Combustion, Explosion, and Shock Waves, Vol. 37, No. 4, pp. 418 422, 2001
Reactivity of Aluminum Powders
A. P. Il in,1 A. A. Gromov,1 and G. V. Yablunovskii1 UDC 541.16:182
Translated from Fizika Goreniya i Vzryva, Vol. 37, No. 4, pp. 58 62, July August, 2001.
Original article submitted April 11, 2000; revision submitted September 5, 2000.
The reactivity of aluminum powders is determined using the following parameters: the
temperature of the onset of oxidation, maximum oxidation rate, degree of conversion
(degree of oxidation) of aluminum, reduced (conditional) ratio of the thermal effect
to the weight increment. These parameters for estimating the activity of aluminum
powders were chosen from the results of nonisothermal oxidation of powders of various
particle sizes under conditions of programmed heating (oxidizer air). In accordance
with the testing method proposed, the most reactive powder studied was STPA-4
ultrafine aluminum powder produced by electrical explosion of conductors.
INTRODUCTION known, such powders (with equal aluminum contents)
have different oxidation rates, degrees of oxidation (con-
Aluminum powder is an effective fuel: its mass
version), etc., depending on their forms, sizes, and parti-
caloricity is more than twice that of magnesium, and
cle distribution functions. With increase in particle size,
although ranking below boron and beryllium in this
the metal content of such powders usually decreases but
parameter, aluminum greatly surpasses them in den-
the oxidation rate can increase. Agglomeration, incom-
sity [1]. In addition, beryllium and its combustion
plete burning, and two-phase losses are serious problems
products are toxic, and boron is a heat-resistant ma-
in using coarsely dispersed powders, especially, in highly
terial, whose viscous low-melting oxide prevents the ox-
metallized compositions [1]. It is possible to raise the
idizer from entering the combustion zone. Therefore,
activity of coarsely dispersed powders by melting them
aluminum is the most appropriate fuel for metallized
with rare-earth elements [3], doping particle surfaces
mixed compositions [1].
with high-melting metals [4], and introducing oxidation
A problem that arises in the use of aluminum as a
catalysts into the powders [5]. A common disadvan-
fuel is to determine the activity of aluminum powders.
tage of all these methods is that being an energetic bal-
On the one hand, low reactivity of aluminum powder is
last, the additives reduce the heat of combustion fuel
preferred during its production, storage, transportation,
combustion. Another method of increasing the reac-
and processing, and, on the other hand, a high rate of
tivity of aluminum powders is to increase their surface
the process is required during oxidation.
area by using dusts with scaly particles. However, the
flat particles of dusts deteriorate the physicomechanical
properties of fuel compositions and contribute to fire
and explosion hazards during processing. This raises
OBJECT OF STUDY AND EXPERIMENTAL
the question of using new types of spherical aluminum
METHODS
powders, which combine high reactivity with relatively
Traditionally, the reactivity (reactive aluminum) high metal content.
implies the content of metallic aluminum in powders [2]. The goal of the work described here was to show
This does not involve problems for coarsely dispersed experimentally that the reactivity of aluminum pow-
powders. However, with this approach to the definition ders, especially that of ultrafine powders, depends not
of the reactivity it is difficult to explain the different only on the metal content but on different parameters
behavior of ultrafine powders during oxidation. As is as well. For a rapid analysis of the reactivity of pow-
ders, we propose to use the following parameters: the
1
High-Voltage Institute, Tomsk Polytechnical University,
temperature of the onset of oxidation (Ton [ć%C]), the
Tomsk 634050.
418 0010-5082/01/3704-0418 $25.00 © 2001 Plenum Publishing Corporation
Reactivity of Aluminum Powders 419
maximum oxidation rate (vox, mg/sec), the degree of
conversion (the degree of oxidation) of aluminum in a
specified temperature range (Ä…, %), the reduced (con-
ditional) ratio of the thermal effect to the weight incre-
ment (S/"m). These parameters can be obtained by
processing the results of nonisothermal oxidation under
conditions of programmed heating (oxidizer is air).
In recent years, along with the design of new grades
of powders using the technology of producing spherical
disperse aluminum, there is a tendency for conversion
to ultrafine aluminum powders in fuel systems (particle
characteristic size less than 1 µm) [6]. It should be noted
that we do not use the term  ultradisperse powders for
the powders studied because this term corresponds to
the physical state of a substance [7]. Among the numer-
ous methods of producing ultrafine aluminum powders,
the electric explosion method is distinguished by the
low energy inputs, high output, and high quality of the
powders produced [8]. Determination of the parameters
of nonisothermal oxidation under standard conditions of
programmed heating in air [9] was developed and tested
in sufficient detail. This makes it possible to compare
results obtained and determine the reactivity of a pow-
der using several parameters [10].
Other methodological approaches to studying the
kinetics of oxidation of ultrafine aluminum powders in
Fig. 1. Derivatograms of sample Nos. 2 (a) and 7 (b)
air can also be found in the literature. Thus, Ivanov
(sample numbers correspond to those of Table 1): m =
and Gavrilyuk [11] used derivatography as a research
5 · 10-5 kg, the heating rate in air is 10ć%C/min, and
method. However, their results are presented in the
Ä…-Al2O3 is the reference.
form of kinetic equations, which demonstrate only for-
mal parameters of the processes. This complicates anal-
ysis of the results and comparison of the reactivity pa-
University (sample Nos. 6 and 7). The specific sur-
rameters of powders.
face area (Ssp) of the samples studied was determined
by the BET (Brunauer Emmett Teller) method (low-
temperature adsorbtion of nitrogen). The metal content
was determined by the volumetric technique, i.e., from
EXPERIMENTAL STUDY OF POWDERS
the volume of hydrogen evolved during interaction of
The comparability of the results of derivatography powders with a 5% solution of NaOH. The temperature
of the powders is provided for by identical experimental of the onset of oxidation was determined by the Piloyan
conditions. The standard weight of ultrafine aluminum method [9] from the curve of mass loss of differential
samples was H"5 · 10-5 kg, the heating rate was equal to thermal analysis (DTA).
H"10ć%C/min, and the remaining parameters were found The data given in Table 1 suggest that with rise
during numerous experiments [12]. in Ssp (decrease in surface-average diameter of par-
Table 1 lists characteristics of the powders studied: ticles as), the mass concentration of the unoxidized
commercial aluminum powders (sample Nos. 1 and 2), metal (CAl0) in the powders increases, and, simultane-
PY87 dust ( Pechiney ) (sample No. 3), STPA-IK (ul- ously, the bulk density Á0 decreases. For sample Nos. 1
trafine powder produced by vaporization condensation and 2, the temperature of the onset of intense oxida-
in argon) (sample No. 4), ultrafine aluminum pow- tion far exceeded the melting temperature of aluminum
ders produced by electrical explosion of conductors, (660ć%C). The values of Ton for the remaining samples
the  Alex powder ( Argonide Corp. ) [13] (sample lie below the melting point of aluminum (especially for
No. 5), and STPA-1 and STPA-4 ultrafine aluminum sample No. 7, whose temperature Ton is 120ć%C below
powders produced at the semicommercial division of the melting point of aluminum). For sample Nos. 1
the High-Voltage Institute of the Tomsk Polytechnical and 2, the degree of oxidation of the metal before melt-
420 Il in, Gromov, and Yablunovskii
TABLE 1
Ä…, %
ć%
Sample No. Ssp (BET), m2/g as, µm CAl , % Á0, g/cm3 Ton, C
0
up to 660ć%C up to 1000ć%C
1 0.15 80.0 99.5 1.60 920 0.65 52.2
2 0.38 9.0 98.5 0.87 820 2.5 41.8
3 5.91 Powder 96.0 0.315 580 8.0 40.5
4 11.00 0.20 86.0 0.21 555 39.9 69.3
5 12.10 0.18 94.8  548 39.4 45.0
6 7.80 0.28 91.0 0.13 560 23.9 74.3
7 16.00 0.13 89.0 0.11 540 50.1 78.6
vox, mg/sec
Sample No. S/"m, rel. units Notes
ć%
(in the temperature range, C)
1 0.04 (920 950) 2.1 Sample weight 86.2 mg
2 0.05 (970 980)  
Scale diameter from 2 3 to 5 6 µm;
3 0.025 (580 650) 
width 0.15 µm
4 0.125 (560 570) 7.7 Sample weight 26.8 mg
5 0.05 (541 554)  [13]
6 0.04 (565 590) 7.0 
7 0.05 (550 605) 8.7 
ing did not exceed 3% and for sample No. 3 it did not Fig. 1. The oxidation of sample No. 7 proceeds in two
exceed 10%. For ultrafine aluminum powders (sample macroscopic stages: the first stage begins at a temper-
Nos. 4 7), the degree of conversion of aluminum before ature of 550ć%C and the second, less intense stage be-
the melting point was more than 20% (the maximum gins at 750ć%C and continues up to complete oxidation
value of 50.1% was obtained for sample No. 7). The of aluminum (more than 1000ć%C). For sample No. 2,
zone of the most intense oxidation was determined from it is possible to distinguish four stages of oxidation:
the TG curve (segments AB and A B in Fig. 1). The 1) 560 640ć%C; 2) 810 970ć%C; 3) 970 980ć%C; 4) 980ć%C
highest oxidation rate was observed for sample No. 4. and then until complete oxidation. The degrees of ox-
Sample Nos. 1, 2, 6, and 7 had comparable oxidation idation of sample Nos. 7 and 2 before melting were
rates, whereas intense oxidation of sample Nos. 1 and 2 50.1 and 2.5%, respectively. The first macrostage of
began at 920 and 820ć%C, respectively, and oxidation oxidation of sample No. 7 also includes several stages:
of sample Nos. 6 and 7 began at about 400ć%C below one can see four segments of increase and decrease in
than that for samples 1 and 2. The specific heat re- temperature on the DTA and DTG (differential ther-
lease S/"m was determined by dividing the area of the mogravimetry) curves. Several powder fractions of close
peak of heat release (DTA curve) by the corresponding sizes are unlikely to burn out separately at T > 2000ć%C:
increase in sample weight (mg) (TG curve). The pa- the particle distribution of ultrafine aluminum powders
rameter S/"m is maximal for sample No. 7 and more produced by electrical explosion is not tetramodal but
than four times larger than that for sample No. 1. bimodal (maxima are in regions 1 3 and 0.1 µm). We
can explain this unique phenomenon for ultrafine alu-
minum powders if we assume that combustion proceeds
under quasiadiabatic conditions. Apparently, an abrupt
DISCUSSION OF RESULTS
increase in temperature leads to the  actuation of en-
dothermic processes, primarily, aluminum boiling, ni-
An analysis of the data given in Table 1 shows that
tration with further formation of AlN or AlON, and va-
in accordance with the testing parameters proposed, the
porization and dissociation of aluminum oxide. Thus,
most active powder is sample No. 7 (STPA-4 powder).
the heat expended in vaporization of 1 mole of Al2O3
Derivatograms of sample Nos. 2 and 7 are shown in
Reactivity of Aluminum Powders 421
a CONCLUSIONS
To test ultrafine aluminum powders, one should use
several characteristics that are standard for ordinary
powders: particle shape and particle size distribution,
specific surface area, etc. [1]. At the same time, the
reactivity of ultrafine powders is characterized by the
following parameters:
" temperature of the onset of oxidation;
" maximum oxidation rate;
" degree of conversion (degree of oxidation) of alu-
x 5500
minum;
b
" ratio of the thermal effect to the weight increase
measured under standard conditions (see Fig. 1).
These parameters can be obtained under conditions of
nonisothermal oxidation in air under linear heating.
This set of parameters reflects not only the reactivity
of the powders but also their special features, i.e., can
be used as a test for a particular powder (see Table 1).
If other oxidizers are used, the activity of such pow-
ders can also be determined from the above parameters
taking into account the special features of the system
x 5500  aluminum powder oxidizer.
This work was supported by Ministry of Education
of the Russian Federation (Grant No. 98-8-5.2-74).
Fig. 2. Electron photomicrographs of products of ox-
idation in air for sample Nos. 2 (a) and 7 (b) (sample
numbers correspond to those of Table 1).
REFERENCES
at the boiling point is equivalent to the amount of heat
1. P. F. Pokhil, A. F. Belyaev, Yu. V. Frolov, V. S. Lo-
released in the combustion of 1.2 moles of the metal in
gachev, and A. I. Korotkov, Combustion of Powdered
oxygen under adiabatic conditions [1]. Consuming the
Metals in Reactive Media [in Russian], Nauka, Moscow
heat of the reactive system, endothermic processes de-
(1972).
crease the temperature, and this can occur several times
(for example, four times for sample No. 7). The self- 2. S. Sarner, Propellant Chemistry, New York (1966).
organization processes involved in the high-rate com- 3. V. G. Shevchenko, V. I. Kononenko, I. N. Latosh, et al.,
 Effect of the size factor and alloying on oxidation of
bustion of the powders occur via the feedback described.
aluminum powders, Fiz. Goreniya Vzryva, 30, No. 5,
The possibility of endothermic reactions is confirmed by
68 71 (1994).
chemical analysis and electron microscopy. In the pho-
tographs shown in Fig. 2, the combustion products of 4. A. L. Breiter, V. M. Mal tsev, and E. I. Popov  Means
sample No. 7 are submicron needles, and the combus- of modifying metallic fuel in condensed systems, Fiz.
tion products of sample No. 2 are spheres. Thus, the Goreniya Vzryva, 26, No. 1, 97 104 (1990).
microstructure of the combustion products of STPA-4 5. V. G. Shevchenko, V. L. Volkov, V. I. Kononenko, et
ultrafine powder changed radically in comparison with al.,  Influence of sodium and potassium polyvanadates
the initial powder, while the microstructure of the com- on aluminum-powder oxidation, Fiz. Goreniya Vzryva,
bustion products of the commercial powder remained 32, No. 4, 91 94 (1996).
unchanged. Obviously, the appearance of combustion 6. F. Tepper, G. Ivanov, M. Lerner, and V. Davidovich,
products in the form of needles of submicron diameters  Energetic formulations from nanosize metal powders,
results from participation of the gaseous phase in the in: Proc. Int. Pyrotechn. Seminars, No. 24, Chicago
formation of the final products [15]. (1998), pp. 519 530.
422 Il in, Gromov, and Yablunovskii
7. V. F. Petrunin and L. D. Ryabev,  State and prospects 12. A. P. Il in and L. T. Proskurovskaya,  Oxidation of met-
of the problem of ultradisperse (nano-) systems, in: als in the ultradisperse state. Part II: High-temperature
Physicochemistry of Ultradisperse Systems: Proc. IV oxidation of aluminum: Size and structural factors,
All-Union Conf., Moscow Eng.-Phys. Inst., Moscow Cherkassy (1988). Deposited at ONII TÉKhIM XII,
(1998), pp. 15 20. No. 905.
8. N. A. Yavorovskii,  Production of ultradisperse powders 13. M. M. Mench, K. K. Kuo, C. L. Yeh, and Y. C. Lu,
by electric explosion of conductors, Izv. Vyssh. Uchebn.  Comparison of thermal behaviour of regular and ultra-
Zaved., Fiz., No. 4, 114 135 (1996). fine aluminum powders (Alex) made from plasma ex-
9. W. Wendtland, Thermal Methods of Analysis, Wiley,
plosion process, Combust. Sci. Technol., 135, 269 292
New York (1974).
(1998).
10. A. P. Il in,  On the excess energy of ultradisperse pow- 14. M. I. Lerner,  Control of the formation of ultrafine
ders produced by explosion of wires, Fiz. Khim. Obrab.
particles under conditions of electric explosions of con-
Mater., No. 3, 94 97 (1994)
ductors, Candidate s Dissertation in Tech. Sci., Tomsk
11. V. G. Ivanov and O. V. Gavrilyuk,  Regularities of ox-
(1988).
idation and self-ignition of electric explosive ultrafine
15. A. P. Il in, G. V. Yablunovskii, A. A. Gromov, et al.,
metal powders in air, Fiz. Goreniya Vzryva, 35, No. 6,
 Combustion of mixtures of ultrafine powders of alu-
53 60 (1999).
minum and boron in air, Fiz. Goreniya Vzryva, 35,
No. 6, 61 64 (1999).