process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20

Contents lists available at ScienceDirect

Process Safety and Environment Protection

journal homepage: www.elsevier.com/locate/psep

Dust explosions: How should the influence of humidity

be taken into account?

M. Traoré, O. Dufaud ∗ , L. Perrin, S. Chazelet, D. Thomas

CNRS - Laboratoire des Sciences du Génie Chimique, UHP-INPL – ENSIC, 1 rue Grandville, BP 20451, 54001 Nancy, France

a b s t r a c t

In this work, the influence of humidity on dust explosions of metallic (aluminium) and organic materials (icing sugar,
polyethylene and magnesium stearate) has been studied. The impact of pre-humidification of powders on their
ignition sensitivity, their volume resistivity and charge decay time has been assessed. The influence of humidity on
explosion severity has also been studied by two methods: on the one hand, the dust sample was stored in a controlled
workstation at constant relative humidity; on the other hand, the dry dust was dispersed in a humidity controlled
atmosphere in the vessel.
As expected, the effect of humidity strongly depends on the chemical nature of the particles. Experiments on
powders volume resistivity and charge decay time have shown typical trends but have especially pointed out the
inadequacy of some standards. Inhibition phenomena have been verified for polyethylene and magnesium stearate,
whereas both inhibition and promotion have been observed for icing sugar and could be explained by an evolution of
sucrose structure. Dry aluminium dust explosions in humid atmosphere show that water vapour inerts the explosion.
However, when aluminium is stored at controlled humidity, the maximum rise of pressure rate increases with the
water content, which is probably due to hydrogen generation.
© 2008 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.

Keywords: Dust explosions; Ignition sensitivity; Equilibrium relative humidity; Static electricity; Standards

1. Introduction oxygen transfer to particle surface, it could also have a pro-

Like fire and ice, humidity and dusts explosions have often
been considered as antagonistic forces. A glance at case his-
tories and statistics tends to corroborate this assertion. For
instance, Count Morozzo’s account of the first quoted dust
explosion in a flour warehouse in Turin on the 14 December
1785, mentioned that this accident could be attributed “to the
extraordinary dryness of the corn” due to a “remarkably dry”
weather (Eckhoff, 2003). Li et al. (1995) also underlined that
statistics on dust explosions in the United States from 1979 to
1986 clearly show that most of the accidents occurred during
winter months when the atmosphere has the lowest humidity
content.
However, the relationship between dust explosivity and
powders moisture content (and thus equilibrium relative
humidity) is not as univocal as it superficially seems. If humid-
ity could affect the ignition and explosion processes, notably
by dust agglomeration, thermal sink, inerting or hindering
moting effect (Laurent, 2003). In particular, fermentation for
grain dusts and oxidation–reduction reactions of metal pow-
ders with water are phenomena generating combustible gas,
which should not be neglected in risk analyses (Otsuka and
Itagaki, 2004).
Despite these facts, the overwhelming majority of stan-
dards and procedures related to dust explosion characteri-
sation neglects the impact of humidity and only specifies
that the equilibrium relative humidity should be checked and
noted down, as for instance IEC 1241-2-1 (MIT) or IEC 1241-2-3
(MIE) (IEC standard, 1994a, 1994b).
In this work, the effects of both water activity (linked
to equilibrium relative humidity) and ambient humidity on
dust explosions characteristics have been checked. Natural
and synthetic organic compounds but also metal powders
have been tested. The proportion of electrostatic discharges
as ignition sources in dust explosion accidents being signifi-
cant (Planas-Cuchi et al., 1997), notably for plastic compounds,

∗
Corresponding author. Tel.: +33 3 83 17 53 33; fax: +33 3 83 32 29 75.
E-mail address: dufaud@ensic.inpl-nancy.fr (O. Dufaud).
Received 12 February 2008; Received in revised form 17 July 2008; Accepted 14 August 2008
0957-5820/$ – see front matter © 2008 The Institution of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.psep.2008.08.001
 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20 15

Table 2 – Explosion sensitivity of the dried powders

Nomenclature
Samples d50 (␮m) MIT (◦ C) MIE (mJ) MEC
aw water activity (g m−3 )
dx particle diameter at which x% of the particles
Aluminium 7 900 13 30
are smaller (␮m) Icing sugar 22 380 18 50
d3,2 Sauter’s mean diameter—diameter of a sphere Magnesium stearate 6 420 4 30
having the same volume to surface area ratio Polyethylene 136 480 500 60
as the particle (␮m)
ERH equilibrium relative humidity
sphere having the same volume to surface area ratio as the
Pm maximum overpressure (bar)
particle, is also indicated in Table 1.
Pmax maximum of the maximum overpressures over
Scanning electron microscope (SEM) was also performed
a wide range of concentrations (bar)
on powders samples to characterise the particles shapes and
(dP/dt)m maximum rate of pressure rise (bar s−1 )
the surface roughness (Fig. 1).
(dP/dt)max maximum of the maximum rates of pres-
Aluminium particles are rather ellipsoidal than spherical
sure rise over a wide range of concentrations
and their surface is quite smooth. Icing sugar is composed of
(bar s−1 )
crystal-shape particles, whereas magnesium stearate powder
 charge decay time (s)
looks like flakes. Polyethylene pellets are rather spherical with
a porous and highly irregular surface.
The samples were systematically dried at 50 ◦ C under vac-
Table 1 – Particles sizes of the powders uum during 2 h before handling. Two methods were used to
Samples d10 (␮m) d50 (␮m) d90 (␮m) d3,2 (␮m) demonstrate the influence of the equilibrium relative humid-
Aluminium 3 7 13 3
ity (ERH) on aluminium dust explosions: (i) the dry dust was
Icing sugar A 7 22 41 7 dispersed in a humidity controlled atmosphere in the explo-
Icing sugar B 17 45 76 12 sion vessel, or (ii) the dust sample was stored in a controlled
Magnesium stearate 3 6 15 3 workstation (glove box) at constant relative humidity dur-
Polyethylene 34 136 249 58 ing a sufficient period of time allowing the equilibrium to be
reached.
the influence of humidity on electrostatic properties has also The relation between the water activity – ratio between the
been investigated. vapour pressure of water in the compound to the vapour pres-
sure of distilled water under the same conditions; similar to
2. Experimental equipments and tests the equilibrium relative humidity in percent divided by 100
procedures – and the water content in the powder is called adsorption
isotherm. It was characterised by using an electromagnetic
2.1. Samples characteristics and preparation suspension microbalance (Fig. 2). The time to equilibrium was
also determined.
Experiments were carried out with four materials chosen for
their various natures: icing sugar—a natural organic powder, 2.2. Explosion sensitivity
polyethylene dust and magnesium stearate: synthetic organic
compounds and aluminium powder: a metal. At first, the minimum ignition energy (MIE), minimum ignition
Aluminium powders were provided by Goodfellow with temperature (MIT) and minimum explosible concentration
99% purity and a specific maximum particle size of 15 ␮m. (MEC) have been determined on dried samples. The measure-
Icing sugar was provided by Erstein and polyethylene (PE) ment of minimum ignition temperature has been carried out
powders were provided by Innovene; various particle-size dis- thanks to a Godbert-Greenwald furnace (Chilworth Technol-
tributions were obtained by sieving. Magnesium stearate was ogy), whereas the minimum ignition energy and minimum
supplied by Sigma-Aldrich. The particle-size distribution of explosible concentration have been determined by using a
the dusts was determined by using a laser diffraction ana- modified Hartmann tube (Kühner AG) (Siwek and Cesana,
lyzer (Mastersizer, Malvern Instrument). The samples were 1995). Tests were carried out in accordance with the princi-
characterised by the d10 , d50 and d90 quantiles of the volumet- ples of IEC standards 1241-2-1 and 1241-2 3. The 20 L sphere
ric distribution as indicated in Table 1; the dx diameter, being apparatus has also been used to confirm MEC data. The results
defined as the size at which x% of the particles are smaller. of the various tests performed on dried powders are listed in
The Sauter mean diameter d3,2 , defined as the diameter of a Table 2.

Fig. 1 – Images by SEM of aluminium, icing sugar, magnesium stearate and polyethylene particles.
16 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20
Fig. 2 – Adsorption isotherms of (a) icing sugar and PE; (b)
aluminium and magnesium stearate at 25 ◦ C.
Table 2 shows that the aluminium, icing sugar and mainly
magnesium stearate could be considered as highly ignitable
dusts, whereas polyethylene has a far lower ignition sensitiv-
ity (especially its MIE).
Fig. 3 shows the impact of pre-humidification of magne-
sium stearate on its ignition sensitivity. Both MIE and MIT
are strongly affected by the variation of the equilibrium rel-
ative humidity of the storage atmosphere; they respectively
increase from 4 mJ and 420 ◦ C for the dried powder to 12 mJ and
Fig. 3 – Evolution of the minimum ignition temperature and
minimum ignition energy of magnesium stearate as a
function of the equilibrium relative humidity.
570 ◦ C for particles stored at 90% ERH. It should be noted that
the sigmoid form of the sorption isotherm (Fig. 2) is especially
perceptible for MIE evolution, which confirm the importance of
water uptake on energy transfer and ignition processes. Such
trends could be observed for polyethylene or polymers.
More generally, it has been demonstrated that humidity
tends to lower the ignition sensitivity of organic materials,
depending on the hydrophobic or hydrophilic nature. MIT and
MEC especially increase when the water content of the dust
rises (Nagy and Verakis, 1983; Nifuku et al., 2006).
Concerning the influence of humidity on aluminium igni-
tion and in particular on the MIE, no significant evolution has
been observed: the MIE ranges from 13 mJ for the dried powder
to 12 mJ at 60% and 75% ERH.
2.3. Explosion severity
The explosion severity parameters Pmax , maximum pressure,
and (dP/dt)max , maximum rate of pressure rise, have been
determined in a 20 L apparatus (Kühner AG) (Callé et al., 2005).
The test chamber is a hollow sphere made of stainless steel.
A cooling water jacket dissipates the heat of the explosion.
During a test, the dust is dispersed into the sphere from a pres-
surised storage chamber via the outlet valve and a disperser.
Tests have been performed with two pyrotechnic ignitors of
5 kJ each as ignition source for the determination of Pmax and
(dP/dt)max . In order to take potential overdriving effects into
account, whereas only one 5 kJ has been used for MEC char-
acterisation. The explosion pressure is recorded as a function
of time using piezoelectric pressure sensors. The dust con-
centration is then varied over a 60–3000 g m−3 range. This
apparatus also allows the determination of minimum explo-
sive concentration, below which explosion is not possible.
Tests were carried out according to ISO 6184-1 (ISO standard,
1985).
Some modifications were made on the sphere: (i) a thermo-
couple was introduced in the centre of the sphere to record the
temperature during an explosion; (ii) the temperature of the
water outlet was also recorded, (iii) a system of gaseous efflu-
ents treatment was added, (iv) the sphere has been grounded
to avoid uncontrolled electrostatic discharges. In order to con-
trol the equilibrium relative humidity, the 20 L sphere has
also been modified with an injection system which allows the
vapourisation of a precise amount of water in the vessel before
ignition. For operating temperatures lower than water boiling
point, the injected water content ranged from 0 to the vapour
saturated pressure. Attention should be paid to the increase of
temperature (up to 360 ◦ C) due to the 10 kJ ignition source. Even
if water is introduced over the saturation volume (0.28 mL),
the remaining liquid could be vapourised during ignition and
before flame propagation. The pressure rise induced by the
ignition source is systematically subtracted from the overall
pressure.
2.4. Electrostatic hazards quantification
The knowledge of the volume resistivity (˝) and the charge
decay time () of a powder could bring valuable information
on its electrostatic hazards (Laurent, 2003). The volume resis-
tivity (in  m) of a material is the electrical resistance across a
cube of unit cross sectional area and unit length. The volume
resistivity is determined by placing the powder sample in a
resistivity cell comprising two electrodes between which an
electric current may be passed through the sample (Chilworth
 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20 17

Technology Ltd). Measurements were carried out according to

IEC standard 1241-2-2 (IEC standard, 1993).
Charge decay time is measured by recording the residual
charge on the powder as a function of time as it is conducted
to earth. The charge decay time is then defined as the char-
acteristic time taken for the charge to decrease to 1/e (37%)
of its initial value. It was measured by a static monitor “field
mill” instrument JCI 140 (John Chubb Instrumentation). How-
ever, powders seldom behave in this way and if the decay is
not a true exponential, such as where the material is non-
ohmic, thus there will not be a simple relationship between
the relaxation time and the decay time. Tests were carried out
according to BS7506 standard (BS standard, 1996).
Temperature and humidity greatly affect electrical flow
and adhesion properties of powders. For this reason it is
essential to perform many electrostatic measurements under
controlled temperature and humidity conditions. The temper-
ature inside our glove box chamber can be controlled between
−10 and 80 ◦ C, and the equilibrium relative humidity between
0.1 and 99.9%.

3. Results and discussion

3.1. Humidity and static electricity

It is common knowledge that the presence of static electricity

could be significantly reduced by increasing the equilibrium
relative humidity up to 60%. However true this assertion
is, this solution can be contestable because it also mod-
ifies the powders properties, causes equipment problems
and introduces biological contaminants into the powder and
equipment.
Fig. 4 shows the equilibrium relative humidity dependency
on volume resistivity at 25 ◦ C. Behaviours of icing and magne-
sium stearate are clearly different, owing to the hydrophilic
nature of sugar and hydrophobic one of magnesium stearate.
Thus, the volume resistivity of icing sugar is 10 times lower
at 30% ERH than at 15% ERH; whereas the resistivity of
magnesium stearate is not humidity dependent. The sorp-
tion isotherms show no significant differences between water
sorption on icing sugar and on magnesium stearate (up to 75%
ERH—Fig. 2). However, a chemical modification of the particles
surface is confirmed by TDA for icing sugar.
Fig. 5 presents the equilibrium relative humidity depen-
dency on the charge decay time at 25 ◦ C. The icing sugar decay
time is about 10 s for 40% humidity while it is equal to 1 h for
15% humidity. For the magnesium stearate the charge decay
time falls from 135,000 s (more than 37 h) at 17% ERH to 60,000 s
(16.6 h) at 75% ERH. Even if an influence is henceforth notewor-
thy on magnesium stearate, the sensitivity to water activity is
once again more pronounced on sugar.
The previously quoted standards, especially the one con-
cerning volume resistivity, only specify that a powder sample
can “be tested directly” (without conditioning). The previous
results show that it is indispensable to work at the process
most representative humidity and that experiments have to
be carried out more accurately and carefully than specify by
the existing standards (Perrin et al., 2007).
3.2. Effect of humidity on explosion severity
The explosion severities of four dusts have been studied. As a
function of their nature, distinct behaviours have been iden-
tified.
Fig. 4 – Influence of the equilibrium relative humidity on
the volume resistivity (25 ◦ C) on icing sugar—22 ␮m (a) and
on magnesium stearate—6 ␮m (b).
Fig. 6 shows that humidity tends to drastically lower the
explosivity of magnesium stearate: for instance at 250 g m−3 ,
(dP/dt)max drops from 1035 bar s−1 for a dried powder to
640 bar s−1 at 90% ERH. This trend is also perceptible on the
maximum explosion pressure, which decrease from 8.1 to
6.8 bars on the same range. Thermal sink effect and particle
agglomeration, observed by microscopy, could notably be put
forward.
Both moisture content of the particles and humidity of
the explosion atmosphere have an influence on polyethylene
dusts explosion. Similarly to the behaviour of magnesium
stearate, the water activity inhibits the explosion and oxida-
tion reaction of PE.
Fig. 7 shows that both maximum explosion pressure and
rate of pressure rise also tends to lower when the water con-
tent of the explosion atmosphere increases. The calculation
of adiabatic flame temperature in presence and absence of
water demonstrates that the injection of water up to satu-
ration implies a drop of maximum pressure of approximately
0.3 bar, which corresponds to the observed decrease. The effect
of humidity on polyethylene explosion is thus essentially due
to heat consumption by water vapourisation; phenomenon
which is more important since the water content in polyethy-
lene dust is significant (Fig. 2).
It should be pointed out that, a slight pressure increase
could be expected for high water volumes due to water
vapourisation. This effect would induce a 6% rise of the max-
imum pressure for a 1 mL water injection and could explain
the pressure increase noticed for water volumes greater than
saturation.
18 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20

Fig. 7 – Effect of atmosphere humidity on dry polyethylene

explosions severity.

behaviours have been observed for various concentrations

and particles-size distribution. As regard to thermogravimet-
ric and thermo-differential analyses, a slight change in icing
sugar thermal response – and thus structure – as a function of
the storage humidity has been demonstrated. The hypothe-
sis of a sugar maturation and sucrose inversion (from sucrose
to glucose and fructose), leading to greater particles ignitabil-
ity, could be put forward (Traoré et al., 2004). Humidity, which

Fig. 5 – Influence of the equilibrium relative humidity on

the charge decay time (25 ◦ C) on icing sugar—22 ␮m (a) and
on magnesium stearate—6 ␮m (b).

No considerable variations of icing sugar explosivity have

been observed as a function of the particles moisture; how-
ever, significant and reproducible trends have been noticed.
From 0 to 30% ERH (from 0 to 0.34% moisture content), both
maximum pressure and rate of pressure rise slightly increase,
whereas when the equilibrium relative humidity raises to 75%
ERH (1.7% moisture content, before irreversible agglomeration
at 80%) the dust explosivity significantly drops (Fig. 8). Those

Fig. 8 – Influences of the equilibrium relative humidity on

Fig. 6 – Influence of the equilibrium relative humidity on (a) the maximum explosion pressure and on (b) the
the maximum rate of pressure rise of magnesium stearate maximum rate of pressure rise of icing sugar at various
at various dust concentrations. dust concentrations.
 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20
Fig. 9 – (a) Evolution of the maximum explosion pressure
as a function of the dust concentration and the equilibrium
relative humidity of storage. (b) Influence of water injection
on the maximum explosion pressure Pm —for 7 ␮m
diameter aluminium dusts.
influences interfacial gas adsorption and desorption generates
an oxygen saturated “sirup” layer on particle surfaces, increas-
ing the liability to spontaneous combustion. This surface
modification promotes the combustion kinetics of sugar par-
ticles. This impact will only be perceptible for low humidities
and gives way to water inhibition phenomena when the mois-
ture content increases. The previous trends are not observed
when the explosivity of dry icing sugar is tested in a humid
atmosphere.
Experiments were then performed on aluminium particles
with a volume median diameter of 7 ␮m. Fig. 9a represents
the maximum pressure as a function of the equilibrium rel-
ative humidity of the dust storage. It could be noticed that
storage conditions have a significant effect on the maximum
explosion pressure, especially at 60 and 76% of equilibrium
relative humidity. This trend is even more noticeable on the
maximum rate of pressure rise, which raises from 1530 bar s−1
for dry powders to 2100 and 2150 bar s−1 at respectively 60 and
76% ERH for 250 g m−3 dusts concentration.
As no sensible modification of the ignition sensitivity has
been noticed, the previous evolutions of aluminium explo-
sion severity could not be explained only by the alteration
of thermal transfers or by heat sink effect due to the pres-
ence of water. It should be stressed that molten aluminium
is known to react violently with water. If the protective layer
19
of alumina which is formed on the metallic surface is broken,
a rapid chemical reaction occurs between the metal and the
water generating a large heat release and hydrogen production
(Cho et al., 1995). Thus, the promotion of explosion violence
is probably due to the following reaction between water and
aluminium (Kwok et al., 2002):
2Al (s) + 3H2 O (g) → Al2 O3 (s) + 3H2 (g)
Nevertheless, does the same evolution occur if the water
is initially present in the oxidant atmosphere as if it is
adsorbed directly onto the particle surface? Explosions of dry
aluminium were then carried out in humid atmosphere by
injecting a specific volume of water. Here, water molecules
have not enough time to adsorb onto aluminium particles
surface. In fact the contact between water molecules and
aluminium particles lasts only the duration of particles dis-
persion and of combustion (from 20 to 70 ms); whereas the
time of sorption equilibrium is about 2 h.
Water amounts are very low (the saturation volume is
about 0.3 mL) but when injected in the sphere, they reduce the
explosion pressures from at least 1 bar at saturation (Fig. 9b).
Nevertheless, considering the drop of pressure due to the
water enthalpy of vapourisation does not seem to be suffi-
cient to explain the observed reduction of explosion pressure.
The maximum rate of pressure rise also decreases for large
water volumes from 900 (dried powder) to 400 bar s−1 (at sat-
uration) for 250 g m−3 dusts concentration. Here, the presence
of water also seems to hinder the oxidation kinetics (Traoré,
2007).
The differences of behaviours observed between the two
kinds of experiments show that water reacts differently
with aluminium when dispersed in the atmosphere or when
adsorbed onto the particles. At first, it should be under-
lined that the water quantities are approximately 10 times
lower (depending on the dust concentration) when adsorbed
than when dispersed in the sphere (Fig. 2). Then, by con-
sidering the processes of aluminium oxidation described
by Trunov et al. (2005), some assumptions could be made:
when the water is adsorbed onto the particles surface, the
phase transformations of alumina layer or its growth could
be strongly modified by the presence of water; moreover it
could also have an impact on the diffusion flux of oxygen
from the atmosphere or the diffusion flux of the metal from
the particle core. Besides, some authors, who have studied
the combustion of aluminium with water, stressed that the
combustion regime could change from a vapour phase reac-
tion for pure aluminium in air (vapourisation followed by a
gas-phase reaction) to a heterogeneous process with water
(Risha et al., 2007). Experiments will then be carried out
to show a possible alumina phase modification during the
oxidation due to water adsorption. On the contrary, when
the humidity of the explosion atmosphere is increased, the
physico-chemical structure of the dust could not be modified
and water does not react with aluminium to form hydro-
gen; the processes of thermal transfer inhibition and inerting
prevail.
It emerges from this study that the influence of water
on aluminium dusts explosivity could be decoupled: when
water adsorption occurs, hydrogen generation leads to an aug-
mentation of the explosion severity, whereas an inhibiting
effect could be put forward when the effective contact time
between water and aluminium is reduced to the explosion
duration.
20 process safety and environment protection 8 7 ( 2 0 0 9 ) 14–20
4. Conclusions
Depending on their chemical nature, the equilibrium relative
humidity and thus the moisture content of powders could
inhibit or promote the dusts ignitability and explosion sever-
ity. On the one hand, the inhibition could notably be caused
by dust agglomeration, heat sink effect due to the water heat
capacity and heat of vapourisation, water vapour inerting in
gas phase or inhibition of mass transfer (especially of oxygen)
at the particle surface. On the other hand, an increase of the
ignition and explosion characteristics could be observed when
the water presence leads to violent chemical reactions with
the powder (combustible gas generation) or when its adsorp-
tion chemically modified the particle surface and improves the
species diffusion. These last-mentioned influences are some-
times very strong and could lead to explosion risks higher than
for the dry powder.
However, it should be underlined that standards often
neglect the impact of this essential parameter. As an exam-
ple, ISO 6184-1 standard specifies that the determination of
the dust explosion indices should be carried out at humidity
lower than 10%; which is not the worst case scenario especially
for some metallic powders. As a conclusion, a peculiar atten-
tion should be paid to this parameter and standards should
sometimes been bypassed or modified to take the influence of
humidity into account.
References
BS standard 7506, 1996, Methods for measurement in
electrostatics: part 2. Guide to Basic Electrostatics (British
Standards Institution).
Callé, S., Klaba, L., Thomas, D., Perrin, L. and Dufaud, O., 2005,
Influence of size distribution and concentration on wood dust
explosion. Experiments and reaction modelling. Powder
Technology, 157: 144–148.
Cho, D.H., Armstrong, D.R. and Anderson, R.P., 1995, Combined
vapor and chemical explosions of metals and water. Nuclear
Engineering and Design, 155: 405–412.
Eckhoff, R.K., (2003). Dust explosions in the process industries (3rd
edition). (Gulf Professional Publishing, Amsterdam,
Netherlands).
IEC standard 1241-2-1, 1994a, Electrical apparatus for use in the
presence of combustible dust. Part 2: Test method section 1:
Methods for determining the minimum ignition temperature
of dust (International Electrotechnical Commission).
IEC standard 1241-2-2, 1993, Method for determining the
electrical resistivity of dust in layers (International
Electrotechnical Commission).
IEC standard 1241-2-3, 1994b, Method for determining the
minimum ignition energy of dust/air mixtures (International
Electrotechnical Commission).
ISO standard 6184-1, 1985, Explosion protection systems. Method
for determination of explosion indices of combustibles dust in
air, part 1 (International Standard Organization).
Kwok, Q., Fouchard, R., Turcotte, A., Lightfoot, P., Bowes, R. and
Jones, D., 2002, Characterization of aluminium nanopowder
compositions. Propellants, Explosives, Pyrotechnics, 27:
229–240.
Laurent, A., (2003). Sécurité des Procédés Chimiques. (Tec&Doc, Paris,
France).
Li, Y.C., Kauffman, C.W. and Sichel, M., 1995, An experimental
study of deflagration to detonation transition supported by
dust layers. Combustion and Flame, 100:
505–515.
Nagy, J. and Verakis, H.C., (1983). Development and Control of Dust
Explosions. (Marcel Dekker, New York, USA).
Nifuku, M., Tsujita, H., Fujino, K., Takaichi, K., Barre, C., Paya, E.,
Hatori, M., Fujiwara, S., Horiguchi, S. and Sochet, I., 2006,
Ignitability assessment of shredder dusts of refrigerator and
the prevention of the dust explosion. Journal of Loss
Prevention in the Process Industries, 19(2–3):
181–186.
Otsuka, T. and Itagaki, H., 2004, The dust explosion of aluminium
dust in water vapour, Special research report no. 29-03,
National Institute of Industrial Safety, Japan.
Perrin, L., Laurent, A., Falk, V., Dufaud, O. and Traore, M., 2007,
Dust and electrostatic hazards: could we improve the current
standards? Journal of Loss Prevention in the Process
Industries, 20: 207–217.
Planas-Cuchi, E., Montiel, H. and Casal, J., 1997, A survey of the
origin, type and consequences of fire accidents in process
plants and in the transportation of hazardous materials.
Transactions of Institution of Chemical Engineers, 75:
3–8.
Risha, G.A., Son, S.F., Yetter, R.A., Yang, V. and Tappan, B.C., 2007,
Combustion of nano-aluminum and liquid water. Proceedings
of the Combustion Institute, 31(2): 2029–2036.
Siwek, R. and Cesana, C., 1995, Ignition behavior of dusts:
meaning and interpretation. Process Safety Progress, 14:
2.
Traoré, M., Dupont, J., Dufaud, O., Perrin, L., Callé, S., Vallières, C.
and Thomas, D., 2004, Dust explosibility characteristics of
icing sugar. Chemical Engineering Transactions, 5:
119–124.
Traoré, M., 2007, Explosions de poussières et de mélanges
hybrides. Etude paramétrique et relation entre la cinétique de
combustion et la violence de l’explosion, Ph.D. Thesis, (INPL,
Nancy, France).
Trunov, M.A., Schoenitz, M., Zhu, X. and Dreizin, E.L., 2005, Effect
of polymorphic phase transformations in Al203 film on
oxidation kinetics of aluminum powders. Combustion and
Flame, 140(4): 310–318.