of the
thermites in the heating process were investigated via a Bruker
Advance D8 X-ray diffractometer, using Cu K␣ radiation at
40 kV and 30 mA.
2. Experimental
2.1. Materials
The bulk Al powders (with 99% purity, average particle size
d50 = 5 ␮m) were purchased from Anshan superfine powders
Ltd., China. CuSO4·5H2O (A.R.), NH4F (C.P.), gelatin (C.P.)
and sodium diethyldithiocarbamate trihydrate (SDT) (A.R.) are
obtained from Beijing chemical Ltd., China. The WO3 pow-
ders (with 99.99% purity, average particle size d50 = 12 ␮m) and
sodium dioctylsulfosccinate (SDS) (C.P.) are purchased from
Shanghai chemical Ltd., China.
3. Results
3.1. Characterization of the samples
As shown in Fig. 1(a), raw Al particles are spherical in shape
and smooth on the surface except few tiny Al grains, whereas
the surface of the particles shown in Fig. 1(b) are covered with
sphericalparticles. Itcanbeseenmoreclearlyin Fig. 1(c)thatthe
surface of the arrow-pointed large particle shown in Fig. 1(b) is
depositedbythetightlyagglomeratednanoparticlesof20–30 nm
in size, showing honeycomb-like coarse surface. When as-
prepared Al/Cu core-shell nanocomposites are immersed into
diluted hydrochloric acid until there is no effervescence, the
porous shell-shaped products can be obtained, as revealed in
Fig. 1(d). It is more clear in Fig. 1(e) that the etching-obtained
spherical particles are hollow and composed of agglomerated
nanoparticles. To detect the surface element of Al/Cu core-shell
nanocomposites before and after etching, EDS measurements
were carried out. Fig. 2 is EDS spectra of Al/Cu nanocompos-
ites before and after etching. It indicated that the surface of
Al/Cu nanocomposites contains Cu and Al. Due to the relative
low fraction (3 wt.%) and the thinner shell of Cu, the presence of
Al element was made clear. However, the element of Cu is dom-
inant. For Al/Cu nanocomposites after etching, only Cu element
is present and the Al element is removed. It can be concluded
that Al/Cu nanocomposites with micron-sized Al particle in the
core and Cu nanoparticles as the shell can be prepared by the
displacement method [13].
2.2. Fabrication of Al/Cu core-shell nanocomposites and
preparation of the thermites
The core-shell Al/Cu nanocomposites were fabricated via a
displacement method because EAθ l3+/Al (−1.66 V) is quite lower
than ECθ u2+/Cu (+0.34 V) [13]. The designed mass fraction of Cu
to the whole Al/Cu nanocomposites is 3 wt.%. As a protective
agent of active Al, the gelatin solution with proper concentration
was introduced to CuSO4·5H2O solution. After the pH value
of the solution was adjusted at about 6.4, Al powders ultra-
sonicated in cool distilled water were rapidly added into the
above-mentioned solution kept at 35 ◦C. Then NH4F solution
was tardily dripped into Al suspension to remove the Al2O3
layers on the surface of Al powders until the suspension cannot
change the color of SDT. After filtration, lavation and evapora-
tion in vacuum at 40 ◦C, the Al/Cu core-shell nanocomposites
were fabricated.
In the present study, the stoichiometric thermite reaction can
be expressed as the following equation [1]:
2Al + WO3 = Al2O3 + W
(2)
The equivalence ratio Φ for this reaction is calculated accord-
ing to [14]:
3.2. DSC analysis of the thermites
(F/Ox)Actual
Φ =
(3)
Fig. 3 shows DSC curves of Al–WO3 with different equiva-
lence ratio Φ recorded both in air and in argon. In each curve, the
endothermic peak at about 660 ◦C, i.e. the theoretical melting
point of aluminum [15,16], is observed and may be associated
with the melting of aluminum. The exothermic peaks are located
at 993–1056 ◦C, which represents the thermite reaction of Al
with WO3. Apparently, the figures of these peaks vary with the
atmosphere changes. In air, each DSC curve shows just one
exothermic peak. When the reaction atmosphere is converted
from air to argon, these curves present three exothermic peaks,
corresponding to three stages in the thermite reaction. It can also
(F/Ox)Stoichiometric
where F represents the mass of fuel (Al), Ox the mass of oxidizer
(Al2O3).
The thermites were prepared by mixing the fuel and oxi-
dizer in the required proportions. With 1.5 wt.% SDS added, the
mixtures of the fuel (Al or Al/Cu nanocomposites) and WO3
powders at the desired equivalence ratio Φ were ultrasonicated
in n-hexane, then heated at low temperature in vacuum to evapo-
rate n-hexane and finally sifted through a wire mesh of 200-␮m
to obtain the thermites used for investigation.

Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
71
Fig. 1. SEM micrographs of (a) raw Al particles; (b) Al/Cu core-shell nanocomposites prepared by the displacement method; (c) partial surface of the arrow-pointed
particle shown in Fig. 1(b); (d) products obtained by etching Al/Cu core-shell nanocomposites with diluted hydrochloric acid; (e) products shown in Fig. 1(b) at
higher magnification.
be observed from Fig. 3 that the correspondent exothermic peak
of Al–WO3 at different equivalence ratio Φ appears at higher
temperature in air than in argon.
Fig. 4 presents DSC curves of core-shell Al/Cu–WO3 with
different equivalence ratio Φ recorded under the different flow-
ing atmosphere. As is demonstrated in Fig. 4(a), when the
thermite reaction for Al/Cu–WO3 occurs in air, DSC curves
almost level off in the temperature range of 400–850 ◦C and
only in the intercepted DSC curves on enlarged scale can the
endothermic peak assigned to Al melting at about 660 ◦C be
observed, whereas the only exothermic peaks ranging from
906 ◦C to 988 ◦C in each DSC curve are attributed to thermite

72
Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
Fig. 2. EDS spectra for (A) Al/Cu core-shell nanocomposites prepared by
a displacement method; (B) products obtained by eroding Al/Cu core-shell
nanocomposites with diluted hydrochloric acid.
Fig. 3. DSC curves of Al–WO3 with different equivalence ratio Φ at heating
rate of 20 ◦C/min, respectively, recorded (a) in air and (b) in argon.
reaction for Al/Cu–WO3. In sharp contrast, when the atmo-
sphere is converted from air to argon, three exothermic peaks
and one endothermic peak in the curves for Φ range from 0.8
to 1.6 can be observed, as shown in Fig. 4(b). Interestingly,
one weak exothermic peak, which is more apparent in the inter-
cepted DSC curves in Fig. 4(b), appears prior to Al melting and
hence may be related to some solid phase reaction. This finding
will be discussed in more detail subsequently. In addition, for
Al/Cu–WO3 in argon, the exothermic peaks broaden and appear
at higher temperature compared with those in air.
than that of Al–WO3 (114.4 ◦C) by 108.69 ◦C. All these suggest
that Al/Cu–WO3 in air is a more passive energetic material but
has more concentrative exothermic course than Al–WO3. From
Fig. 5(b), in argon, the average values of onset temperatures for
two thermites are almost equal while the average value of peak
temperature of Al/Cu–WO3 is lower than that of Al–WO3 by
59.2 ◦C. In summary, compared with Al–WO3, whether in air
or in argon, Al/Cu–WO3 is more passive but much more con-
centrative in terms of energy release and may be used as a novel
kind of passive and effective energetic materials for some spe-
cific martial applications, such as projectiles, energetic-shaped
charge, active fragments, detonator, etc.
3.3. Thermal performance analysis of thermite reactions
The reaction heats for Al–WO3 and core-shell Al/Cu–WO3
in air and in argon can be obtained from the integrated heats of
the exothermic peaks in DSC curves. The results are shown in
Fig. 6 as a function of equivalence ratio Φ. It can be seen that
the reaction heats of Al/Cu–WO3 are generally higher than those
of Al–WO3 systems and arrive at the maximum for the slightly
fuel rich (Φ = 1.2) mixture whether in air or in the argon. How-
ever, all measured reaction heats are considerably lower than
the theoretical value (2910 J/g) [17]. It is worth noting that the
reaction heat in air for Al/Cu–WO3 with Φ = 1.2 is larger than
that for Al–WO3 by 56% and accounts for approximately 86%
of the theoretical value. As is shown in Fig. 6(a and b), when the
Onset and peak temperatures of Al–WO3 and core-shell
Al/Cu–WO3 as a function of equivalence ratio are presented
in Fig. 5. It can be seen that in the most cases, variation in Φ
for each Al–WO3 and core-shell Al/Cu–WO3 has few effects
on the onset and peak temperatures whether in air or in argon.
As shown in Fig. 5(a), in air, the average value of onset temper-
atures of Al/Cu–WO3 at different equivalence ratios is higher
than that of Al–WO3 by 31.7 ◦C, while the average value of peak
temperature of Al/Cu–WO3 is lower than that of Al–WO3 by
77.04 ◦C. Moreover, the average variation value of Al/Cu–WO3
from onset temperature to peak temperature (5.7 ◦C) is lower

Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
73
Fig. 4. DSC curves of core-shell Al/Cu–WO3 with different equivalence ratio
Φ at heating rate of 20 ◦C/min, respectively, recorded (a) in air and (b) in argon.
Inserted graphs are the intercepted DSC curves on enlarged scale.
Fig. 5. The onset and peak temperatures of Al–WO3 and core-shell Al/Cu–WO3
(a) in air and (b) in argon as a function of equivalence ratio Φ. The values
plotted and error bars are the average and standard deviation, respectively, of the
temperature obtained at different equivalence ratio. The dashed and solid lines
are only intended to show the average of onset or peak temperatures.
atmosphere is converted from air to argon, the reaction heats for
Al–WO3 and core-shell Al/Cu–WO3 at investigated equivalence
ratio are dramatically reduced, for example, the reaction heats in
air for core-shell Al/Cu–WO3 is 2508 J/g, for Al–WO3 1169 J/g.
Sun et al. [18] reported that due to the incompleteness of the
thermite reactions, the reaction heats for metastable intermolec-
ular composites (MIC) which consist of a nano-sized Al-MoO3
systems are approximately 35% of the theoretical value.
The maximum reaction rate per gram of aluminum reacted in
air and in argon for Al–WO3 and core-shell Al/Cu–WO3, which
is determined by the heat flow rate at the dominant exothermic
peak of Al–WO3 and Al/Cu–WO3, is plotted against equiva-
lence ratio Φ. As shown in Fig. 7(a and b), whether in air or in
argon, the values of Qmax for Al/Cu–WO3 are generally higher
than those for Al–WO3 and reach their respective maximum
at Φ = 1.2. Remarkably, in air, the maximum reaction rate at
Φ = 1.2 has a more than 10-fold increase from 6.95 W/g for
Al–WO3 to 84.91 W/g for Al/Cu–WO3, whereas for Al–WO3
and Al/Cu–WO3 in argon, their maximum reaction rates have
little difference. Furthermore, the maximum reaction rate for
Al–WO3 and Al/Cu–WO3 is considerably higher in air than in
argon. In summary, the reactivity of Al/Cu nanocomposites with
WO3 is considerably higher than those of bulk Al particles under
the same conditions, especially in air.
4. Discussion
Comparing two inserted graphs in Fig. 4, it can be eas-
ily observed that exothermic reaction of Al/Cu–WO3 in argon
occurs prior to Al melting point, suggesting presence of some
solid–solid state reaction. These reactions did not complete
before reaching the melting point of Al. Therefore, the heats
of these reactions are inaccurate due to the competing positive
and negative energy signals. This phenomenon indicates that
partial Al/Cu nanocomposites have been partially oxidized by
WO3 in argon before the primary thermite reaction occurs.
To further confirm the possible solid state reaction prior to Al
melting, we collected the quenched products obtained by heat-
ing the thermites to 700 ◦C both in air and in argon, and then
conducted a XRD analysis. Fig. 8 shows the XRD patterns of
the quenched products. As is indicated in Fig. 8, the peaks of W
and Al2O3 appear in the XRD pattern for Al/Cu–WO3 in argon

74
Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
Fig. 6. Reaction heats for Al–WO3 and core-shell Al/Cu–WO3 (a) in air and
(b) in argon as a function of equivalence ratio Φ. The values plotted and error
bars are the averages and standard deviations, respectively, of the reaction heat
obtained at the heating rates of 5, 10 and 20 ◦C/min. The dashed and doted lines
are only intended to show the trends of data.
Fig. 7. Maximum reaction rates for Al–WO3 and core-shell Al/Cu–WO3 (a) in
air and (b) in argon as a function of equivalence ratio Φ. The values plotted and
error bars are the averages and standard deviations, respectively, of the reaction
heat obtained at the heating rates of 5, 10 and 20 ◦C/min. The dashed and doted
lines are only intended to show the trends of data.
whereas in other three patterns, just peaks of WO3 and Al can
be noted. It is also observed that the peak intensities for starting
materials WO3 and Al shown in the pattern (b) decrease com-
pared with other three patterns. However, the peak intensities
of the reaction products (e.g. W and Al2O3) are fairly faint due
to this short-time reaction. Combined with the exothermic peak
in the range of 640–650 ◦C shown in Fig. 4(b), the above XRD
results also confirm that the existence of the solid state reaction,
i.e. 2Al(s) + WO3(s) = Al2O3(s) + W(s).
In general, each Al particle was enwrapped by a dense Al2O3
layer. As the heating is proceeding, Al2O3 layer will shrink due
to the 20% density increase resulted from phase transforma-
tion from amorph-Al2O3 to ␥-Al2O3 and active Al is exposed
[19]. For Al–WO3 system, the melt Al will flood from fracted
Al2O3 shell, contact with WO3 and lead to occurring of the
thermite reaction. As a result, a new dense Al2O3 layer forms
and gets thicker and thicker. If this oxide layer were not bro-
ken again, the reaction would cease before the total mass of Al
core is consumed completely. For Al/Cu–WO3 system, however,
each Al/Cu core-shell nanocomposite has many Cu nanoparti-
Fig. 8. XRD patterns for the reaction products obtained by heating the Φ = 1.2
different thermites to 700 ◦C and quenching under different atmosphere: (a)
Al–WO3 in argon; (b) Al/Cu–WO3 in argon; (c) Al–WO3 in air; (d) Al/Cu–WO3
in air.

Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
75
Table 1
removal rate for Al2O3 is accelerated. Only when Al2O3 layer
on the surface of raw Al particles is removed to expose the active
Al, Cu2+ in solution can be reduced into nanometer Cu particles.
Therefore, the Al/Cu core-shell nanocomposites obtained via the
displacement method almost do not contain Al2O3 layer.
As showed in Table 1 and Fig. 9, in spite of the almost same
mass fraction of Cu nanoparticles as T3, T2 has lower reac-
tion heat and higher peak temperature, which is due to the fact
that Al2O3 layer on the surface of raw Al particles prevents
Cu nanoparticles from intimate contact with active Al, whereas
in preparation process of Al/Cu core-shell nanocomposites via
the displacement method, Al2O3 layer on the surface of raw
Al particles is removed to expose the active Al particles, which
are subsequently deposited by Cu nanoparticles resulting from
reduction reaction between Al and Cu2+. Thus, T3 has the low-
est peak temperature and the highest reaction heat owing to its
peculiar core-shell structure.
Thermites and their reaction heats
Thermite no.
Φ = 1.2 Thermite components
excluding WO3
Heats of reaction,
ꢀHr (J/g)
T1
T2
Al powders
805.4
838.6
Mixture of 3wt.% Cu nanoparticles
and 97wt.% Al powders
Al/Cu nanocomposites prepared via a
displacement method
T3
1198
cles on the surface of Al particle, which replace bulk of Al2O3
layer. This will result in the intimate contact between nanometer
Cu particles and active Al core and largely increase the spe-
cific surface area of Al particles. Meanwhile, these nanometer
Cu particles will act as many active centers in the course of
thermite reaction. Therefore, it is the unique microstructure of
Al/Cu core-shell nanocomposites may improve the performance
of thermite reaction in Al–WO3 system.
To further investigate effect of microstructure of Al/Cu core-
shell nanocomposites on reactivity of thermite reaction for
Al/Cu–WO3, we prepared a thermite by mixing 3 wt.% Cu
nanoparticles and 97 wt.% raw Al particles with WO3 in propor-
tion of Φ = 1.2 and made a comparative study on three thermites
including the as-prepared thermite via DSC analysis. Three ther-
mites and their reaction heats are summarized in Table 1. DSC
curves of the thermites listed in Table 1 are shown in Fig. 9.
According to our previous research, synthesis of Al/Cu core-
shell nanocomposites via a displacement method is based on the
following reactions
5. Conclusions
SEM images and EDS spectra indicate that Al/Cu core-
shell nanocomposites can be prepared via a facile displacement
method and that 20–30 nm Cu nanoparticles have been deposited
on the surface of raw Al particles successfully. The analysis of
DSC shows that based on onset and peak temperatures, maxi-
mum reaction rate (Qmax) and reaction heat (ꢀHr), the reactivity
of Al/Cu–WO3 both in air and in argon is higher than that
of Al–WO3 but depends strongly on the equivalence ratio Φ,
which is optimal when the fuel in the thermites is slightly
rich, i.e. Φ = 1.2. Moreover, the atmosphere of thermite reac-
tion has great influence on reactivity. Remarkably, the reactivity
of Al/Cu–WO3 in air is enhanced by large. It is also observed
that the solid phase reaction 2Al(s) + WO3(s) = Al2O3(s) + W(s)
may be irritated prior to Al melting when heating Al/Cu–WO3
in argon, which is confirmed by DSC and XRD. Finally, the
reasons of increase in the reactivity of thermites are discussed.
We consider that it is the unique microstructure of Al/Cu core-
shell nanocomposites that leads to a great improvement in the
reactivity for Al/Cu–WO3.
Al2O3 + 6H+ → 2Al3+ + 3H2O
Al3+ + 6F− → [AlF6]3−
(4)
(5)
(6)
2Al + 3Cu2+ → 2Al3+ + 3Cu
Firstly, Al2O3 layer on the surface of raw Al particles is
removed in acid solution to produce Al3+, and then NH4F is
added into solution and the stable complex [AlF6]3− is formed.
As a result, the balance of the reaction (4) move rightward and
Acknowledgements
This work is the combination of several research groups in
our center. The authors would like to thank Chengfang Che
of National Special Superfine Powders Engineering Research
Center in Nanjing University of Science & Technology for her
enthusiastic support in DSC analysis and Dr. Edward L. Dreizin
of New Jersey Institute of Technology who provided many help-
ful suggestions for our experiments.
References
[1] R. Ricceri, P. Matteazzi, J. Alloy Compd. 358 (2003) 71–75.
[2] N. Uzunov, R. Ivanov, Appl. Surf. Sci. 225 (2004) 72–77.
[3] K.W. Ciurowa, K. Gamrat, Z. Sawlowicz, J. Therm. Anal. Calorim. 80
(2005) 619–623.
Fig. 9. DSC curves of the thermites listed in Table 1, recorded in argon at heating
rate of 20 ◦C/min.
[4] X.L. Zhe, C.A. Dioka, A. Hendry, J. Eur. Ceram. Soc. 25 (2005) 695–702.
[5] R.H. Fan, H.L. Lu, K.N. Sun, Thermochim. Acta 440 (2006) 129–131.

76
Y. Wang et al. / Thermochimica Acta 463 (2007) 69–76
[6] S.M. Umbrajkar, M. Schoenitz, E.L. Dreizin, Thermochim. Acta 451
(2006) 34–43.
[13] Z.P. Cheng, Y. Yang, X.D. Liu, Acta Chim. Sinica 65 (2007)
81–85.
[7] J. Mei, R.D. Halldearn, P. Xiao, Scripta Mater. 41 (1999) 541–548.
[14] J.J. Granier, M.L. Pantoya, Combust. Flame 138 (2004) 373–
[8] T.M. Tillotson, A.E. Gash, R.L. Simpson, J. Non-Cryst. Solids 285 (2001)
383.
338–345.
[15] S.X. Wang, K.M. Liang, X.H. Zhang, J. Mater. Sci. Lett. 22 (2003)
855–856.
[9] S. Valliappan, J. Swiatkiewicz, J.A. Puszynski, PowderTechnol. 156(2005)
164–169.
[16] L. Lin, Q.J. Xue, H.W. Liu, J. Inorg. Mater. 11 (1996) 745–748.
[17] R.H. Perry, D.W. Green, Perry’s Chemical Engineers Handbook, vol. 2,
Sci. Press, Beijing, 2001.
[10] T.M. Yue, K.J. Huang, H.C. Man, Surf. Coat. Technol. 194 (2005) 232–237.
[11] K.B. Plantier, M.L. Pantoya, A.E. Gash, Combust. Flame 140 (2005)
299–309.
[18] J. Sun, M.L. Pantoya, S.L. Simon, Thermochim. Acta 444 (2006) 117–
127.
[19] M.L. Pantoya, J.J. Granier, J. Therm. Anal. Calorim. 1 (2006) 1–7.
[12] S. Majumdar, G.B. Kaleb, I.G. Sharma, J. Alloy Compd. 394 (2005)
168–175.